PDA

View Full Version : Dork Poll tracking (Pomeroy-Sagarin)



Pages : [1] 2

JasonEvans
11-15-2010, 11:09 AM
Each week, I will put a post in this thread with the KenPom and Sagarin rankings of each ACC team. Food for thought...



Team Sag Pom Dif
Duke 1 1 0
UNC 9 16 -7
Clemson 20 29 -9
Maryland 29 23 +6
Fla St 32 27 +5
Miami 46 46 0
Va Tech 51 25 +26
Ga Tech 53 78 -25
Virginia 58 57 +1
BC 63 38 +25
NC State 70 41 +29
Wake 90 132 -42

Additionally, We can look at how each school was ranked the week before and where they are now in the two most widely regarded computer rankings.


Sagarin Today Last wk Change
Duke 1 1 --
UNC 9 5 -4
Clemson 20 25 +5
Maryland 29 29 --
Fla St 32 32 --
Miami 46 47 +1
Va Tech 51 46 -5
Ga Tech 53 43 -10
Virginia 58 77 +19
BC 63 64 +1
NC State 70 65 -5
Wake 90 37 -53

Virginia is the big mover in the Saragin rankings. Wake's strangely high preseason ranking has been overtaken by their horrid first week results.


Pomeroy Today Last wk Change
Duke 1 1 --
UNC 16 12 -4
Maryland 23 24 +1
Va Tech 25 17 -8
Fla St 27 20 -7
Clemson 29 28 -1
BC 38 41 +3
NC State 41 33 -8
Miami 46 39 -7
Virginia 57 62 +5
Ga Tech 78 83 +5
Wake 132 107 -25

A lot of folks may have thought Pom had Clemson too high, but his preseason ranking may have been right. Disappointing to see so many of the ACC's middle tier sliding down 7 or 8 spots, but most of them have big chances this week to put on an eye-opening performance.

-Jason "whew, that took a while... hope ya'll enjoy it ;)" Evans

Faustus
11-15-2010, 12:00 PM
I do appreciate the efforts Jason. I can also remember the days when the only stats around to see were the weekly AP and UPI top 20 (not even top 25) rankings that would show up in the newspapers on Tuesdays. (But then the players' uniforms had shorter pants then too.) The ebb and flow of ACC members could be very interesting to watch this season. There's one team at the top whose position I don't want to see ebb, however.

Indoor66
11-15-2010, 01:12 PM
Like last week, I will put a post in this thread with the KenPom and Sagarin rankings of each ACC team. Food for thought...



Team Sag Pom Dif
Duke 1 1 0
UNC 9 16 -7
Clemson 20 29 -9
Maryland 29 23 +6
Fla St 32 27 +5
Miami 46 46 0
Va Tech 51 25 +26
Ga Tech 53 78 -25
Virginia 58 57 +1
BC 63 38 +25
NC State 70 41 +29
Wake 90 132 -42

Additionally, We can look at how each school was ranked the week before and where they are now in the two most widely regarded computer rankings.


Sagarin Today Last wk Change
Duke 1 1 --
UNC 9 5 -4
Clemson 20 25 +5
Maryland 29 29 --
Fla St 32 32 --
Miami 46 47 +1
Va Tech 51 46 -5
Ga Tech 53 43 -10
Virginia 58 77 +19
BC 63 64 +1
NC State 70 65 -5
Wake 90 37 -53

Virginia is the big mover in the Saragin rankings. Wake's strangely high preseason ranking has been overtaken by their horrid first week results.


Pomeroy Today Last wk Change
Duke 1 1 --
UNC 16 12 -4
Maryland 23 24 -1
Va Tech 25 17 -8
Fla St 27 20 -7
Clemson 29 28 +1
BC 38 41 +3
NC State 41 33 -8
Miami 46 39 -7
Virginia 57 62 +5
Ga Tech 78 83 +5
Wake 132 107 -25

A lot of folks may have thought Pom had Clemson too high, but his preseason ranking may have been right. Disappointing to see so many of the ACC's middle tier sliding down 7 or 8 spots, but most of them have big chances this week to put on an eye-opening performance.

-Jason "whew, that took a while... hope ya'll enjoy it ;)" Evans

If this is going to be a regular thing (and I hope it is), shouldn't it have its own thread - maybe titled "Pomeroy-Sagarin Movement" or something?

JasonEvans
11-15-2010, 01:41 PM
If this is going to be a regular thing (and I hope it is), shouldn't it have its own thread - maybe titled "Pomeroy-Sagarin Movement" or something?

Done!

-JE

Indoor66
11-15-2010, 01:48 PM
Thanks. I think this will be a great addition to our enjoyment of the season.... It will for me!

juise
11-15-2010, 01:49 PM
I don't know. I think I'm going to miss the phrase "Dork Polls" in the title of the thread, like the one we had last year. :)

(Seriously, though... thanks for the effort, Jason.)

JasonEvans
11-16-2010, 12:53 PM
I don't know. I think I'm going to miss the phrase "Dork Polls" in the title of the thread, like the one we had last year. :)

(Seriously, though... thanks for the effort, Jason.)

Ask and ye shall receive!

-Jason

JasonEvans
11-16-2010, 01:48 PM
In the wake of their horrid performance last night, Ga Tech dropped from 78th in Pomeroy to 107th. They dropped from 53rd to 106 in the Sagarin rankings.

Yikes!

--Jason "is it too early to start the Paul Hewitt replacement list?" Evans

Bob Green
11-16-2010, 02:14 PM
--Jason "is it too early to start the Paul Hewitt replacement list?" Evans

Too early? That watch should've been started a couple of years ago. Once he is fired, perhaps Gary Williams should hire him to be lead recruiter at Maryland. It is a frightening thought to think about the success Williams could attain if he had Hewitt's recruits.

JasonEvans
11-16-2010, 02:55 PM
Too early? That watch should've been started a couple of years ago. Once he is fired, perhaps Gary Williams should hire him to be lead recruiter at Maryland. It is a frightening thought to think about the success Williams could attain if he had Hewitt's recruits.

Hewitt was in trouble about two years ago but he landed a huge recruiting class -- highlighted by Favors -- to preserve his job. Everyone said it was such a good class, he deserved the chance to coach it and see what happened.

Well, what happened was a good, but not nearly as good as everyone had expected year. It was not a bad enough season to have anyone say he should be fired, but it was not good enough to buy him the good will to survive much longer.

Unlike that situation, Hewitt does not have any saviors coming in his recruiting for next year. His best recruit is forward Julian Royal, a top 100 prospect, but not the kind of kid who can save your program (the way Favors could).

--Jason "if there was a dork poll for coaching, Hewitt would have an exceedingly low ranking right now" Evans

Im4howdy
11-16-2010, 03:04 PM
Jason-

As one who has been critical (more than once) of your excesses on this board, I have to publicly express my appreciation for your Dork Poll Tracking and look forward to following it in the future. Good job.

JasonEvans
11-17-2010, 04:33 PM
In the wake of their horrid performance last night, Ga Tech dropped from 78th in Pomeroy to 107th. They dropped from 53rd to 106 in the Sagarin rankings.

Apparently, Wake has decided to tell Ga Tech, "anything you can do, I can do worse!"

In the aftermath of Wake's horrid loss last night, they fell from--

Pomeroy - 132 to 152
Sagarin - 90 to 164

The KenPom move is not that huge but the Sagarin plunge is quite impressive. That takes some work! Of course, Sag had Wake waaaay too high in the preseason. This kind of correction is not all that surprising.

Still, even the bad ACC teams tend to only be in the 100 or so kind of range in these polls. To see an ACC team down in the 150s and 160s is really unusual.

Here is the lowest rated ACC team in the KenPom rankings in recent years--


2010 - Virginia - 76
2009 - Virginia - 105
2008 - NC St - 96
2007 - Wake - 87
2006 - Ga Tech - 96
2005 - FSU - 87
2004 - Clemson - 97
2003 - Clemson - 117

-Jason "sadly, Ken has no archived numbers for the woeful 2002 UNC club" Evans

hurleyfor3
11-17-2010, 04:42 PM
I don't know. I think I'm going to miss the phrase "Dork Polls" in the title of the thread, like the one we had last year. :)


*ahem* And who here popularized the term in the first place? You're welcome. :)

I have a hard time buying into the use of Bayesian priors for this kind of thing, but I guess Pomeroy is ok with them now, so who am I to argue. ("Bayesian prior" is dorkspeak for "guessing".)

SCMatt33
11-17-2010, 05:06 PM
Still, even the bad ACC teams tend to only be in the 100 or so kind of range in these polls. To see an ACC team down in the 150s and 160s is really unusual.

Here is the lowest rated ACC team in the KenPom rankings in recent years--


2010 - Virginia - 76
2009 - Virginia - 105
2008 - NC St - 96
2007 - Wake - 87
2006 - Ga Tech - 96
2005 - FSU - 87
2004 - Clemson - 97
2003 - Clemson - 117

-Jason "sadly, Ken has no archived numbers for the woeful 2002 UNC club" Evans

This could actually end up being a good thing as far as ACC perception goes. Whether it's right or not, the media (and sometimes members of the committee, even if they don't admit it), rates conferences on two things. First, they look at how many "top teams" a conference has. This generally means top 10-15 teams seen as legit Final Four contenders. If the Big East has 4 and the ACC has 3, it's seen as better for the Big East, even though both conferences have 25% of their teams in that group. The second is how many teams are in contention for a tourney berth. If the Big East has 8 teams on the bubble or better, it would be seen as better than 6 for the Pac-10, even though the Pac-10 has a much better proportion of it's league in contention.

Bottom teams are completely ignored in conference evaluation. The Big East has always been in the top conference conversation the last few years, even though it has had 4 teams (25% of its league) outside the top 100 on KenPom in 3 of the last 4 years. Last year, the Big Ten was seen as one of the strongest conferences despite having two teams ranked below 170 (the lower half of D-I)! BTW, neither one of those teams was named Penn State, who was 97. This is how the ACC has always looked good in the computers in the last five years, despite often being described as "down" by the media. The computers give fair weight to the worst teams.

If the ACC has a couple of punching bags who both win just 1 or 2 games in conference, it can help build the rest of the conference up. Last year, the worst teams were winning 4-5 games in conference and hanging "bad losses" on the middle of the conference, even if those losses shouldn't have been considered as bad as the bottom teams of another conference.

MChambers
11-17-2010, 05:26 PM
This could actually end up being a good thing as far as ACC perception goes. Whether it's right or not, the media (and sometimes members of the committee, even if they don't admit it), rates conferences on two things. First, they look at how many "top teams" a conference has. This generally means top 10-15 teams seen as legit Final Four contenders. If the Big East has 4 and the ACC has 3, it's seen as better for the Big East, even though both conferences have 25% of their teams in that group. The second is how many teams are in contention for a tourney berth. If the Big East has 8 teams on the bubble or better, it would be seen as better than 6 for the Pac-10, even though the Pac-10 has a much better proportion of it's league in contention.

I certainly agree with you, but note that Pomeroy predicts that behind Duke in the ACC will be five teams jammed at 9-7, not exactly the stuff of top 15 teams. Of course, someone may break out of the pack.

SCMatt33
11-17-2010, 05:45 PM
I certainly agree with you, but note that Pomeroy predicts that behind Duke in the ACC will be five teams jammed at 9-7, not exactly the stuff of top 15 teams. Of course, someone may break out of the pack.

Even without the top ten teams, it will help the ACC tremendously in category 2, as KenPoms predictions likely lead to 9 teams who are at least on the bubble, which would be a really giant number. The "top teams" factor is usually more important so you would need a few to break out the pack.

Also, I just saw that he predicts 7 conference wins between Gtech and Wake. For my scenario to play out, that number needs to be down at 4, including two games against each other.

ThePublisher
11-17-2010, 07:01 PM
Thanks for this. I check KenPom after every game. I find them to be much more reliable than sargin. I have a funny feeling that Duke will remain number 1 on pomeroy all year long. Just a hunch.

billyj
11-17-2010, 07:27 PM
Thanks for this. I check KenPom after every game. I find them to be much more reliable than sargin. I have a funny feeling that Duke will remain number 1 on pomeroy all year long. Just a hunch.

Ohio is looking awfully good though. They are not far behind.

loran16
11-17-2010, 07:51 PM
Ohio is looking awfully good though. They are not far behind.

Pretty sure Ohio lost to Oakland last week. (Oh you meant another school in the state?)

JasonEvans
11-17-2010, 10:34 PM
As I wrote in an email earlier today--

Sullinger is the nation's best freshman. If I had to put money on anyone other than Duke to win the national title, it would be The Ohio State University. I think they are the clear #2 and the team I fear the most.

-Jason "Thad has really built a quality program... and his recruiting is studly" Evans

BattierBattalion
11-18-2010, 09:04 AM
As I wrote in an email earlier today--

Sullinger is the nation's best freshman. If I had to put money on anyone other than Duke to win the national title, it would be The Ohio State University. I think they are the clear #2 and the team I fear the most.

-Jason "Thad has really built a quality program... and his recruiting is studly" Evans

They scored 93 points on 65 possessions at Florida. That's incredible.

loran16
11-20-2010, 08:17 PM
Well. After Yesterday Duke remains #1 (and #1 in both Offensive and Defensive efficiency), while UNC dropped to #8 (they'd been bumped up due to the Hofstra game.)

Meanwhile, potential opponent KState dropped to #20 (#24 on both O and D) after it's horrible game to Presbyterian.

MChambers
11-20-2010, 10:12 PM
Well. After Yesterday Duke remains #1 (and #1 in both Offensive and Defensive efficiency), while UNC dropped to #8 (they'd been bumped up due to the Hofstra game.)

Meanwhile, potential opponent KState dropped to #20 (#24 on both O and D) after it's horrible game to Presbyterian.
You're right, as far as Pomeroy goes, but somehow Duke dropped in Sagarin, at least in the blended rating. Still #1 in the Predictor, but #3 in the Rating.

It's over.

budwom
11-20-2010, 11:00 PM
Pomeroy and Sagarin are fun to check out, but seriously, this early in the season these kind of ratings mean zipsquat. In another ten days we'll have some much more meaningful data to pick over.

77devil
11-21-2010, 08:47 AM
You're right, as far as Pomeroy goes, but somehow Duke dropped in Sagarin, at least in the blended rating. Still #1 in the Predictor, but #3 in the Rating.

It's over.

Likely because the ELO Chess component considers the strength of the opponents but not the margin of victory. Pitt and WVU played strong teams in the their last game(or two for Pitt), while Duke has played cupcakes thus far. That will change dramatically starting Monday.

As Budworm notes below, as does Sagarin, the ratings are not statistically meaningful(aka theory of "zipsquat") at this point.


Pomeroy and Sagarin are fun to check out, but seriously, this early in the season these kind of ratings mean zipsquat. In another ten days we'll have some much more meaningful data to pick over.

tieguy
11-21-2010, 11:45 AM
Jason "sadly, Ken has no archived numbers for the woeful 2002 UNC club" Evans

Google cache has partial caches of Ken's data from 2002. Looks like they were 136 at the end of the season, though I seem to recall his methodology changed significantly in 200(4?) and so these numbers aren't directly comparable.

tieguy
11-21-2010, 11:49 AM
Likely because the ELO Chess component considers the strength of the opponents but not the margin of victory.

As Sagarin puts it, ELO is "politically correct"; I seem to recall that he only does it because otherwise he can't participate in the BCS rankings.

Note, though, that all he considers in the "better" rankings is the score. There is no understanding of the underlying structure of the game, which makes his numbers much less useful when comparing two teams who play the game at very different paces. Pomeroy takes that into account. So really Pomeroy > Sagarin Predictor >> Sagarin ELO.

JasonEvans
11-22-2010, 11:48 AM
Rankings as of Monday 11/22



Team Sag Pom Dif
Duke 7 1 -6
UNC 25 13 -12
Clemson 34 30 -4
Fla St 36 27 -9
Maryland 43 29 -14
NC State 49 37 -12
Va Tech 55 39 -16
Miami 94 60 -34
Virginia 97 68 -29
BC 113 52 -61
Ga Tech 121 81 -40
Wake 146 138 -8

What jumps out is how much higher Pomeroy thinks of the ACC than Sagarin does. There is not a single team that is ranked higher in Sag than it is in Pom. In some cases, the differences are quite stark. Pom sees Miami and BC as a bubble teams while Sag has them barely in the hunt for an NIT bid. Time will tell, but I have to say that the performance of the ACC thus far does not make me confident about the quality of the conference.

Now, here is how each team moved from week to week, starting with Sagarin.


Sagarin Today Last wk Change
Duke 7 1 -6
UNC 25 9 -16
Clemson 34 20 -14
Fla St 36 32 -4
Maryland 43 29 -14
NC State 49 70 +21
Va Tech 55 51 -4
Miami 94 46 -48
Virginia 97 58 -39
BC 113 63 -50
Ga Tech 121 53 -69
Wake 146 90 -56

Wow! Every single team in the conference moved down except NC State. That is a bad week! Many of the moves were dramatic. There are 5 ACC teams, almost half the conference, ranked lower than 90. That's stunning and quite unusual for this conference.


Pomeroy Today Last wk Change
Duke 1 1 --
UNC 13 16 +3
Fla St 27 27 --
Maryland 29 23 -6
Clemson 30 29 -1
NC State 37 41 +4
Va Tech 39 25 -14
BC 52 38 -14
Miami 60 46 -14
Virginia 68 57 -11
Ga Tech 81 78 -3
Wake 138 132 -6

I don't even know what to say about a ranking that thinks more highly of Carolina today than it did a week ago. I generally trust Ken's rankings a great deal but I find it mystifying that he could still be ranking UNC this high. Still, the obvious trend for the conference as a whole is clear. It was a bad week for the conference as most ACC teams went down in the rankings.

Well, that's all for now. As always, I welcome more analysis and discussion of this fun (but dorky) subject. Lets hope the conference fares a bit better in the coming week!

-Jason "Wake sucks royally, at least we can all agree on that!" Evans

JasonEvans
11-22-2010, 11:58 AM
One more thing worth noting -- the ACC was the #1 conference in Pomeroy's rankings last week. They are now #2 and the gap to the Big Ten is quite large. Sagarin had the ACC as the #2 conference a week ago. We are now #5. Blech!!

-Jason "aside from Duke, this is not looking like a very strong league" Evans

COYS
11-22-2010, 12:03 PM
Rankings as of Monday 11/22



Team Sag Pom Dif
Duke 7 1 -6
UNC 25 13 -12
Clemson 34 30 -4
Fla St 36 27 -9
Maryland 43 29 -14
NC State 49 37 -12
Va Tech 55 39 -16
Miami 94 60 -34
Virginia 97 68 -29
BC 113 52 -61
Ga Tech 121 81 -40
Wake 146 138 -8

What jumps out is how much higher Pomeroy thinks of the ACC than Sagarin does. There is not a single team that is ranked higher in Sag than it is in Pom. In some cases, the differences are quite stark. Pom sees Miami and BC as a bubble teams while Sag has them barely in the hunt for an NIT bid. Time will tell, but I have to say that the performance of the ACC thus far does not make me confident about the quality of the conference.

Now, here is how each team moved from week to week, starting with Sagarin.


Sagarin Today Last wk Change
Duke 7 1 -6
UNC 25 9 -16
Clemson 34 20 -14
Fla St 36 32 -4
Maryland 43 29 -14
NC State 49 70 +21
Va Tech 55 51 -4
Miami 94 46 -48
Virginia 97 58 -39
BC 113 63 -50
Ga Tech 121 53 -69
Wake 146 90 -56

Wow! Every single team in the conference moved down except NC State. That is a bad week! Many of the moves were dramatic. There are 5 ACC teams, almost half the conference, ranked lower than 90. That's stunning and quite unusual for this conference.


Pomeroy Today Last wk Change
Duke 1 1 --
UNC 13 16 +3
Fla St 27 27 --
Maryland 29 23 -6
Clemson 30 29 -1
NC State 37 41 +4
Va Tech 39 25 -14
BC 52 38 -14
Miami 60 46 -14
Virginia 68 57 -11
Ga Tech 81 78 -3
Wake 138 132 -6

Wow. I don't even know what to say about a ranking that thinks more highly of Carolina today than it did a week ago. I generally trust Ken's rankings a great deal but I find it mystifying that he could still be ranking UNC this high. Still, the obvious trend is clear. It was a bad week for the conference as most ACC teams went down in the rankings.

Well, that's all for now. As always, I welcome more analysis and discussion of this fun (but dorky) subject. Lets hope the conference fares a bit better in the coming week!

-Jason "Wake sucks royally, at least we can all agree on that!" Evans

I think Pomeroy at least gives UNC credit for blowing out Hofstra and posting decent defensive and offensive efficiency stats against other weak competition. Pomeroy is less concerned with wins and losses than overall efficiency. Minnesota, for example, has been pretty poor on offense and just decent enough on defense to have a pretty mediocre offensive/defensive efficiency differential, despite beating UNC. I think Pomeroy still sees a higher ceiling for UNC, which is probably true. That being said, if UNC continues to struggle against quality opponents, Kenpom's preseason predictors and the Heels' early season success against cupcakes will stop bailing them out.

The historical data Pomeroy uses is probably what is keeping the ACC afloat despite a horrible early season. It doesn't help that NCState had to play their toughest game without their veteran star.

DoubleDuke Dad
11-22-2010, 12:10 PM
Rankings as of Monday 11/22



Pomeroy Today Last wk Change
Duke 1 1 --
UNC 13 16 +3
Fla St 27 27 --
Maryland 29 23 -6
Clemson 30 29 -1
NC State 37 41 +4
Va Tech 39 25 -14
BC 52 38 -14
Miami 60 46 -14
Virginia 68 57 -11
Ga Tech 81 78 -3
Wake 138 132 -6

I don't even know what to say about a ranking that thinks more highly of Carolina today than it did a week ago.

-Jason "Wake sucks royally, at least we can all agree on that!" Evans

How many more losses in a row will it take before they are ranked #1?:rolleyes:

pfrduke
11-22-2010, 12:23 PM
The historical data Pomeroy uses is probably what is keeping the ACC afloat despite a horrible early season. It doesn't help that NCState had to play their toughest game without their veteran star.

This is the first season that the Pomeroy ratings have involved "bias" early in the season, rather than rely only on this season's numbers. My guess is that there will be some kinks that need to work themselves out, and it will improve over the next couple of years. But I think you're right that the bias is inflating the ACC's numbers.

tieguy
11-22-2010, 02:21 PM
I don't find the Pomeroy/Sagarin delta all that interesting (especially not at this point in the season), but a top-25 poll differential would be very interesting- there must be some teams that the coaches and media are over/underrating in the opinion of the dorks.

FWIW, I'm also working on shoving all this into Google Docs to visualize the change over the course of the season, but so far not enough data to make it work.

sagegrouse
11-22-2010, 02:26 PM
Congratulations to the OP and the posters for keeping perspective and making reflective and not heated comments. What Sagarin and KenPom are doing is to rig up early season projections for their data-driven models, when -- quite frankly -- there is no data.

Tune back in January 15.

sagegrouse

camion
11-22-2010, 02:35 PM
At this point in the season the dorks are using EWAGs (http://www.urbandictionary.com/define.php?term=EWAG) in their rankings just like the pollsters. I start paying attention to Kenpom and Sagarin about the time serious conference play starts in January.

Sagarin's version of the current dork disclaimer is:
"For the first few weeks of the season, the starting ratings have weight
in the process(BAYESIAN), but once the teams are all WELL CONNECTED, then
the starting ratings are no longer used and all teams are started equal
and the ratings are then done in an UNBIASED manner from that point on.
The teams are now NOT WELL CONNECTED and so the ratings are BAYESIAN."


Finally, I don't consider the RPI an accurate indicator of teams' relative strengths, ever. That isn't it's purpose. The formula shown below tells you something about a team's strength, but mainly it penalizes teams (like Va Tech) that play cupcake schedules.

From Wikipedia: (http://en.wikipedia.org/wiki/Ratings_Percentage_Index)
The current and commonly used formula for determining the RPI of a team at any given time is as follows.
RPI = (WP * 0.25) + (OWP * 0.50) + (OOWP * 0.25)
where WP is Winning Percentage, OWP is Opponents' Winning Percentage and OOWP is Opponents' Opponents' Winning Percentage.

loran16
11-22-2010, 02:52 PM
At this point in the season the dorks are using EWAGs (http://www.urbandictionary.com/define.php?term=EWAG) in their rankings just like the pollsters. I start paying attention to Kenpom and Sagarin about the time serious conference play starts in January.

Sagarin's version of the current dork disclaimer is:
"For the first few weeks of the season, the starting ratings have weight
in the process(BAYESIAN), but once the teams are all WELL CONNECTED, then
the starting ratings are no longer used and all teams are started equal
and the ratings are then done in an UNBIASED manner from that point on.
The teams are now NOT WELL CONNECTED and so the ratings are BAYESIAN."


Finally, I don't consider the RPI an accurate indicator of teams' relative strengths, ever. That isn't it's purpose. The formula shown below tells you something about a team's strength, but mainly it penalizes teams (like Va Tech) that play cupcake schedules.

From Wikipedia: (http://en.wikipedia.org/wiki/Ratings_Percentage_Index)
The current and commonly used formula for determining the RPI of a team at any given time is as follows.
RPI = (WP * 0.25) + (OWP * 0.50) + (OOWP * 0.25)
where WP is Winning Percentage, OWP is Opponents' Winning Percentage and OOWP is Opponents' Opponents' Winning Percentage.

No one is suggesting the Pomeroy or Sagarin rankings are in any way predictive right now, but they're interesting. Also, while they do take into account right now early season projections, at least Pomeroy's is based upon a statisical system, and both of their adjustments right now are done not by gut reaction (like the pollsters) but due to their mathematical formulas.

So if anything, they're at least as relevant as the current Top 25 polls, and Pomeroy is even more interesting, because it makes evaluations of specific factors of teams (O and Defensive Efficiency, Tempo, etc.).

No one is suggesting that they're in any way definitive or a really accurate method of classifying who the best team is yet, but they're better than the polls and the movements are interesting.

MChambers
11-22-2010, 03:04 PM
No one is suggesting the Pomeroy or Sagarin rankings are in any way predictive right now, but they're interesting. Also, while they do take into account right now early season projections, at least Pomeroy's is based upon a statisical system, and both of their adjustments right now are done not by gut reaction (like the pollsters) but due to their mathematical formulas.

So if anything, they're at least as relevant as the current Top 25 polls, and Pomeroy is even more interesting, because it makes evaluations of specific factors of teams (O and Defensive Efficiency, Tempo, etc.).

No one is suggesting that they're in any way definitive or a really accurate method of classifying who the best team is yet, but they're better than the polls and the movements are interesting.
Pomeroy's new system for starting ratings tries to take into account new players. One potential weakness in that is that freshmen in most cases will need time to adjust to college play. In other words, they'll be a lot better in March than November, but Pomeroy's new system may not reflect that in November.

loran16
11-22-2010, 03:47 PM
Pomeroy's new system for starting ratings tries to take into account new players. One potential weakness in that is that freshmen in most cases will need time to adjust to college play. In other words, they'll be a lot better in March than November, but Pomeroy's new system may not reflect that in November.

This applies for all players, not just freshmen. If pomeroy is overrating the freshmen at first, the teams in question will drop down as their results show that and then rise up as the freshmen reach potential. Same with other players. Really shouldn't make a difference...the ratings are NOT supposed to indicate how good teams are in March right now, but how good they are right now. They're not particularly reliable right now, but they're better than other things.

JasonEvans
11-22-2010, 10:30 PM
Another thing to note is that Pomeroy has indicated that as he gets more and more data, he weights the pre-season SWAG less and less. He says the SWAG won't be completely gone until early January but that it will account for just a tiny fraction of a percent of a team's total ranking by the time we get to January.

At this point, with most teams on game #4 or so, there is obviously a strong weighting of the SWAG, but it is diminishing and, I would bet, it won't be a very significant factor any more in just a couple weeks.

I can absolutely compare the Sag and KenPom numbers to the AP and ESPN polls. I'll start that next week... or maybe sooner.

I also love that tieguy (a welcome old-timer who is coming back STRONG!!) is going to try to put all this into a graph. In think it could be very interesting to watch. I wonder a bit about how it is going to look though with one team at #1 and another at #150. I would imagine it is going to be hard to see real movement on a graph where there is that wide a gap in the values of the data.

--Jason "as always, thanks to everyone for participating in this dorky exercise ;) " Evans

loran16
11-22-2010, 10:32 PM
Another thing to note is that Pomeroy has indicated that as he gets more and more data, he weights the pre-season SWAG less and less. He says the SWAG won't be completely gone until early January but that it will account for just a tiny fraction of a percent of a team's total ranking by the time we get to January.

At this point, with most teams on game #4 or so, there is obviously a strong weighting of the SWAG, but it is diminishing and, I would bet, it won't be a very significant factor any more in just a couple weeks.

I can absolutely compare the Sag and KenPom numbers to the AP and ESPN polls. I'll start that next week... or maybe sooner.

I also love that tieguy (a welcome old-timer who is coming back STRONG!!) is going to try to put all this into a graph. In think it could be very interesting to watch. I wonder a bit about how it is going to look though with one team at #1 and another at #150. I would imagine it is going to be hard to see real movement on a graph where there is that wide a gap in the values of the data.

--Jason "as always, thanks to everyone for participating in this dorky exercise ;) " Evans

Eh, last week The Preseason projections counted for 5 games. He says he's reducing the impact every week. So by next week, the preseason projections are gong to be outweighed byt he acutal games.

JasonEvans
11-23-2010, 07:34 AM
(note-- I meant to post this last night, but forgot to hit the submit key-- duuh!)

By the way, if anyone cares, Marquette was the #33 team in Pomeroy's rankings before tonight. Ken had forecast a final score of 80-67 (which would have been darn close to right had Duke not gaaacked up a fair amount of the lead in the final 30 seconds or so).

Just so we can see how it changes-- Marquette's overall rating was .8929. Its offensive rating was 112.0 (#27) and its defense was 93.1 (#49).

Duke was the #1 team with an overall rating of .9844. Duke's offense was 121.7 (#1) and the D was 84.8 (also #1).

--Jason "geek much?" Evans

77devil
11-23-2010, 07:55 AM
(note-- I meant to post this last night, but forgot to hit the submit key-- duuh!)

By the way, if anyone cares, Marquette was the #33 team in Pomeroy's rankings before tonight. Ken had forecast a final score of 80-67 (which would have been darn close to right had Duke not gaaacked up a fair amount of the lead in the final 30 seconds or so).

Just so we can see how it changes-- Marquette's overall rating was .8929. Its offensive rating was 112.0 (#27) and its defense was 93.1 (#49).

Duke was the #1 team with an overall rating of .9844. Duke's offense was 121.7 (#1) and the D was 84.8 (also #1).

--Jason "geek much?" Evans

Numbers have been updated and Marquette moved down a notch, although its rating edged up to .9009. Duke lost a little ground in all 3 ratings, but remained #1 in them.

I am surprised that Marquette's O and Duke's D #s didn't move a bit more given the Golden Eagles offensive efficiency in yesterday's game.

COYS
11-23-2010, 08:16 AM
Numbers have been updated and Marquette moved down a notch, although its rating edged up to .9009. Duke lost a little ground in all 3 ratings, but remained #1 in them.

I am surprised that Marquette's O and Duke's D #s didn't move a bit more given the Golden Eagles offensive efficiency in yesterday's game.

Marquette didn't manage a point per possession against us. While that is still significantly more than our other opponents, it is not particularly impressive. Our D allowed a little bit more per possession than it probably should have, but the 19 turnovers were the real culprits in hurting our Pomeroy stats. That killed our efficiency. Actually, we were pretty efficient when we were actually able to get a shot off without coughing it up (I'll take 55% shooting any night!). I'll say more in the post game thread, but I'm honestly not too worried about the turnovers at this point and would expect to see them come back down as the team gets accustomed to recognizing when it should play fast and when it should slow it down a bit.

gam7
11-23-2010, 02:35 PM
Do the offensive and defensive efficiency ratings take into account strength of schedule? That is, would a given set of offensive and defensive statistics be weighted more heavily against a good team than it would against a bad team (e.g., if Duke has an eFG% of 50% against Colgate, it's not as impressive as it is against K-State). Or, do you have to look at strength of schedule separately, as in "Team A is ranked #5 in offensive efficiency, but they've done it against a #1 ranked strength of schedule."

Also, what does the "pyth" measure actually mean in layman's terms?

loran16
11-23-2010, 03:17 PM
Do the offensive and defensive efficiency ratings take into account strength of schedule? That is, would a given set of offensive and defensive statistics be weighted more heavily against a good team than it would against a bad team (e.g., if Duke has an eFG% of 50% against Colgate, it's not as impressive as it is against K-State). Or, do you have to look at strength of schedule separately, as in "Team A is ranked #5 in offensive efficiency, but they've done it against a #1 ranked strength of schedule."

Also, what does the "pyth" measure actually mean in layman's terms?

Yes, the ADJUSTED O and D efficiency ratings take into account strength of schedule. That's a good reason why Duke was ranked #1 in offensive efficiency last year...as our our opponent's had the best D in the NCAA (the ACC was well, kind of good defensively). Pomeroy used to have listed the raw offensive and defensive efficiency numbers, which don't take into account opponent's, but I can't find it up there on the new version of the site.

Pyth (Short for Pythagorean, like the triangle. Long story.) is the result of a formula which takes the offensive and defensive efficiency to find an expected winning percentage over a regular season.

gam7
11-23-2010, 03:26 PM
Yes, the ADJUSTED O and D efficiency ratings take into account strength of schedule. That's a good reason why Duke was ranked #1 in offensive efficiency last year...as our our opponent's had the best D in the NCAA (the ACC was well, kind of good defensively). Pomeroy used to have listed the raw offensive and defensive efficiency numbers, which don't take into account opponent's, but I can't find it up there on the new version of the site.

Pyth (Short for Pythagorean, like the triangle. Long story.) is the result of a formula which takes the offensive and defensive efficiency to find an expected winning percentage over a regular season.

Thanks very much, Loran16. Wanted to give you a positive comment, but it told me that I had to spread around some other comments before giving you one.

loran16
11-23-2010, 04:37 PM
Thanks very much, Loran16. Wanted to give you a positive comment, but it told me that I had to spread around some other comments before giving you one.

Glad to help. (And i don't really care about the comment system, so don't worry about it.)

MChambers
11-24-2010, 08:21 AM
I'm not sure I can be as dorky as Jason, but I'll try. ;)

Duke is back to #1 in the two leading dork polls after last night's impressive win. (In Sagarin, we're still #1 in Predictor and #2 in Pure Chess, but #1 overall.)

JasonEvans
11-25-2010, 10:20 AM
I found this column interesting (http://kenpom.com/blog/index.php/weblog/what_can_win_distributions_do_for_you/)from Ken talking about how little we know about the season and how wide a range there is for team's results at this point.


Take Boston College for example. They lost to Yale last week. Yet if the ratings are accurate right now, there’s a 3% chance BC goes 11-5 in ACC. In 10,000 runs, there was one trial where they went 0-16 and one where they went 14-2. The model isn’t saying BC is going to improve or implode in those cases. It’s the same team that is currently rated seventh-best in the ACC providing that range of results.

--Jason "to me, this column hearkens back to 1994-95, when Duke was very good early in the year and then... well... you know" Evans

JasonEvans
11-25-2010, 03:16 PM
A cool function on Ken's site...

He currently gives Duke a 6.79% chance of going unbeaten in the regular season wth a 12.72% chance of going perfect in the ACC regular season.

That's about a 1-in-15 chance of unbeaten regular season, 1-in-8 in the ACC.

-Jason "insane? Perhaps, perhaps not" Evans

ice-9
11-26-2010, 04:36 AM
I don't like that kenpom uses preseason bias in his rankings. I understand his premise that preseason rankings are the most accurate way to predict postseason finish, but IMO only as it applies to the top 25, or perhaps even just the top 15. Past the top 25, the preseason poll loses its predictive power and I'd prefer if kenpom just kept his rankings numerically "pure" and let his readers make adjustments upwards or downwards.

superdave
11-26-2010, 09:49 AM
A cool function on Ken's site...

He currently gives Duke a 6.79% chance of going unbeaten in the regular season wth a 12.72% chance of going perfect in the ACC regular season.

That's about a 1-in-15 chance of unbeaten regular season, 1-in-8 in the ACC.

-Jason "insane? Perhaps, perhaps not" Evans

If you figure that we have 6-7 games this year where our opponent has, say, a 1/3 or better chance of beating us and we just won one of those games (K State), then our odds of going undefeated should have ticked up. If we beat Izzo's team then they should go up again a little, right?

The ACC is never as bad as the pundits enjoy imagining though.

arnie
11-26-2010, 10:46 AM
Just checked RPI and Duke was 24th behind such stalwarts Appalachian State and Vermont. App has a 1-1 record. I realize its very early, but any ideas as to how this ranking system could be that far off??

-jk
11-26-2010, 11:25 AM
Just checked RPI and Duke was 24th behind such stalwarts Appalachian State and Vermont. App has a 1-1 record. I realize its very early, but any ideas as to how this ranking system could be that far off??

RPI is a very different beast. It's less about what you do, and more about who you play. Or the reason VaTech missed the tourney last year.

-jk

SuperTurkey
11-26-2010, 11:39 AM
RPI is a very different beast. It's less about what you do, and more about who you play. Or the reason VaTech missed the tourney last year.

-jk

Normally, I would say that our RPI will shoot towards 1 once we enter conference play. Given the state of the ACC, that may not be so true this year.

loran16
11-26-2010, 02:04 PM
If you figure that we have 6-7 games this year where our opponent has, say, a 1/3 or better chance of beating us and we just won one of those games (K State), then our odds of going undefeated should have ticked up. If we beat Izzo's team then they should go up again a little, right?

The ACC is never as bad as the pundits enjoy imagining though.

KState was actually still predicted as like a 75% chance of winning for us.

http://kenpom.com/team.php?team=Duke


Most likely losses: (First Number is chance of losing, 2nd is to whom and when)
29% @UNC on 3/5
25% @FSU on 1/12
23% @MD on 2/2
20% @VT on 2/26
20% @NCSt on 1/19
14% @St.Johns on 1/30
14% @Miami on 2/13
13% MSU on 12/1 (First Home Game on list)
12% Butler on 12/4 (Neutral Game)
10% UNC on 2/9
8% Clemson on 3/2
8% @UVA on 2/16
7% Temple on 2/23
7% MD on 1/9
6% @Oregon on 11/27
6% NCSt. on 2/5
5% Boston College on 1/27
4% Miami on 1/2
4% @Wake on 1/22
2% St. Louis on 12/11
2% UAB on 1/5
2% UVA on 1/15
2% GT on 2/20
1% Bradley on 12/8
1% @UNCG on 12/29
0.3% Elon at home.
-------------------------------------
So yeah, we only have 5 games where kenpom predicts a 20% or higher chance of a loss, and non where the odds are 30%.

JasonEvans
11-26-2010, 04:41 PM
RPI is a very different beast. It's less about what you do, and more about who you play. Or the reason VaTech missed the tourney last year.

The RPI is purely mathematical and is made up of your win percentage (25%), your opponents' win percentage (50%) and your opponents' opponents' win percentage (25%).

Because it is only mathematical, there is no "pre-season" bias built in. As a result, at this early point in the season when there have only been a very limited number of games, the RPI is pretty darn haphazard and in no way indicative of who the best teams are. In fact, even at the end of the year, the RPI is not designed to show who the best team is. It is designed to measure which teams have played well against a tough schedule.

--Jason "if there was no preseason bias included in the other 'Dork Polls' you would see Sag and KenPom looking out-of-whack early in the year too" Evans

pantone287
11-27-2010, 03:12 AM
Early as can be, but Kansas at a very strong .9858 (above what we ended the year at last year) in Pomeroy's (http://kenpom.com/rate.php?y=2011) ratings.

The way they've played thus far plus the addition of Selby makes them a very serious source of concern. Caveats all over the place, but early results have 4 National Championship level squads in Kansas, Duke, Ohio State & Pitt and a pretty good drop off after that.

Sagarin (http://www.usatoday.com/sports/sagarin/bkt1011.htm) in general agreement about nat'l contenders but is less sold on Ohio State. Sagarin also has the teams more closely bunched.

Both systems working with assumptions about strength as part of their formulas at this point.

BattierBattalion
11-27-2010, 12:41 PM
Early as can be, but Kansas at a very strong .9858 (above what we ended the year at last year) in Pomeroy's (http://kenpom.com/rate.php?y=2011) ratings.



KU is definitely a contender, but they've only played cupcake teams so far. These rankings aren't going to be truly meaningful until conference games start.

loran16
11-27-2010, 01:47 PM
KU is definitely a contender, but they've only played cupcake teams so far. These rankings aren't going to be truly meaningful until conference games start.

How you play against Cupcakes is a very relevant measure of success. Yes conference play can reveal true talents as well, but being able to hold weaker teams to low scores and blow them out tells you whether you're a good team or not.

So don't discount that totally. That said, there are plenty of relevant non-cupcake games on everyone's non-conference schedule. Kansas' next 6 games are against:
#12 Arizona
#46 UCLA
#21 Memphis
#90 Colorado State
#72 USC
#67 Cal.

(All rankings are Pomeroy rankings. Arizona and Memphis are at true neutral sites). KU hasn't played anyone ranked better than 120 just yet for comparison. Duke has played two (#36 Marquette and #19KSU).

If KU is still ahead of Duke after that point, it'll be very relevant. And of course, they're playing WITHOUT Josh Selby....which supposedly will be a Kyrie Irving-like impact on their team.

Ultrarunner
11-27-2010, 02:19 PM
How you play against Cupcakes is a very relevant measure of success. Yes conference play can reveal true talents as well, but being able to hold weaker teams to low scores and blow them out tells you whether you're a good team or not.

So don't discount that totally. That said, there are plenty of relevant non-cupcake games on everyone's non-conference schedule. Kansas' next 6 games are against:
#12 Arizona
#46 UCLA
#21 Memphis
#90 Colorado State
#72 USC
#67 Cal.

(All rankings are Pomeroy rankings. Arizona and Memphis are at true neutral sites). KU hasn't played anyone ranked better than 120 just yet for comparison. Duke has played two (#36 Marquette and #19KSU).

If KU is still ahead of Duke after that point, it'll be very relevant. And of course, they're playing WITHOUT Josh Selby....which supposedly will be a Kyrie Irving-like impact on their team.

I'll be interested to see what happens against Cal and USC. They appear to the only defensively minded teams that KU will face whereas Duke has already played against several teams that are defensively oriented and look to slow the game down.

KU is currently ranked as the #1 team on defense. I expect that to change with better competition.

The difference in OOC SOS is stark. Duke is at 108. KU is at 226. That's a pretty big spread though it's pretty early in the season and I expect that to even out in the next month or so.

loran16
11-27-2010, 02:59 PM
I'll be interested to see what happens against Cal and USC. They appear to the only defensively minded teams that KU will face whereas Duke has already played against several teams that are defensively oriented and look to slow the game down.

KU is currently ranked as the #1 team on defense. I expect that to change with better competition.

The difference in OOC SOS is stark. Duke is at 108. KU is at 226. That's a pretty big spread though it's pretty early in the season and I expect that to even out in the next month or so.

Do remember that SoS is accounted for in the efficiency ratings. Now the SoS (and thus the efficiency ratings) might be skewed by a small sample size, but if KU's numbers are better, it's their numbers including sample size.

BattierBattalion
11-27-2010, 04:28 PM
How you play against Cupcakes is a very relevant measure of success. Yes conference play can reveal true talents as well, but being able to hold weaker teams to low scores and blow them out tells you whether you're a good team or not.

So don't discount that totally. That said, there are plenty of relevant non-cupcake games on everyone's non-conference schedule. Kansas' next 6 games are against:
#12 Arizona
#46 UCLA
#21 Memphis
#90 Colorado State
#72 USC
#67 Cal.

(All rankings are Pomeroy rankings. Arizona and Memphis are at true neutral sites). KU hasn't played anyone ranked better than 120 just yet for comparison. Duke has played two (#36 Marquette and #19KSU).

If KU is still ahead of Duke after that point, it'll be very relevant. And of course, they're playing WITHOUT Josh Selby....which supposedly will be a Kyrie Irving-like impact on their team.

You're absolutely right. What I should have said was that we just haven't had enough games for the KenPom stuff to be meaningful.

CLW
11-28-2010, 10:07 AM
For whatever it is worth, we have fallen to #2 behind Kansas in Ken Pom's ratings.

http://kenpom.com/rate.php

tieguy
11-28-2010, 01:10 PM
For whatever it is worth, we have fallen to #2 behind Kansas in Ken Pom's ratings.

We've also fallen behind them in Sagarin.

The interesting thing is that we'd fallen behind KU before we crushed Oregon and KU struggled against Arizona, and despite that they stayed ahead of us. Obviously it is still early (and we should get a nice bump if we beat MSU, less so Butler) but clearly Kenpom's numbers are liking the very, very big thumpings KU has been giving lesser opponents of late.

TexHawk
11-29-2010, 01:12 AM
We've also fallen behind them in Sagarin.

The interesting thing is that we'd fallen behind KU before we crushed Oregon and KU struggled against Arizona, and despite that they stayed ahead of us. Obviously it is still early (and we should get a nice bump if we beat MSU, less so Butler) but clearly Kenpom's numbers are liking the very, very big thumpings KU has been giving lesser opponents of late.
To be fair, "struggling" against Arizona in an 8 point win is not shocking. Kenpom moved them up to #5, even after the loss. Given the state of the Pac10, they could very well end up with a fantastic record and a Top 2-3 seed in March. Derrick Williams is a stud.

And if you're looking for a place for KU to regress, it's on offense, not defense. Final Kenpom rankings for KU's defensive efficiency the last 5 years: 2, 1, 1, 7, and 8.

JasonEvans
11-29-2010, 08:49 AM
Pom currently gives Duke a 7.62% chance of going unbeaten in the regular season.
He gives Kansas an 11.58% chance of doing the same.

Ohio State and Pitt have much smaller chances (2% and less than 1% each) due to the tougher nature of the Big Ten and Big East.

--Jason "can you imagine the hype if both teams were able to do this?!?!" Evans

JasonEvans
12-02-2010, 10:12 AM
Ooops! I totally forgot to do this on Monday. Been a busy week.

Anyway, I did save the numbers on Monday morning so the numbers I am about to post are all prior to the ACC-Big Ten Challenge results.

Without any further ado, here are the...

Rankings as of Monday 11/29



Team Sag Pom Dif
Duke 1 2 +1
UNC 24 22 -2
Clemson 34 32 -2
Va Tech 43 46 -3
Fla St 44 29 -15
Maryland 50 35 -15
BC 57 40 -17
NC State 66 49 -17
Miami 78 68 -10
Ga Tech 88 76 -12
Virginia 128 91 -37
Wake 179 152 -27

Just like last week, Pomeroy likes the ACC a lot more than Sagarin does. I don't mind this as I think Pom's numbers are better and seem more accurate than Sagarin's. Sag only sees 3 ACC teams as solid tournament teams at this point (ranked in top 40) while Pom has 6.

Now, here is how each team moved from week to week, starting with Sagarin.


Sagarin Today Last wk Change
Duke 1 7 +6
UNC 24 25 +1
Clemson 34 34 ---
Va Tech 43 55 +12
Fla St 44 36 -8
Maryland 50 43 -7
BC 57 113 +56
NC State 66 49 -17
Miami 78 94 +16
Ga Tech 88 121 +33
Virginia 128 97 -31
Wake 179 146 -33

Impressive move back into relevance by BC and, to a lesser extent, GA Tech. Stunnign to see Wake drop that much after being at #146 last week. I doubt an ACC team has ever been ranked this close to #200 in the nation. Just pathetic.

Ok, here is a look at the weekly movement in the Pomeroy--


Pomeroy Today Last wk Change
Duke 2 1 -1
UNC 22 13 -9
Fla St 29 27 -2
Clemson 32 30 -2
Maryland 35 29 -6
BC 40 52 +12
Va Tech 46 39 -7
NC State 49 37 -12
Miami 68 60 -8
Ga Tech 76 81 +5
Virginia 91 68 -23
Wake 152 138 -14

Despite Pom liking the ACC more than Sag does, almost everyone in the conference saw their ranking slip during the week. Not good. Again, BC stands out as the team that had a good week.


Again, sorry to have forgotten to get to this earlier in the week. The numbers feel a tad stale as a result. For example, NC State's embarrassment at the hands of Wiscy are not reflected in these results. Anyway, we'll see how the Big 10 challenge results impact the conference next week ;)

-Jason "in the next few weeks I'll look at how teams have moved compared to their pre-season rankings" Evans

gw67
12-02-2010, 10:23 AM
After getting beat up by the Big Ten, the ACC is now the 4th rated conference by Pomeroy and the 5th rated conference by Sagarin (after the Big Ten, Big East, Big 12 and [yikes!] the Mountain West). I expect the ACC to maintain this lowly position as long as the teams picked by the ACC sportswriters to finish near the top of the league (UNC, Virginia Tech, Florida State) continue to lose to OOC teams. Duke continues to be ranked #1 by Sagarin and #2 by Pomeroy. There are no other ACC teams in the top 25 of either rating service. Maryland is the next best ACC team. The Terps are 30th in Sagarin and 26th in Pomeroy. I’ve watched several of the Terps’ games and, as of now, their weaknesses outnumber their strengths.

gw67

JasonEvans
12-02-2010, 04:34 PM
After getting beat up by the Big Ten, the ACC is now the 4th rated conference by Pomeroy and the 5th rated conference by Sagarin (after the Big Ten, Big East, Big 12 and [yikes!] the Mountain West). I expect the ACC to maintain this lowly position as long as the teams picked by the ACC sportswriters to finish near the top of the league (UNC, Virginia Tech, Florida State) continue to lose to OOC teams. Duke continues to be ranked #1 by Sagarin and #2 by Pomeroy. There are no other ACC teams in the top 25 of either rating service. Maryland is the next best ACC team. The Terps are 30th in Sagarin and 26th in Pomeroy. I’ve watched several of the Terps’ games and, as of now, their weaknesses outnumber their strengths.

Interesting to note that The Mountain West, while highly regarded by Sagarin, remains a second tier conference in the Pomeroy rankings. Ken has the MW 7th, fairly far back behind the #6 Pac Ten.

In Pomeroy's rankings, the Big Ten is the clear #1 with a rating of .9056. The Big 12, Big East, and ACC are closely bunched after that at .8791, .8786, ,8704. Then there is a gap to the SEC (.8304) and Pac Ten (.8283). The Mountain West comes in with a rating of .7784.

-Jason "who knows how much preseason bias is still in these-- probably not all that much" Evans

sagegrouse
12-02-2010, 04:42 PM
After getting beat up by the Big Ten, the ACC is now the 4th rated conference by Pomeroy and the 5th rated conference by Sagarin (after the Big Ten, Big East, Big 12 and [yikes!] the Mountain West). I expect the ACC to maintain this lowly position as long as the teams picked by the ACC sportswriters to finish near the top of the league (UNC, Virginia Tech, Florida State) continue to lose to OOC teams. Duke continues to be ranked #1 by Sagarin and #2 by Pomeroy. There are no other ACC teams in the top 25 of either rating service. Maryland is the next best ACC team. The Terps are 30th in Sagarin and 26th in Pomeroy. I’ve watched several of the Terps’ games and, as of now, their weaknesses outnumber their strengths.

gw67

Not to quibble, but the ACC was one bonehead Virginia Tech play away from beating the Big Ten in the Challenge. Holy cow, Hokies, finish the job!!!!! As it was, the final outcome was 6-5. I don't call this "getting beat up by the Big Ten."

Of course, if I were an NC State fan, I certainly wouldn't argue with the term, since "beat up" is not as bad as "poleaxed" or "annihilated."

sagegrouse

sagegrouse

pantone287
12-02-2010, 06:29 PM
Interesting to note that The Mountain West, while highly regarded by Sagarin, remains a second tier conference in the Pomeroy rankings. Ken has the MW 7th, fairly far back behind the #6 Pac Ten.

In Pomeroy's rankings, the Big Ten is the clear #1 with a rating of .9056. The Big 12, Big East, and ACC are closely bunched after that at .8791, .8786, ,8704. Then there is a gap to the SEC (.8304) and Pac Ten (.8283). The Mountain West comes in with a rating of .7784.



In Sagarin system Duke would be favored by only 4 against St. Mary's on a neutral floor, would be underdogs at San Diego St. Vegas would tend to disagree.

Early season issue factors into both systems, but Pomeroy better passes the smell test.

tieguy
12-02-2010, 06:41 PM
Early season issue factors into both systems, but Pomeroy better passes the smell test.

Pomeroy just had a post about that: http://kenpom.com/blog/index.php/weblog/prediction_tracker/

He attributes the better early-season performance to the offseason steps that were taken to "preseed" the system with recruiting data, last year's performance, etc.

uh_no
12-02-2010, 09:09 PM
Not to quibble, but the ACC was one bonehead Virginia Tech play away from beating the Big Ten in the Challenge. Holy cow, Hokies, finish the job!!!!! As it was, the final outcome was 6-5. I don't call this "getting beat up by the Big Ten."

Of course, if I were an NC State fan, I certainly wouldn't argue with the term, since "beat up" is not as bad as "poleaxed" or "annihilated."

sagegrouse

sagegrouse

I think the argument can be made that despite the final tally being close, the ACC was blown out in several games and mostly won their games by a small margin....indicating that if there was ever a sort of round robin amongst the teams, the big 10 would win by a large margin

UVA 87 Minn 79 +8
WF 76 IA 73 +3
GT 71 NW 91 -20
FSU 44 OSU 58 -14
clem 61 mich 69 -8
nc 67 ill 79 -12
BC 88 ind 76 +12
NCST 48 wisc 87 -39
VT 55 Purd 58 -3
md 62 PSU 39 +23
Duke 84 msu 79 +5
Total: B10 by 55 points

they won 4 of their games by 10 points or more, 2 by 20 or more, and 1 by 30 or more
in contrast the ACC won 2 by 10 or more and 1 by 20 or more

All in all, this tournament doesn't mean much since it is based on last years rankings.....here's why

lets say we have OSU wins the big 10 one year and loses 7 of their best players to the NBA draft and has no good incoming freshman, so despite being best in the league last year, they are effectively last this year (suppose for the moment that all other teams are about the same and that the best team int eh big 10 is equal to the best in the acc, and so and so down the line)

therefore OSU by virtue of winning last year would get their tails kicked by the best team in the ACC, and the new best team in the big10 would play the second best team in the ACC and beat them (ans so and so down the line) meaning the big 10 would win 10-1 despite the conferences being equivalent

now in theory the OSU game would be a huge blowout win by the big east (lets say 50 points) and all the other games would be close (<10 pts) so despite the big 10 winning 10-1, the point total would be about even

now OBVIOUSLY this is an absolutely ridiculous scenario which could not play out in reality, but it points to the errors of judging overall conference quality based on a head to head tournament. In any case, you get a better picture of conference comparability (regardless of individual matchups) by looking at the overall scoring margin, and in that metric, the big10 blew us out this year. It is also highly indicative that the teams regarded as being in the top of the league lost

UNC, VT, NCST, FSU all lost, with only VT being close to winning, (yeah these are preseason rankings, which someone will call me out on....but w/e)

so effectively what this tournament showed is that the acc has all of 1 team that can hang with the top 5-6 teams in the big 10 (2 if you count VT)....and maybe you can say maryland would have hung in there, but there's no way to know atm, and that all our wins (except duke) came from teams in the bottom half of the conference, where the teams are fighting over the scraps of relevency anyway and don't have a huge impact on the conference 'goodness'

so once I've looked at the numbers this way, it seems impossible to make an argument that the ACC is even remotely close to the Big 10 this year and that if there was a tournament where every team in the big 10 played every team in the big east, i think the big 10 might win 80 or more of the 132 games

let the games continue! things can change

whether the top of the big east can rival the big 10 is a wholly different question

Kedsy
12-02-2010, 11:50 PM
I think the argument can be made that despite the final tally being close, the ACC was blown out in several games and mostly won their games by a small margin....indicating that if there was ever a sort of round robin amongst the teams, the big 10 would win by a large margin

UVA 87 Minn 79 +8
WF 76 IA 73 +3
GT 71 NW 91 -20
FSU 44 OSU 58 -14
clem 61 mich 69 -8
nc 67 ill 79 -12
BC 88 ind 76 +12
NCST 48 wisc 87 -39
VT 55 Purd 58 -3
md 62 PSU 39 +23
Duke 84 msu 79 +5
Total: B10 by 55 points

they won 4 of their games by 10 points or more, 2 by 20 or more, and 1 by 30 or more
in contrast the ACC won 2 by 10 or more and 1 by 20 or more

All in all, this tournament doesn't mean much since it is based on last years rankings.....here's why

lets say we have OSU wins the big 10 one year and loses 7 of their best players to the NBA draft and has no good incoming freshman, so despite being best in the league last year, they are effectively last this year (suppose for the moment that all other teams are about the same and that the best team int eh big 10 is equal to the best in the acc, and so and so down the line)

therefore OSU by virtue of winning last year would get their tails kicked by the best team in the ACC, and the new best team in the big10 would play the second best team in the ACC and beat them (ans so and so down the line) meaning the big 10 would win 10-1 despite the conferences being equivalent

now in theory the OSU game would be a huge blowout win by the big east (lets say 50 points) and all the other games would be close (<10 pts) so despite the big 10 winning 10-1, the point total would be about even

now OBVIOUSLY this is an absolutely ridiculous scenario which could not play out in reality, but it points to the errors of judging overall conference quality based on a head to head tournament. In any case, you get a better picture of conference comparability (regardless of individual matchups) by looking at the overall scoring margin, and in that metric, the big10 blew us out this year. It is also highly indicative that the teams regarded as being in the top of the league lost

UNC, VT, NCST, FSU all lost, with only VT being close to winning, (yeah these are preseason rankings, which someone will call me out on....but w/e)

so effectively what this tournament showed is that the acc has all of 1 team that can hang with the top 5-6 teams in the big 10 (2 if you count VT)....and maybe you can say maryland would have hung in there, but there's no way to know atm, and that all our wins (except duke) came from teams in the bottom half of the conference, where the teams are fighting over the scraps of relevency anyway and don't have a huge impact on the conference 'goodness'

so once I've looked at the numbers this way, it seems impossible to make an argument that the ACC is even remotely close to the Big 10 this year and that if there was a tournament where every team in the big 10 played every team in the big east, i think the big 10 might win 80 or more of the 132 games

let the games continue! things can change

whether the top of the big east can rival the big 10 is a wholly different question

Well, first of all, if you add up the numbers on your chart it only comes to +45, not +55, for the Big Ten. And 39 of those 45 points were in one game.

Putting that aside, I think you're wrong about the tourney being based on last year's rankings. In fact, I always heard the coaches from both conferences voted on the rankings and then the two leagues matched up. Even if I'm wrong about that, why would Maryland, who finished tied for first in the ACC last season, be playing Penn State? Why would UNC, who finished near the bottom of the league last year, be playing Illinois? I think the matchups were supposed to be as even as possible, and that your 10 to 1 example doesn't fit the facts.

As far as them blowing us out and us winning close games, if you take the NC State outlier game out of the equation, then the difference is only 6 points in the Big Ten's favor, over 10 games, which is pretty much even. So I think it is possible to make the argument. Especially since Wisconsin is so good at home and State was playing without their best player. If Smith plays and the game's on a neutral court, it's probably a pretty close game, and then there's almost no way to make the argument that the Big 10 is clearly better.

Just depends on how you look at it.

pfrduke
12-03-2010, 12:26 AM
Yeah, the only way last year's rankings come into play is that the 12th place ACC team doesn't get the chance to play in the challenge.

JasonEvans
12-06-2010, 11:29 AM
Rankings as of Monday 12/6



Team Sag Pom Dif
Duke 1 2 +1
UNC 28 28 --
BC 32 34 +2
Maryland 50 27 -23
Miami 56 54 -2
Fla St 57 31 -26
Virginia 65 80 +15
Clemson 70 44 -26
Va Tech 75 50 -25
NC State 76 68 -8
Ga Tech 115 86 -29
Wake 207 132 -75

The trend continues as Pomeroy continues to like the ACC a lot more than Sagarin does. There are 6 ACC teams who are 20+ spots higher in Pom than they are in Sagarin. Virginia is the only ACC team who is ranked meaningfully higher in Sag than they are in Pom.

Now, here is how each team moved from week to week, starting with Sagarin.


Sagarin Today Last wk Change
Duke 1 1 --
UNC 28 24 -4
BC 32 57 +25
Maryland 50 50 --
Miami 56 78 +22
Fla St 57 44 -13
Virginia 65 128 +63
Clemson 70 34 -36
Va Tech 75 43 -32
NC State 76 66 -10
Ga Tech 115 88 -27
Wake 207 179 -28

Another week where most of the league drops in the rankings. Sag now only sees 3 ACC teams being worthy of the NCAA tourney. With the number of non-conference games left dwindling, it is going to become hard for ACC teams to distinguish themselves as in-conference victories (other than teams who beat Duke) are just not going to be as meaningful as they have been in other seasons.

BC's move is really impressive, they were ranked #113 just 2 weeks ago. It is obviously easier to make a big move early in the season when there are fewer data points but this is still a nice indication of how well BC has played lately. Virginia's move is the other positive one that jumps off the page. The Cavs still have a ways to go, but their past couple games certainly have put them into the conversation about mid-level ACC teams as opposed to being a team chained to the bottom of the conference.

I continue to be amazed at how bad Wake is. For them to drop into the 200s is just shocking. ACC teams are never rated that low!

Ok, here is a look at the weekly movement in the Pomeroy--


Pomeroy Today Last wk Change
Duke 2 2 --
Maryland 27 35 +8
UNC 28 22 -6
Fla St 31 29 -2
BC 34 40 +6
Clemson 44 32 -12
Va Tech 50 46 -4
Miami 54 68 +14
NC State 68 49 -19
Ga Tech 86 76 -10
Virginia 80 91 +11
Wake 132 152 +20

I am not sure if this is a sign of a better or worse ranking model, but you can see that there is a bit more stability in the Pom rankings as opposed to Sagarin. It is tough for a team to make a 20+ place movement in Pom while 30+ movements are almost common for Sagarin.

Also worth noting that because there are more ACC teams ranked reasonably high in Pom, there are a few more chances for ACC teams to post impressive wins.

-Jason "sigh, the ACC is just so uninspiring this year" Evans

roywhite
12-06-2010, 11:38 AM
So Pomeroy now has Kansas #1, and that's without Josh Selby, who will join the team soon.

Hmmmm...

superdave
12-06-2010, 11:44 AM
So Pomeroy now has Kansas #1, and that's without Josh Selby, who will join the team soon.

Hmmmm...

Where's Crimson& Blue? We may need some insight. I have not seen KU play yet this year. Selby should be back on the court soon.

loran16
12-07-2010, 01:11 AM
Pomeroy's player rankings are now up! Some interesting things:

Andre Dawkins is #6 NATIONALLY in Offensive efficiency, however in limited usage (Usage is not characterized as minutes, but % of possessions used).

Duke has not made up for Zoubek's offensive boarding talents, as the following is how our offensive boarding% leaders are:
Mason Plumlee: 12.2% (189th in the Country)
Miles Plumlee: 13.1%
Ryan Kelly: 7.2%
Josh Hairston (Very small sample size): 7.7%
Kyle Singler: 7.5%

(For Comparison: Miami's Reggie Johnson leads the entire COUNTRY in OR% at 23.0%. John Henson is at 16%, while Jordan Williams is at 13.6%. Tyler Zeller is only at 9.1%, with Barnes at 8.7%)

Another Interesting things is seeing the %Used of players in the ACC. Malcolm Delaney (28.5%) and Reggie Jackson of BC (28.3%) lead the ACC in how often a player is used in a possession (the #s are the amounts of possessions used). For Comparison sake, the most used BCS player is UConn's Kemba Walker (#18- 32.8%)

Duke on the other hand, is really really spread out: Nolan Smith is #5 in conference in usage at 27.5%, then Kyrie Irving is at 25.2% (15th in conference), Kyle Singler is at 21.2% (#31 in conference), Mason Plumlee at 21.0% (#33), Seth Curry at 18% (#52), and Andre Dawkins at 13.2%
---Yes you read that right, when Seth is in the game, he touches the ball more often than Andre Dawkins. These are interesting numbers...one would figure Kyle's usage should increase and Nolan's Decrease.

Oh and in-conference, with a minimum of 20% of possessions used, Kyrie Irving is #2 in Offensive Rating, a metric that measures the overall efficiency of a player on the court on offense. Nolan Smith shows up at #11 and Kyle Singler at #13. For comparison's sake, Tyler Zeller shows up at #5, Harrison Barnes at #27, and John Henson at #33 (The FTs KILL Henson).

Other Random Stats:
Seth Curry is #5 in Steals' Percentage in the ACC. The top 8 in that stat in the ACC oddly enough includes FIVE Georgia Tech players (at #s 2-4 and #s 7-8). Chris Singleton unsurprisingly leads in this statistic.

Two Dukies, Kyrie Irving #4 and Nolan Smith #10, are in the top 10 in the ACC in Fouls Drawn. Not really a surprise there.

Andre Dawkins is #2 in the ACC in Turnover %, with Curry at #9, Singler at #14, Irving at #34, Smith at #42, and Mason at #45. For comparison sake, Larry Drew comes in at #69.

Kyrie Irving comes in at #3 in the ACC and Nolan Smith at #5 in Assist Rate.

pantone287
12-07-2010, 01:53 AM
In terms of our competitors on the national scene: Kansas's (http://kenpom.com/team.php?team=Kansas) worst 2-pt shooter is Markieff Morris at .553. Wish I had a little more insight into how they've been managing so proficiently to fill it up. With Selby on the horizon they continue to loom out on the prairie.

On the home front, Kyle's rebounding percentages are down by about a third compared to past years, which given his evident assiduity around the hoop seems like a case of unlucky bounces, but given our mediocre board work thus far something to keep an eye on.

Disappointing to see Mason and Miles down at 100 offensive efficiency and below. When you see the flashes that each has shown recently its hard not to be optimistic, but turnovers and (in Miles's case) fouls continue to present themselves to be whittled down.

Kyrie and Dre have been unbelievable. Kyle and Nolan's #s look excellent, particularly given each has had stretches of early uneven play.

Kedsy
12-07-2010, 11:20 AM
Pomeroy's player rankings are now up! Some interesting things:

Duke has not made up for Zoubek's offensive boarding talents, as the following is how our offensive boarding% leaders are:
Mason Plumlee: 12.2% (189th in the Country)
Miles Plumlee: 13.1%
Ryan Kelly: 7.2%
Josh Hairston (Very small sample size): 7.7%
Kyle Singler: 7.5%

(For Comparison: Miami's Reggie Johnson leads the entire COUNTRY in OR% at 23.0%. John Henson is at 16%, while Jordan Williams is at 13.6%. Tyler Zeller is only at 9.1%, with Barnes at 8.7%)


If Jordan Williams is at 13.6, then Miles Plumlee's 13.1 seems pretty impressive for a guy who most people around here think is seriously underachieving.

MChambers
12-07-2010, 11:33 AM
If Jordan Williams is at 13.6, then Miles Plumlee's 13.1 seems pretty impressive for a guy who most people around here think is seriously underachieving.
I think Miles's play is a lot like Zoubek's last year -- doing the little things that win, but not always looking smooth, especially on the offensive end. So he doesn't pass the eye test of the fans, but he's actually quite effective. It would be great if he could make a similar leap to the one Brian made last February, especially as to fouls.

Kedsy
12-07-2010, 11:34 AM
I think Miles's play is a lot like Zoubek's last year -- doing the little things that win, but not always looking smooth, especially on the offensive end. So he doesn't pass the eye test of the fans, but he's actually quite effective. It would be great if he could make a similar leap to the one Brian made last February, especially as to fouls.

I agree. I think fouls are the main thing holding him back at this point.

Delaware
12-07-2010, 11:45 AM
Interesting that before the season started KenPom had what we all thought might be low in PPG for this team.
For reference before the season started the projection of PPG for Duke was 79.6. Through 8 games the actual PPG is 89.3 and the new projection (actual for 8 games plus further projections) is now at 84.9.

The data shows Duke (10.4%) and Kansas (9.4%) as having about a 1 in 10 chance of going undefeated hrough the regular season, pretty high. There are three teams (includng Duke, Kansas and Ohio St.) all projected to win every individual game. Pitt is projected to lose only one game.

MChambers
12-07-2010, 12:01 PM
Interesting that before the season started KenPom had what we all thought might be low in PPG for this team.
For reference before the season started the projection of PPG for Duke was 79.6. Through 8 games the actual PPG is 89.3 and the new projection (actual for 8 games plus further projections) is now at 84.9.

The data shows Duke (10.4%) and Kansas (9.4%) as having about a 1 in 10 chance of going undefeated hrough the regular season, pretty high. There are three teams (includng Duke, Kansas and Ohio St.) all projected to win every individual game. Pitt is projected to lose only one game.

I think KenPom's projections on points and tempo were simply based on the last 2 or 3 years and not adjusted to reflect roster changes, so it's not surprising that his projections were low.

loran16
12-07-2010, 12:25 PM
Interesting that before the season started KenPom had what we all thought might be low in PPG for this team.
For reference before the season started the projection of PPG for Duke was 79.6. Through 8 games the actual PPG is 89.3 and the new projection (actual for 8 games plus further projections) is now at 84.9.

The data shows Duke (10.4%) and Kansas (9.4%) as having about a 1 in 10 chance of going undefeated hrough the regular season, pretty high. There are three teams (includng Duke, Kansas and Ohio St.) all projected to win every individual game. Pitt is projected to lose only one game.


I think KenPom's projections on points and tempo were simply based on the last 2 or 3 years and not adjusted to reflect roster changes, so it's not surprising that his projections were low.

This is correct. In preseason, our tempo (pace) was still a below average pace (slow). Now Duke's pace is around 10 possessions higher per game (actually i think it was about 8, but whatever), putting us as the 21st fastest team in the country.

Extra 10 possessions plus a team that scores at least 1.15 per possession = an additional 11.5 points per game.

loran16
12-07-2010, 12:34 PM
I think Miles's play is a lot like Zoubek's last year -- doing the little things that win, but not always looking smooth, especially on the offensive end. So he doesn't pass the eye test of the fans, but he's actually quite effective. It would be great if he could make a similar leap to the one Brian made last February, especially as to fouls.


If Jordan Williams is at 13.6, then Miles Plumlee's 13.1 seems pretty impressive for a guy who most people around here think is seriously underachieving.

First the Zoubek comparison doesn't work. In Zoubek's Freshman through Sophomore years, Zoubek's Offensive Rebounding % was over 16. For comparison's sake, being over 16 would've put Zou in the top ten of the COUNTRY, had he played enough minutes to qualify. 13% barely puts you into the top 100. The two players are very different...Miles is only similar in that he fouls a lot this year and plays a big man position.

Second, the Miles and Kelly #s are small sample sizes, and are likely to fluctuate wildly.



Of Note: Small Sample size here with Miles.

JasonEvans
12-07-2010, 12:52 PM
There are three teams (includng Duke, Kansas and Ohio St.) all projected to win every individual game. Pitt is projected to lose only one game.

Yeah, but some of the projections are fairly close.

For example, among the games in which Ohio St are favored are 2 games where they are a slight 62% fave (at Illinois and at Wisconsin) and one where they are a 64% fave (at Purdue). Frankly, winning at Wisconsin is really, really tough and I would make OSU an underdog in that game, for sure. Kohl is one of the 3-5 toughest places in America to win, IMO. OSU is fortunate that the unbalanced Big Ten sked allows them to play Mich State at home as I think they would be an underdog if the game was played on the road.

Kansas only has one game where they are less than a 70% fave -- when they go to Baylor they are only a 64% fave. Their next closest game is the one at KSt in which Pom says Kansas is a 73% favorite.

As for Duke, there are zero games left in which we are seen as less than a 70% favorite. Pom says our best chances to lose are the games at Maryland (74%), at UNC (74%), and at Florida St (76%). The other really risky game is at St John in which we are only an 80% favorite. Impressive how well Lavin has St John playing so far (though I think they may fall back a bit in BEast play).

--Jason "you won't see Duke's unbeaten percentage move a lot over the next few games until the New Year as they are largely 99% chance games" Evans

MChambers
12-07-2010, 01:34 PM
First the Zoubek comparison doesn't work. In Zoubek's Freshman through Sophomore years, Zoubek's Offensive Rebounding % was over 16. For comparison's sake, being over 16 would've put Zou in the top ten of the COUNTRY, had he played enough minutes to qualify. 13% barely puts you into the top 100. The two players are very different...Miles is only similar in that he fouls a lot this year and plays a big man position.
I think I may not have been clear. I wasn't trying to say that Miles is a good an offensive rebounder as Zoubs, although he is a decent offensive rebounder. I was just trying to say that, as with Zoubs, the team is quite effective when he's on the floor, even if he doesn't always pass the eye test, and that it would be great if he could reduce his fouls.

SilkyJ
12-07-2010, 02:07 PM
Duke on the other hand, is really really spread out: Nolan Smith is #5 in conference in usage at 27.5%, then Kyrie Irving is at 25.2% (15th in conference), Kyle Singler is at 21.2% (#31 in conference), Mason Plumlee at 21.0% (#33), Seth Curry at 18% (#52), and Andre Dawkins at 13.2%
---Yes you read that right, when Seth is in the game, he touches the ball more often than Andre Dawkins. These are interesting numbers...one would figure Kyle's usage should increase and Nolan's Decrease.


Great post! Lot's of interesting info and stats in there. Since much of it has been expounded on, I thought I'd touch on this part.

This doesn't surprise me at all (I doubt it surprises anyone). With all the weapons we have, the ball is going to be spread around. Additionally, your last point Re: Seth's usage vs Andre's, that shouldn't surprise anyone either. Seth is a reserve ball handler for us and a playmaker. He'll play some "PG" in the sense of running plays, using ball-screens etc. Andre, while he is playing great, is still mostly a spot up shooter. Sure he can come off screens/curls, use pump-fakes, etc., but we aren't going to put ball-screens out there for him and have him create. Not his game.

phaedrus
12-07-2010, 07:05 PM
Additionally, your last point Re: Seth's usage vs Andre's, that shouldn't surprise anyone either. Seth is a reserve ball handler for us and a playmaker. He'll play some "PG" in the sense of running plays, using ball-screens etc. Andre, while he is playing great, is still mostly a spot up shooter. Sure he can come off screens/curls, use pump-fakes, etc., but we aren't going to put ball-screens out there for him and have him create. Not his game.

On top of that, I think Seth has spent more time out there with our second team than Andre and Andre has spent more time with our first team than Seth. Thus Seth tends to be higher on the food chain while he's out there, while Andre is sharing shots with Kyle, Kyrie, and Nolan.

tieguy
12-09-2010, 10:03 AM
"you won't see Duke's unbeaten percentage move a lot over the next few games until the New Year as they are largely 99% chance games"

Interestingly, we did just that last night, because Pomeroy's computer liked our defensive effort so much last night even against a mediocre opponent. We went from 5th to 4th defensively, and increased our lead in the offensive rating, which was enough to push us back to #1 (over Kansas) and raise our unbeaten record chance to nearly 14%. His graph is also showing our most likely record as 30-1 now.

Obviously, we have to take this all with a grain of salt (for example, the ratings take no account of Kyrie being out) but interesting to see what can happen by beating up on even a very lightly rated opponent. If we do the same to higher-ranked St. Louis we may see our undefeated probability pass 15%.

loran16
12-09-2010, 11:58 AM
More interestingly, Duke played its slowest game (65 possessions) of the year yesterday, perhaps unsurprisingly.

Should be interesting to see how much we slow down.

jipops
12-09-2010, 01:04 PM
Obviously, we have to take this all with a grain of salt (for example, the ratings take no account of Kyrie being out)

yeah, shouldn't the Luck factor go way down into the negative now?

Double DD
12-09-2010, 02:37 PM
In football, beating up on bad or medicore teams is a better predictor of future success than a close win over a good team. I'm not sure if the same holds true for college basketball, but I'd bet it would to a certain extent and so crushing teams like Bradley without Kyrie is a good sign.

CLW
12-09-2010, 09:33 PM
yeah, shouldn't the Luck factor go way down into the negative now?

Ironically SLU is the "unluckiest" team in college hoops according to Ken Pom.

Perhaps they have the cure for turf toe in the Billiken's belly?

tieguy
12-10-2010, 02:16 AM
In football, beating up on bad or medicore teams is a better predictor of future success than a close win over a good team. I'm not sure if the same holds true for college basketball, but I'd bet it would to a certain extent and so crushing teams like Bradley without Kyrie is a good sign.

Basketball Prospectus (which everyone should subscribe to) recently had a related article in the NBA context: http://basketballprospectus.com/article.php?articleid=1319

Delaware
12-12-2010, 12:44 PM
Yeah, but some of the projections are fairly close.

For example, among the games in which Ohio St are favored are 2 games where they are a slight 62% fave (at Illinois and at Wisconsin) and one where they are a 64% fave (at Purdue). Frankly, winning at Wisconsin is really, really tough and I would make OSU an underdog in that game, for sure. Kohl is one of the 3-5 toughest places in America to win, IMO. OSU is fortunate that the unbalanced Big Ten sked allows them to play Mich State at home as I think they would be an underdog if the game was played on the road.

Kansas only has one game where they are less than a 70% fave -- when they go to Baylor they are only a 64% fave. Their next closest game is the one at KSt in which Pom says Kansas is a 73% favorite.

As for Duke, there are zero games left in which we are seen as less than a 70% favorite. Pom says our best chances to lose are the games at Maryland (74%), at UNC (74%), and at Florida St (76%). The other really risky game is at St John in which we are only an 80% favorite. Impressive how well Lavin has St John playing so far (though I think they may fall back a bit in BEast play).

--Jason "you won't see Duke's unbeaten percentage move a lot over the next few games until the New Year as they are largely 99% chance games" Evans

Jason.... I would have predicted the same thing that you did above... that Duke's unbeaten percentage would not move up much... but we would have both been wrong. As of this morning.... Duke's projected record and chance of unbeaten season on KenPom:
Projected record: Overall = 29-2 ACC = 15-1
Chance of unbeaten record: Overall: 19.82% ACC=23.80%

It was only about 10% a week ago, now it is almost 20%! I don't understand it other than all the others on our schedule have gotten a lot worse and the individual percentages for each game have gone up?

MChambers
12-12-2010, 01:27 PM
Jason.... I would have predicted the same thing that you did above... that Duke's unbeaten percentage would not move up much... but we would have both been wrong. As of this morning.... Duke's projected record and chance of unbeaten season on KenPom:
Projected record: Overall = 29-2 ACC = 15-1
Chance of unbeaten record: Overall: 19.82% ACC=23.80%

It was only about 10% a week ago, now it is almost 20%! I don't understand it other than all the others on our schedule have gotten a lot worse and the individual percentages for each game have gone up?
Although the last two games were games Duke was projected to win easily, Duke was even more efficient (even given the opponents) than it has been most of the season. So our efficiency numbers went up. I think much of our opposition has been struggling, too, as you suggest, but I think the main thing is that Duke has been more efficient, on both ends.

I'd still rather have Kyrie, of course, but it is interesting that the team in some ways is playing even better without him.

mr. synellinden
12-12-2010, 03:21 PM
Although the last two games were games Duke was projected to win easily, Duke was even more efficient (even given the opponents) than it has been most of the season. So our efficiency numbers went up. I think much of our opposition has been struggling, too, as you suggest, but I think the main thing is that Duke has been more efficient, on both ends.

I'd still rather have Kyrie, of course, but it is interesting that the team in some ways is playing even better without him.

Even though it's a small sample size, the games without Irving are somewhat revealing and, I think, confirm what a lot of fans were noticing - that the new offensive style of running at every opportunity was leading to a lot of erratic and sloppy play. Now, it may be that this was inevitable in the sense that there is always going to be a learning curve when adjusting to a new style and that during the course of the season we would have become even more efficient and comfortable with the "high octane" offense.

However, it may be that even with Irving, our offense is better suited to a more traditional half court/motion offense with selective fast breaks. When you have five deadly three point shooters and the most potent drive, draw and dish point guard Duke has had since Williams (and maybe ever), you should be almost unguardable on offense. What limits your advantage is turning the ball over too much. A more controlled tempo should theoretically cut down on turnovers and allow for the "talent advantage" to have more of an impact. As an example, despite Irving's phenomenal game, we probably would have lost at home to Michigan St. if they hadn't turned the ball over so many times, a lot of which were unforced turnovers, and shot FTs better. If we had lost, it would have been because WE turned the ball over too many times.

I understand wanting to increase the number of possessions when you have so much offensive firepower, but that strategy becomes a poor one if it makes the team less efficient per possession.

I am by no means questioning Coach K's strategy in implementing more of a running offense. I am just curious to see how our offensive efficiency numbers during the next 10 games or so compare to the first 8. And if our team is more efficient in a more traditional offense, I wonder if Coach K will change his strategy if and when Irving comes back.

Think of playing closer to last year's style but adding Curry and a much improved Dawkins - swapping Thomas/Zoubek for Plumlee, Plumlee and Kelly (better offense, not as good defense but better shot blocking, less offensive rebounds) - and, of course, adding Irving for Scheyer which gives us a more disruptive point guard who can break down a defense and create an easy basket or get fouled.

Of course I hope Irving comes back soon and we get to see if there is any change in strategy.

Bob Green
12-12-2010, 03:40 PM
I am just curious to see how our offensive efficiency numbers during the next 10 games or so compare to the first 8. And if our team is more efficient in a more traditional offense, I wonder if Coach K will change his strategy if and when Irving comes back.

Good post! We will not see the full effect of playing without Irving until the level of competition increases. Over our first eight games, with Irving, we beat Top 10 opponents Michigan State and Kansas State. Knocking off Bradley and St. Louis just isn't going to tell us a lot about life without Irving. Over the next 10 games you mention, there are exactly zero Top 25 teams scheduled. However, we will play tough conference road games at Florida State and at N.C. State. By the time 2/2/11 rolls around, and we head off to College Park to play Maryland, if Irving is still sidelined, we should have a pretty clear picture of the team's style and capabilities.

camion
12-12-2010, 09:46 PM
Here's an interesting titdbit from kenpom rankings after Saturday's games.

119 - Long Beach State.
120 - St. Louis.

Just a dork morsel to munch on.

Duvall
12-12-2010, 10:18 PM
Here's an interesting titdbit from kenpom rankings after Saturday's games.

119 - Long Beach State.
120 - St. Louis.

Just a dork morsel to munch on.

That's still skewed by preseason ratings, though. Saint Louis was expected to have a much better team coming into the season than the one they have been able to put on the floor, and their computer ratings are inflated because of it.

AlaskanAssassin
12-12-2010, 10:29 PM
Can anyone tell me what the smaller size double digit number next to the projected scores mean? I can't seem to figure it out. thanks

loran16
12-12-2010, 10:32 PM
That's still skewed by preseason ratings, though. Saint Louis was expected to have a much better team coming into the season than the one they have been able to put on the floor, and their computer ratings are inflated because of it.

Errrr, not really. Preseason rankings are not hugely impacting things too much at this point. They originally counted for 5 games' worth of value, but their weight has dropped each weak an unspecified amount. LBSU was ranked 103 to start the year, while St. Louis was ranked 95. Both have dropped, Stl. Louis by a far greater amount, but the two were always projected to be similarly meh to bad teams.

loran16
12-12-2010, 10:33 PM
Can anyone tell me what the smaller size double digit number next to the projected scores mean? I can't seem to figure it out. thanks

That's the pace (Tempo) of the game in total possessions each team will get.

AlaskanAssassin
12-12-2010, 10:41 PM
gotcha! thanks


That's the pace (Tempo) of the game in total possessions each team will get.

DukieTiger
12-13-2010, 12:22 AM
Pomeroy is now giving Duke almost 20% chance of going undefeated. Thought that was interesting. Just drives home that Duke has probably played its toughest opponents (although toughest games will probably be @FSU, @NCST, @MD, @VT, @UNC) of the year at this point in the schedule. He gives Duke almost 24% chance to run the table in the ACC.

I also thought it was interesting that Duke hopped back over Kansas to reclaim the #1 spot when Irving went out.

-jk
12-13-2010, 10:14 AM
Pomeroy is now giving Duke almost 20% chance of going undefeated. Thought that was interesting. Just drives home that Duke has probably played its toughest opponents (although toughest games will probably be @FSU, @NCST, @MD, @VT, @UNC) of the year at this point in the schedule. He gives Duke almost 24% chance to run the table in the ACC.

I also thought it was interesting that Duke hopped back over Kansas to reclaim the #1 spot when Irving went out.

I can only surmise "High Octane Freshman PG" is semantically equivalent to "Mistakes Will Happen".

-jk

JasonEvans
12-13-2010, 11:19 AM
Rankings as of Monday 12/13



Team Sag Pom Dif
Duke 1 1 --
UNC 28 28 --
BC 31 37 +6
Maryland 45 25 -20
Miami 49 57 +12
Fla St 52 35 -17
NC State 68 71 +3
Va Tech 69 50 -19
Clemson 70 47 -23
Virginia 72 85 +13
Ga Tech 152 88 -65
Wake 229 155 -74

Pom continues to think more highly of the ACC than Sag does, though the gaps are getting a bit smaller in some cases.

Sagarin Rankings


Sagarin Today Last wk Change
Duke 1 1 --
UNC 28 28 --
BC 31 32 +1
Maryland 45 50 +5
Miami 49 56 +7
Fla St 52 57 +5
NC State 68 76 +8
Va Tech 69 75 +6
Clemson 70 70 --
Virginia 72 65 -7
Ga Tech 152 115 -37
Wake 229 207 -22

Wow, the ACC actually had a good week in the Sag rankings?!?!?! Everyone in the middle tier improved a bit. It is still stunning to see how poorly GaTech and especially Wake are rated. Most years, the bottom of the ACC is in the 90-110 range. I can't recall seeing an ACC team rated lower than 150. Seeing one in the mid-200s... stunning.

Allow me to put Wake's #229 ranking in perspective -- Wake would be the 2nd worst team in the Ivy League. They would be the 8th place team in the Southland Conference. They'd be 7th place in the Summit League (who?!?!). Just sad. Not typical of the ACC.


Pomeroy


Pomeroy Today Last wk Change
Duke 1 2 +1
Maryland 25 37 +12
UNC 28 28 --
Fla St 35 31 -4
BC 37 34 -3
Clemson 47 44 -3
Va Tech 50 50 --
Miami 57 54 -3
NC State 71 68 -3
Virginia 85 80 -5
Ga Tech 88 86 -2
Wake 155 132 -23

Maryland is really the only ACC team to make a meaningful move in the Pomeroy rankings this past week, going from a bubble team to a solid tournament pick and a top 25 ranking. Interesting that the Pom numbers have stabilized for now. Sure, there were not all that many games last week, but it is still interesting that the movements were so small.

On real issue I have-- there is just no way Ga Tech is so close to Virginia at this point. No way! Virginia is playing far better ball.

--Jason "whew, this takes some work... but I am a dork so I enjoy it" Evans

Bob Green
12-13-2010, 12:06 PM
Maryland is really the only ACC team to make a meaningful move in the Pomeroy rankings this past week, going from a bubble team to a solid tournament pick and a top 25 ranking.

Maryland moves up in the rankings from 37 to 25, while Boston College drops from 34 to 37. Who won the BC - Maryland game? Color me confused.

JasonEvans
12-13-2010, 04:11 PM
Maryland moves up in the rankings from 37 to 25, while Boston College drops from 34 to 37. Who won the BC - Maryland game? Color me confused.

I must admit, it does seem counter-intuitive. But, I think that is what makes Ken's ratings so impressive. Wins and losses are not all that matters. He is largely looking at how effective and efficient you are on offense and defense going by the assumption that the most efficient teams are the ones who, over the course of many games, will be the winners more often than not.

It may also be a sign of bunching in the rankings. I am not familiar enough with Ken's numbers to say for certain what a wide margin is between teams, but the following is worth noting--

The difference between the #1 team (Duke) and the #5 team (Kentucky) is .0266 in Ken's ratings. That is greater than the difference between Kentucky and the #20 team (West Virginia). The gap from #1 to #5 is larger than the gap from #5 to #20. Similarly, the gap between Maryland, #25 at .9291, and BC, #37 at .9052, is .0239 -- less than that same gap between Duke and Kentucky.

Of course, much of what that may prove is that Duke (and Kansas and maybe Ohio State) are really a cut above all the rest of the teams around, even the other supposedly good teams like Kentucky, Syracuse, and the such.

-Jason "I feel really dorky when I dissect differences between .9236 and .9164 ;) " Evans

sammy3469
12-13-2010, 04:27 PM
I must admit, it does seem counter-intuitive. But, I think that is what makes Ken's ratings so impressive. Wins and losses are not all that matters. He is largely looking at how effective and efficient you are on offense and defense going by the assumption that the most efficient teams are the ones who, over the course of many games, will be the winners more often than not.

It may also be a sign of bunching in the rankings. I am not familiar enough with Ken's numbers to say for certain what a wide margin is between teams, but the following is worth noting--

The difference between the #1 team (Duke) and the #5 team (Kentucky) is .0266 in Ken's ratings. That is greater than the difference between Kentucky and the #20 team (West Virginia). The gap from #1 to #5 is larger than the gap from #5 to #20. Similarly, the gap between Maryland, #25 at .9291, and BC, #37 at .9052, is .0239 -- less than that same gap between Duke and Kentucky.

Of course, much of what that may prove is that Duke (and Kansas and maybe Ohio State) are really a cut above all the rest of the teams around, even the other supposedly good teams like Kentucky, Syracuse, and the such.

-Jason "I feel really dorky when I dissect differences between .9236 and .9164 ;) " Evans

I believe MD was actually 24 or 25 before the BC. In any case BC dropped a little because they were a little less efficient offensively than they usually are (for instance they only had a 48% FG when they were frequently above 55%). Basically they're so good offensively (and depend on it so much) that when they are just a little less than perfect, it hurts them.

MChambers
12-13-2010, 04:35 PM
I believe MD was actually 24 or 25 before the BC. In any case BC dropped a little because they were a little less efficient offensively than they usually are (for instance they only had a 48% FG when they were frequently above 55%). Basically they're so good offensively (and depend on it so much) that when they are just a little less than perfect, it hurts them.
The changes also could be due to changes in the ratings of the teams they've previously placed. For example, Temple beat Georgetown, so maybe that helps Maryland a little.

camion
12-13-2010, 07:58 PM
Sagarin:

'The teams are now WELL CONNECTED and so the ratings are UNBIASED."

Not as accurate as they will be later in the season, but no longer Bayesian. That Bayes (http://en.wikipedia.org/wiki/Thomas_Bayes) guy didn't know squat about basketball.

hurleyfor3
12-14-2010, 11:26 AM
Sagarin:

'The teams are now WELL CONNECTED and so the ratings are UNBIASED."

Not as accurate as they will be later in the season, but no longer Bayesian. That Bayes (http://en.wikipedia.org/wiki/Thomas_Bayes) guy didn't know squat about basketball.

I'm not really seeing how the Mormons end up second. Although they are seventh under Pomeroy.

Kedsy
12-14-2010, 12:02 PM
I'm not really seeing how the Mormons end up second. Although they are seventh under Pomeroy.

BYU did really well in last year's computer ratings also (9 in Sagarin, pre-NCAAT; 7 in Pomeroy, pre-NCAAT).

tieguy
12-14-2010, 12:34 PM
Obviously this is disconnected from reality a bit, given Kyrie's toe, but our schedule has weakened enough in Pomeroy's computer brain since Saturday that he now predicts us to go 30-1, and we've crossed the 20% barrier for an undefeated season. I admit I'm going to continue to find this fascinating until such time as we actually lose.

JasonEvans
12-14-2010, 06:24 PM
As we periodically do, lets have a look at how Pomeroy projects the ACC to go this season...

Duke 15-1
FSU 10-6
BC 10-6
Maryland 9-7
UNC 9-7
Clemson 8-8
Va Tech 7-9
Miami 7-9
NCSU 6-10
Virginia 6-10
Ga Tech 6-10
Wake 3-13

In case of a tie, I ranked the team with the higher KenPom ranking higher.

A few comments --

1) If BC goes 10-6 then Steve Donohue will win Coach of the Year hand down... even if Duke goes unbeaten.

2) Would 7-9 get Seth fired at Va Tech? His seat would be getting warm, that's for sure.

3) El Sid would be gonzo if they go 6-10. He might be gonzo if they go 8-8 and miss the NCAAs.

4) Obviously, we do not know the scenario under which these teams are reaching these records, but this sure looks like a situation where the ACC would get 5 NCAA bids.

--Jason "other thoughts?" Evans

JasonEvans
12-17-2010, 11:02 AM
Wow! Here is a great column/blog post by Ken Pomeroy (http://kenpom.com/blog/index.php/weblog/the_head-to-head_fallacy/)on the fallacy that whichever teams wins a head-to-head matchup is the better team.

He looks at teams that played twice in the same season and found the following--


Home teams that won by one or two points were 16-52 in the rematches, winning just 23.5% of the time. Most fans like to think the results of a close game as just because teams that emerge victorious show grittiness, heart, and toughness. But those teams were almost certain losers when they faced the same team on the road. What happened to the grittiness then? To me, there’s no greater statement to the influence of luck in the outcome of a close game than the struggles of close home winners in a road game against the same team.

If you still doubt, consider this: The average margin for the winners of the first game in those cases was +1.6 points. The average margin for those teams in the rematch was -6.9, almost exactly double what many studies have shown to be typical home court advantage. Was there anything more going on for the close home winners than just simply being at home? It doesn’t appear that way.

--Jason "really cool stuff... the gist is -- lose a close game on the road and it pretty much proves you are the better team" Evans

ice-9
12-18-2010, 06:21 AM
Wow! Here is a great column/blog post by Ken Pomeroy (http://kenpom.com/blog/index.php/weblog/the_head-to-head_fallacy/)on the fallacy that whichever teams wins a head-to-head matchup is the better team.

He looks at teams that played twice in the same season and found the following--

--Jason "really cool stuff... the gist is -- lose a close game on the road and it pretty much proves you are the better team" Evans


Great article! Also goes to reinforce just how important scoring margin is. I wish I had data to support that statement, but looking at win-loss records can be deceiving. Winning by a lot or by a little matters a great deal. You can say a 2 point win is basically playing a team even; someone just has to win. That's something the NCAA selection committee should look at...

Ultrarunner
12-18-2010, 09:11 PM
Looking at today's Pomeroy - which is becoming addictive! - I noticed that Johnny Dawkins has a 0 percent chance for an unbeaten season. Stands to reason since Stanford already has two losses.

Tommy Amaker, however, has a 2.96 percent chance of being unbeaten. And while I think the world of Tommy Amaker, I think that Pomeroy might have this a bit high considering Harvard already has two losses.

But I'm sipping whiskey and giddy over the Carolina game plus I'm probably not nearly as bright as Ken Pomeroy (please note the subtle understatement), so what would I know?

Go Tommy! And Johnny!

loran16
12-18-2010, 10:15 PM
Looking at today's Pomeroy - which is becoming addictive! - I noticed that Johnny Dawkins has a 0 percent chance for an unbeaten season. Stands to reason since Stanford already has two losses.

Tommy Amaker, however, has a 2.96 percent chance of being unbeaten. And while I think the world of Tommy Amaker, I think that Pomeroy might have this a bit high considering Harvard already has two losses.

But I'm sipping whiskey and giddy over the Carolina game plus I'm probably not nearly as bright as Ken Pomeroy (please note the subtle understatement), so what would I know?

Go Tommy! And Johnny!

That 2.96% is the chance of going unbeaten in conference.

JasonEvans
12-18-2010, 10:50 PM
Looking at today's Pomeroy - which is becoming addictive! - I noticed that Johnny Dawkins has a 0 percent chance for an unbeaten season. Stands to reason since Stanford already has two losses.

Tommy Amaker, however, has a 2.96 percent chance of being unbeaten. And while I think the world of Tommy Amaker, I think that Pomeroy might have this a bit high considering Harvard already has two losses.

But I'm sipping whiskey and giddy over the Carolina game plus I'm probably not nearly as bright as Ken Pomeroy (please note the subtle understatement), so what would I know?

Go Tommy! And Johnny!

The whiskey got to you.

Pom is predicting Tommy has a 2.96% chance of going undefeated in the Ivy League conference. If you look, the 2.96% is under the projected conference record. Under the projected overall record, there is nothing because there is no chance of Harvard going unbeaten on the season as they already have 2 losses.

Stanford, with no games played yet in the Pac 10, does have a chance of going unbeaten in the conference. However, that chance rounds to 0.00% (meaning it is something less than 0.004%) so it looks like a zero.

Duke, as another example, has a 19.48% chance of going unbeaten on the entire regular season and a 23.78% chance of going unbeaten in the ACC. With non-conference games left against Elon, UNCG, UAB, St. John, and Temple; one could interpret that data to mean that there is slightly more than a 4% chance we will lose one of those non-conference games.

--Jason "question not, the wisdom of the Pom!" Evans

P.S. - I know Loran16 already mentioned the conference thing, but I wanted to give a more thorough explanation.

SuperTurkey
12-18-2010, 10:58 PM
However, that chance rounds to 0.00% (meaning it is something less than 0.004%) so it looks like a zero.

Don't you mean 'something less than 0.005%?' 0.00499999999% is greater than 0.004% and still rounds to 0.00%, while 0.005% rounds to 0.01%.

Ultrarunner
12-18-2010, 11:15 PM
The whiskey got to you.

Pom is predicting Tommy has a 2.96% chance of going undefeated in the Ivy League conference. If you look, the 2.96% is under the projected conference record. Under the projected overall record, there is nothing because there is no chance of Harvard going unbeaten on the season as they already have 2 losses.

Stanford, with no games played yet in the Pac 10, does have a chance of going unbeaten in the conference. However, that chance rounds to 0.00% (meaning it is something less than 0.004%) so it looks like a zero.

Duke, as another example, has a 19.48% chance of going unbeaten on the entire regular season and a 23.78% chance of going unbeaten in the ACC. With non-conference games left against Elon, UNCG, UAB, St. John, and Temple; one could interpret that data to mean that there is slightly more than a 4% chance we will lose one of those non-conference games.

--Jason "question not, the wisdom of the Pom!" Evans

P.S. - I know Loran16 already mentioned the conference thing, but I wanted to give a more thorough explanation.

Thank you. I'll retire to other activities now.

jipops
12-18-2010, 11:20 PM
Wow! Here is a great column/blog post by Ken Pomeroy (http://kenpom.com/blog/index.php/weblog/the_head-to-head_fallacy/)on the fallacy that whichever teams wins a head-to-head matchup is the better team.

He looks at teams that played twice in the same season and found the following--



--Jason "really cool stuff... the gist is -- lose a close game on the road and it pretty much proves you are the better team" Evans

Let's hope we don't have to play Mich St again then.

JasonEvans
12-19-2010, 05:44 PM
Let's hope we don't have to play Mich St again then.

We beat them by 5. I don't think that is quite the kind of close game Pomeroy is talking about.

Also, Pomeroy's numbers all involve home-and-home series. If we were to play MSU again, it would not be at MSU.

--Jason "I'd be happy to face MSU again as it would almost certainly mean we had advanced pretty far in the NCAA tourney" Evans

tieguy
12-19-2010, 07:15 PM
--Jason "I'd be happy to face MSU again as it would almost certainly mean we had advanced pretty far in the NCAA tourney" Evans

That's #15-pomeroy-ranked, almost-lost-to-Oakland MSU? Or some other MSU that is a serious threat to advance in March? I know Izzo is good, and I know this team can probably play better than they currently are playing, but I'm not seeing them as a great team yet, and I still don't think our win over them was that impressive.

(That said, they are the toughest team we'll see until the Sweet 16, given how weak the ACC looks.)

juise
12-19-2010, 10:47 PM
Stanford, with no games played yet in the Pac 10, does have a chance of going unbeaten in the conference. However, that chance rounds to 0.00% (meaning it is something less than 0.004%) so it looks like a zero.

http://verydemotivational.files.wordpress.com/2010/10/2f0ef064-30c9-4482-bd57-cb4de74cc9cb.jpg
(Someone had to. Well, maybe someone didn't have to... but I did.)

JasonEvans
12-21-2010, 10:04 AM
Rankings as of Tuesday 12/21



Team Sag Pom Dif
Duke 1 1 --
BC 34 42 +8
UNC 35 25 -10
Fla St 46 31 -15
Miami 54 66 +12
Maryland 57 23 -24
Va Tech 62 41 -21
Virginia 70 85 +15
Clemson 80 59 -21
NC State 82 73 -9
Ga Tech 91 77 -14
Wake 239 176 -63

There continues to be a strong anti-ACC bias in the Sagarin rankings... either that or Pom has a strong pro-ACC bias ;) The gaps that had narrowed a bit last week have stretched out again. Sag really thinks the ACC is one great team and a few bubble teams. The difference in the Maryland rankings are especially striking. Pom sees the Terps as a Top 25 club while Sag doesn't even think they should make the NCAA tournament.

Sagarin Rankings


Sagarin Today Last wk Change
Duke 1 1 --
BC 34 31 +3
UNC 35 28 -7
Fla St 46 52 +6
Va Tech 62 69 +7
Miami 54 49 -5
Maryland 57 45 -12
Virginia 70 72 +2
Clemson 80 70 -10
NC State 82 68 -14
Ga Tech 91 152 +61
Wake 239 229 -10

Some of the moves in the Sag rankings are striking and don't seem justified by the limited games we saw last week. If anything, rankings should be somewhat steady at this time of year. We have a decent set of games to build a database and not many games are played over the holidays. And yet Sag really downgraded its opinion of Maryland... in a week in which Maryland did not even play!! Strange.
Pomeroy


Pomeroy Today Last wk Change
Duke 1 1 --
Maryland 23 25 +2
UNC 25 28 +3
Fla St 31 35 +4
Va Tech 41 50 +9
BC 42 37 -5
Clemson 59 47 -12
Miami 66 57 -9
NC State 73 71 -2
Ga Tech 77 88 +11
Virginia 85 85 --
Wake 176 155 -21

The top of the ACC all improved, though some by only small bits, while the lower-middle all slipped a bit. There may be some stratification taking place according to Pom's rankings-- Duke at the top; then a 2nd tier of UNC, Maryland, and FSU; Va Tech and BC just a shade behind that 2nd tier; and then a bunching at the bottom with Wake nowhere near anyone else in the league.

--Jason "the fact that Maryland didn't play a game and barely moved in Pom but jumped around a bit in Sag if further evidence to me that Pom is a better ranking" Evans

HCheek37
12-21-2010, 12:20 PM
Interesting but not surprising to see our Adjusted Defensive Efficiency dropped from first in the country to 4th after last night. Too many points allowed to a weak team and not enough stops will do that. Our offense stayed 2nd behind Pittsburgh.

Our chance for an unbeaten record dropped from the 19% range down to 15.93% as well. The only 3 games he has us winning less than 80% of the time at @ FSU, @ MD, @ UNC

NSDukeFan
12-21-2010, 03:10 PM
[SIZE="5"]...

The top of the ACC all improved, though some by only small bits, while the lower-middle all slipped a bit. There may be some stratification taking place according to Pom's rankings-- Duke at the top; then a 2nd tier of UNC, Maryland, and FSU; Va Tech and BC just a shade behind that 2nd tier; and then a bunching at the bottom with Wake nowhere near anyone else in the league.

--Jason "the fact that Maryland didn't play a game and barely moved in Pom but jumped around a bit in Sag if further evidence to me that Pom is a better ranking" Evans

You may be right that Pom is the better ranking and I follow it more than Sagarin. But, I think a reason Maryland moved around more in Sag's ranking than KenPom may be that Pom still has some pre-season effect in his rankings while Sagarin may be going purely on the games that have been played, which is a small sample size. So, if a team Maryland played had a bad loss, that could have an effect on Maryland's ranking. That is my theory anyway.

loran16
12-21-2010, 03:49 PM
You may be right that Pom is the better ranking and I follow it more than Sagarin. But, I think a reason Maryland moved around more in Sag's ranking than KenPom may be that Pom still has some pre-season effect in his rankings while Sagarin may be going purely on the games that have been played, which is a small sample size. So, if a team Maryland played had a bad loss, that could have an effect on Maryland's ranking. That is my theory anyway.

Well Pomeroy is roughly 80% actual season statistics right now (http://twitter.com/kenpomeroy/status/16190376503156736).

There are other difference as well, as how they handle points. I trust pomeroy given that I understand how he calculates efficiencies and pyths.

Delaware
12-22-2010, 10:56 AM
http://kenpom.com/blog/index.php/weblog/the_kpoy_a_history/

Interesting stuff here from KenPom.

Nolan is 8th so far this year in his POY ratings.

The Landlord finished first and JJ second in 2006.

Jon and Kyle were 5 and 6 in last year's ratings.

91_92_01_10_15
12-28-2010, 01:17 PM
FYI, Ohio State has jumped to #1 overall (Pomeroy) after winning 100-40 last night:

http://kenpom.com/

nocilla
12-28-2010, 03:06 PM
FYI, Ohio State has jumped to #1 overall (Pomeroy) after winning 100-40 last night:

http://kenpom.com/

So, do we need to win big against UNCG to reclaim the #1 spot? Or is UNCG so weak that it won't help?

91_92_01_10_15
12-28-2010, 03:26 PM
So, do we need to win big against UNCG to reclaim the #1 spot? Or is UNCG so weak that it won't help?

Well, the OSU win came against #316 Tennessee Martin, so I think if we outperformed the predicted 99-61 score significantly, it would probably help. The Elon game definitely did not help.

Indoor66
12-28-2010, 03:45 PM
Is there any real meaning to the dork polls at this point in the season? Teams are just now beginning to play meaningful games against meaningful opponents - conference games. IMO it will only be after the next two or so weeks that these stats based rankings will have any results that are an accurate gauge of the relative strengths of the teams.

I think that K uses these early games to learn about his team and combinations - not to run up scores to enhance his dork poll position. Same for other coaches and programs - except maybe the heels. :(

JasonEvans
12-28-2010, 03:52 PM
First of all... whew, I am tired of doing the weekly comparisons of the Sagarin and Pomeroy polls. Do folks really want me to continue? I grabbed the numbers on Monday so I can do them again this week if there is a desire. Are they helping folks very much? I rarely get responses to the posts either on the board or via PM/pitchfork comment so I don't know how much desire there is to see them continue.

Second, I was looking at the Pomeroy numbers this week and was struck at how much Duke, Ohio St, and Kansas have separated themselves from the rest of the field. All three of these teams are in the top 6 in BOTH offense and defense.



Team Offense Defense
Duke 2 4
Oh St 5 1
Kans 6 2

Just so we understand how impressive this is, no other team has an offense and defense that are both ranked in the top 12, let alone both being in the top 6.

The degree to which these three have separated themselves from the pack is also evident in their "unbeaten percentages." Now, obviously these percentages are also a function of the quality of the remaining opponents, but these three really stand out despite playing in BCS conferences.


Unbeaten chance
Duke - 17.15%
Ohio St - 11.85%
Kansas - 8.12%
Central Fla - 0.65%
San Diego St - 0.43%
Syracuse - 0.06%
Cincy - less than 0.01%

Pretty stark, huh?

Finally, one more silly thing I did. I looked at all those unbeaten percentages and thought about how long it had been since we had seen a team from a major conference go unbeaten in the regular season. I reversed the percentages to find out the odds that there are no BCS teams that go unbeaten.

So the chance of Duke losing is 82.85%. If you multiply that by the chance of Ohio St losing (88.15%), times the chance of Kansas losing (91.88%), times the chance of Syracuse losing (99.94%), times the chance of Cincy losing (essentially 100%), and you get the chance that all of them lose. Reverse that and you have the chance that one of them ends the regular season with an unbeaten record.

Ready for the numbers?

There is a 67.06% chance that all of the currently unbeaten BCS teams pick up a loss. That means there is a 32.94% chance that one of them emerges from the season unbeaten.

1 in 3... those are the odds.

--Jason "I can't decide if that is a longshot or not" Evans

SCMatt33
12-28-2010, 04:13 PM
Is there any real meaning to the dork polls at this point in the season? Teams are just now beginning to play meaningful games against meaningful opponents - conference games. IMO it will only be after the next two or so weeks that these stats based rankings will have any results that are an accurate gauge of the relative strengths of the teams.

I think that K uses these early games to learn about his team and combinations - not to run up scores to enhance his dork poll position. Same for other coaches and programs - except maybe the heels. :(

I think that the dork polls are the only ones with any meaning in this point in the season as they fully account for the strength of your opponents, unlike the human polls which reward teams for barely beating mediocre competition and punish them for barely losing to better teams.

As a side note, Dre is now #1 in the country in offensive rating, effective fg%, and true shooting percentage, 24th in TO rate, and 291st in fouls committed/40 min. He is not ranked in any other category. The lesson, don't let Dre catch and shoot. Right now, he is the perfect role player. He can't put the team on his back right now (with only being used on 14.4% of possessions, not getting to the line much, or having many assists), but you better stay honest on everyone else or he will burn you!!!

Indoor66
12-28-2010, 05:31 PM
I think that the dork polls are the only ones with any meaning in this point in the season as they fully account for the strength of your opponents, unlike the human polls which reward teams for barely beating mediocre competition and punish them for barely losing to better teams.

As a side note, Dre is now #1 in the country in offensive rating, effective fg%, and true shooting percentage, 24th in TO rate, and 291st in fouls committed/40 min. He is not ranked in any other category. The lesson, don't let Dre catch and shoot. Right now, he is the perfect role player. He can't put the team on his back right now (with only being used on 14.4% of possessions, not getting to the line much, or having many assists), but you better stay honest on everyone else or he will burn you!!!

My point is that the numbers reflect having played, for the most parts, much lesser teams. Until teams play other teams of some quality and games which have real meaning - like conference games - the stats have little import. Of even less meaning are the voting polls which are popularity contests.

loran16
12-28-2010, 06:01 PM
My point is that the numbers reflect having played, for the most parts, much lesser teams. Until teams play other teams of some quality and games which have real meaning - like conference games - the stats have little import. Of even less meaning are the voting polls which are popularity contests.

This would be true only if the numbers were raw numbers. But as Matt was trying to tell you, the numbers are adjusted for strength of schedule. For example, if Duke beats UNCG by less than 38, then Duke's numbers will drop. If Duke beats them by more, it will rise.

Certain teams are underranked or overranked in pomeroy's #s because while they've won big games, they underperform in the games vs weak competition. That has predictive value of a team being worse than they might appear by record.

In other words, Pomeroy rankings are more relevant than the polls, even at this stage, because a decent # of games have been played and strength of schedule is weighted PROPERLY in Pomeroy, but not by the polls.

Newton_14
12-28-2010, 06:16 PM
This would be true only if the numbers were raw numbers. But as Matt was trying to tell you, the numbers are adjusted for strength of schedule. For example, if Duke beats UNCG by less than 38, then Duke's numbers will drop. If Duke beats them by more, it will rise.

Certain teams are underranked or overranked in pomeroy's #s because while they've won big games, they underperform in the games vs weak competition. That has predictive value of a team being worse than they might appear by record.

In other words, Pomeroy rankings are more relevant than the polls, even at this stage, because a decent # of games have been played and strength of schedule is weighted PROPERLY in Pomeroy, but not by the polls.

The problem with that, which Indoor was pointing out is that K approaches the cupcake games differently. He uses rotations far different from what we see in games like the K State and Mich St games. The end result is while we end up beating those teams anyway, the game plays out entirely different than it would have had K used the normal rotations. Once we get into conference play and settle into what will be the normal rotation and style of play, the numbers will be a better reflection of what the team is capable of.

ice-9
12-28-2010, 08:32 PM
The problem with that, which Indoor was pointing out is that K approaches the cupcake games differently. He uses rotations far different from what we see in games like the K State and Mich St games. The end result is while we end up beating those teams anyway, the game plays out entirely different than it would have had K used the normal rotations. Once we get into conference play and settle into what will be the normal rotation and style of play, the numbers will be a better reflection of what the team is capable of.

I would add to that the following question: does winning by 30 that much different to winning by 40? Both are major blowouts in which the opponent has no chance of winning, and the lower margin is probably more reflective of a coach that put in the bench players early, experimented with combinations and strategies they otherwise wouldn't, etc.

But in Kenpom's system, winning by 40 is definitely better than 30 (and linearly so?).

To me Kenpom's ratings overvalue big blowout wins over cupcake teams -- this is why Kansas is rated so highly.

NSDukeFan
12-28-2010, 08:39 PM
I would add to that the following question: does winning by 30 that much different to winning by 40? Both are major blowouts in which the opponent has no chance of winning, and the lower margin is probably more reflective of a coach that put in the bench players early, experimented with combinations and strategies they otherwise wouldn't, etc.

But in Kenpom's system, winning by 40 is definitely better than 30 (and linearly so?).

To me Kenpom's ratings overvalue big blowout wins over cupcake teams -- this is why Kansas is rated so highly.

I think this is an interesting discussion, as I believe somewhere along the line (I am not going to go through this board, or Pomeroy's blog right now) I believe Pomeroy mentioned that how teams fare against weak opposition can be very telling about how good the team really is. So if Duke should blow teams out and consistently does, vs. a team that just barely beats weaker teams, Duke will likely have greater success in the future. However, Indoor makes some good points that there will be more value to the polls once there are more conference games where teams play quality teams more consistently (except for the Wakes and Auburns of the world?) I believe that Pomeroy is still not 100% using data from this year yet, and it makes sense that will occur in the next couple of weeks after conference play is in full swing.

JasonEvans
12-28-2010, 09:38 PM
I think this is an interesting discussion, as I believe somewhere along the line (I am not going to go through this board, or Pomeroy's blog right now) I believe Pomeroy mentioned that how teams fare against weak opposition can be very telling about how good the team really is. So if Duke should blow teams out and consistently does, vs. a team that just barely beats weaker teams, Duke will likely have greater success in the future.

Yup, Pom has mentioned that fact before. Beating a bad team by a lot says you are much better than them while wining a relatively close game against a poor team can be a sign of weakness from a team.

However, I have always wondered about the value of a 40 point win versus a 20 point win or a 50 point win versus a 30 point win. Is there really all that much of a difference in those games? It is not just the team that keeps on pressing or leaves their starters in to run up the score, it can be a team that gets great production out of some scrubs who stretch out a lead but who will rarely, if ever, play in meaningful games. How much does that matter in how good a team is?

I think that once you get to about a 25 points victory, the rest is pretty meaningless in the grand scheme of things.

-Jason "JMO" Evans

loran16
12-28-2010, 10:32 PM
Yup, Pom has mentioned that fact before. Beating a bad team by a lot says you are much better than them while wining a relatively close game against a poor team can be a sign of weakness from a team.

However, I have always wondered about the value of a 40 point win versus a 20 point win or a 50 point win versus a 30 point win. Is there really all that much of a difference in those games? It is not just the team that keeps on pressing or leaves their starters in to run up the score, it can be a team that gets great production out of some scrubs who stretch out a lead but who will rarely, if ever, play in meaningful games. How much does that matter in how good a team is?

I think that once you get to about a 25 points victory, the rest is pretty meaningless in the grand scheme of things.

-Jason "JMO" Evans

Well the idea is efficiency. A good team like Duke should be extra efficient on both ends against a bad team, resulting in higher blow-outs. That said, pomeroy has said in the past that he was figuring out how to lessen the impact of blow-outs, but i don't think he ever implemented a fix.

SCMatt33
12-28-2010, 10:36 PM
Well the idea is efficiency. A good team like Duke should be extra efficient on both ends against a bad team, resulting in higher blow-outs. That said, pomeroy has said in the past that he was figuring out how to lessen the impact of blow-outs, but i don't think he ever implemented a fix.

I know that Pomeroy and others have long lamented this problem, not just in college basketball, but in all sports. There has not yet been a reliable way to implement diminishing returns on larger blow-outs. This is one of the reasons (along with not promoting running up scores) that margin of victory has been cut of all BCS computers in football.

ice-9
12-29-2010, 01:56 AM
I haven't gone through KenPom's methodology, but I'm guessing that numbers are processed linearly. I.e. that a one point difference between a 2-point win and a 3-point win is treated the same as a one point difference between a 30-point win and a 31-point win in terms of efficiency.

I'd argue that a non-linear curve is more appropriate; i.e. that the former counts a lot more than the latter.

Or maybe take Jason's approach; beyond a certain margin (e.g. 30 points), it doesn't count as much. So linear until 30, non-linear past it or uniformly discounted.

SCMatt33
12-29-2010, 02:21 AM
I haven't gone through KenPom's methodology, but I'm guessing that numbers are processed linearly. I.e. that a one point difference between a 2-point win and a 3-point win is treated the same as a one point difference between a 30-point win and a 31-point win in terms of efficiency.

I'd argue that a non-linear curve is more appropriate; i.e. that the former counts a lot more than the latter.

Or maybe take Jason's approach; beyond a certain margin (e.g. 30 points), it doesn't count as much. So linear until 30, non-linear past it or uniformly discounted.

KenPom uses a pythagorean calculation method explained here (http://kenpom.com/blog/index.php/weblog/ratings_explanation/). It is certainly not linear. He also doesn't simply input scores into the formula but adjusted offensive and defensive efficiencies, which account for both tempo and strength of opponents. He admits that MOV is not capped, but not because it is his preference, but because there is not an adequate mathematical method of limiting it.

Skitzle
12-29-2010, 05:55 AM
The problem with that, which Indoor was pointing out is that K approaches the cupcake games differently. He uses rotations far different from what we see in games like the K State and Mich St games. The end result is while we end up beating those teams anyway, the game plays out entirely different than it would have had K used the normal rotations. Once we get into conference play and settle into what will be the normal rotation and style of play, the numbers will be a better reflection of what the team is capable of.


I would think there is also a problem in the fact that our team hasn't played a game against a significant opponent since we lost a MAJOR contributor to our team.

The "dork" polls don't take that into account...

camion
01-01-2011, 11:18 AM
As we start into the conference season I thought it would be nice to document the dork prediction (kenpom) for the final conference standings. The list reflects the unbalanced schedule and the three games already played. We'll see how things work out. Given how they've played so far I'll be surprised if Wake makes it to 2 wins.


1 Duke 15-1
2 FSU 10-6
2 UNC 10-6
2 UMd 10-6
5 BC 9-7
6 Clemson 8-8
6 Va Tech 8-8
8 Miami 7-9
9 NCSU 6-10
10 Ga Tech 6-10
11 UVa 5-11
12 WFU 2-14

91_92_01_10_15
01-06-2011, 10:25 AM
Duke inches in front by the narrowest of margins:

http://kenpom.com/

davekay1971
01-06-2011, 10:35 AM
Duke inches in front by the narrowest of margins:

http://kenpom.com/

Interesting to see Maryland so high in KenPom, 2nd highest in the ACC. They've been playing well, so we'll see if they deserve that lofty dork poll ranking come Sunday...

JasonEvans
01-06-2011, 11:20 AM
Duke inches in front by the narrowest of margins:

http://kenpom.com/

We were .001 behind Ohio St at the start of yesterday. We are now .001 ahead. I'd argue that the difference is hardly worth mentioning. What remains clear is that these two (along with Kansas, who is .0025 back) have separated themselves from the rest of college basketball. #4 Washington is .0112 behind Kansas, a fairly large margin.

It is also worth noting that as UNC and Maryland have risen in the rankings (now both in the Pom top 20), Duke's odds of going unbeaten have dropped even as Duke has continued to win games. Pom gives the Blue Devils a 13.3% chance of going unblemished on the regular season. The number was up around 18% just a few days ago but has dropped as a result of our future opponents looking stronger (especially MD, UNC, and StJohn).

--Jason "uh-oh... am I talking about going unbeaten again? Someone please stop me!!" Evans

ACCBBallFan
01-06-2011, 11:32 AM
I voted on behalf of Ken Pomeroy who currently sees ACC finish as:

14-2 Duke

11-5 UNC

10-6 MD

9-7 VA Tech
9-7 FSU who play each other this weekend

8-8 BC
8-8 Clemson

7-9 Miami

6-10 NC St who may fare better with Tracy Smith back
6-10 GA Tech

5-11 UVA may be worse if Mike Scott redshirts due to ankle injury

2-14 Wake

The way the Pomeroy system works, those two Duke losses are not any particular game, favord in all, but the sum of a bunch of partial percentages chances of losing, 32% @ MD, 28% at UNC for example sum to 60% and account for one of the virtual losses along with some other partials.

Sun Jan 9 15 Maryland W, 82-69 78 89% Home
Wed Jan 12 43 Florida St. W, 77-66 75 84% Away
Sat Jan 15 102 Virginia W, 82-57 65 99% Home
Wed Jan 19 77 North Carolina St. W, 84-68 73 93% Away
Sat Jan 22 174 Wake Forest W, 95-68 77 98% Away
Thu Jan 27 59 Boston College W, 90-68 69 96% Home
Sun Jan 30 40 St. John's W, 78-68 70 83% Away
Wed Feb 2 15 Maryland W, 78-73 78 68% Away
Sat Feb 5 77 North Carolina St. W, 89-64 73 98% Home
Wed Feb 9 18 North Carolina W, 87-72 79 90% Home
Sun Feb 13 56 Miami FL W, 82-69 72 88% Away
Wed Feb 16 102 Virginia W, 77-60 65 95% Away
Sun Feb 20 88 Georgia Tech W, 88-62 75 98% Home
Wed Feb 23 29 Temple W, 78-62 71 93% Home
Sat Feb 26 33 Virginia Tech W, 75-66 71 79% Away
Wed Mar 2 52 Clemson W, 80-61 71 96% Home
Sat Mar 5 18 North Carolina W, 82-76 79 72% Away
Projected record: 29-2 14-2

ACCBBallFan
01-06-2011, 11:46 AM
On Pomeroy scoring the second place battle is very close between UNC and MD. Though it rounds to a full 1 loss difference, the net difference is only 0.21.

Duke's loss percentages sum to 1.74.

UNC''s sum to 5.39

MD's future losses sum to 4.6 plus their first game loss to fading BC makes it 5.6.

VA Tech sums to 5.79 plus its first game loss to UVA makes it 6.79.

FSU's percentage chances of losses sum to 6.8

superdave
01-06-2011, 11:47 AM
It is also worth noting that as UNC and Maryland have risen in the rankings (now both in the Pom top 20), Duke's odds of going unbeaten have dropped even as Duke has continued to win games. Pom gives the Blue Devils a 13.3% chance of going unblemished on the regular season. The number was up around 18% just a few days ago but has dropped as a result of our future opponents looking stronger (especially MD, UNC, and StJohn).


Anyone watched Unc or Maryland lately? Any insight on how they are playing better?

Jordan Williams and Tyler Zeller will be tough matchups for us. But we have depth at the 4/5 so I have some level of confidence.

ice-9
01-06-2011, 11:49 AM
I voted on behalf of Ken Pomeroy who currently sees ACC finish as:

14-2 Duke

11-5 UNC

10-6 MD



Yet he sees Maryland as the second highest rated ACC team! The closest game he predicts we will have in fact is a 5-point win over MD at MD with a 32% chance of losing. That's 1 in 3.

JasonEvans
01-06-2011, 12:03 PM
Anyone watched Unc or Maryland lately? Any insight on how they are playing better?

Jordan Williams and Tyler Zeller will be tough matchups for us. But we have depth at the 4/5 so I have some level of confidence.

One of the reasons I was so pleased to see Mason step up and have a good game last night was the knowledge that Jordan Williams comes to Cameron this weekend. He and Zeller are probably 2 of the 5 or 10 best big men in the country right now.

-Jason "Williams is a horse!!" Evans

JasonEvans
01-06-2011, 09:30 PM
KenPom recently did a projection (http://kenpom.com/blog/index.php/weblog/pre-conference_preview_blowout_part_1/) of how each conference would turn out running 10000 simulations based on current team ratings. Here is an interesting commentary (http://rivals.yahoo.com/ncaa/basketball/blog/the_dagger/post/Five-big-surprises-from-Ken-Pomeroy-s-conference?urn=ncaab-303792#remaining-content) on Pomeroy's projections.

-Jason "he has the ACC as the biggest blowout of any conference, with Duke winning the title 93.8% of the time" Evans

zack2014
01-06-2011, 10:01 PM
Anyone watched Unc or Maryland lately? Any insight on how they are playing better?


I watch a good bit of Maryland. I think a lot of them looking better has to do the quality of competition. Since the BC loss, they have played teams ranked 243, 332, and 334 according to KenPom. Maryland has trouble whenever they face pressure defense and none of these teams have been able to pressure Maryland's guards very well. We will see what happens when Maryland's guards face a good backcourt.

hurleyfor3
01-10-2011, 11:35 AM
After the Maryland win Pomeroy now projects our conference record at 15-1.

loran16
01-10-2011, 12:07 PM
After the Maryland win Pomeroy now projects our conference record at 15-1.

Despite that, our likelihood of winning at Maryland is down to a season low 67%, and we're expected to pull it out by 4.

mr. synellinden
01-10-2011, 12:13 PM
Despite that, our likelihood of winning at Maryland is down to a season low 67%, and we're expected to pull it out by 4.

And Maryland is up to #13 in the rankings with the 3rd ranked defensive efficiency in the country - so we have at least one top 20 win on our resume.

Michigan St. is down to 25, Kansas St. 39, Marquette 40 and Butler 49! Suddenly, our out of conference schedule doesn't look so great.

UNC is at 21, VT at 26 are the only other ACC teams in the top 40.

Also interesting is that UConn is quickly down to 29.

SCMatt33
01-10-2011, 12:27 PM
And Maryland is up to #13 in the rankings with the 3rd ranked defensive efficiency in the country - so we have at least one top 20 win on our resume.

Maryland is a really interesting case in the computers. They have a ton of wins against mediocre and bad competition. Their best wins are @Penn St. and home vs. CoC. They have 6 wins outside the RPI top 200 and 7 outside of Pomeroy's top 200. They are 0-5 vs. the RPI top 50 (0-4 for Pomeroy as BC is #55), but all of those losses were by single digits, with 3 games lost by 4 points or less. Outside of CoC, they have won all of their other games by a wide margin, which has led to them being loved by margin based computers, but hated by win based computers. Sagarin has Maryland at 45 overall, but with giant splits between the win based ELO Chess (80) and the margin based Pure Points Predictor (21). Their RPI is between 110-120 depending on who you ask. If they don't get a win over Villanova or Duke in the rematch, and finish with 10 or so ACC wins, they will be an interesting case for the committee.

loran16
01-10-2011, 12:34 PM
Maryland is a really interesting case in the computers. They have a ton of wins against mediocre and bad competition. Their best wins are @Penn St. and home vs. CoC. They have 6 wins outside the RPI top 200 and 7 outside of Pomeroy's top 200. They are 0-5 vs. the RPI top 50 (0-4 for Pomeroy as BC is #55), but all of those losses were by single digits, with 3 games lost by 4 points or less. Outside of CoC, they have won all of their other games by a wide margin, which has led to them being loved by margin based computers, but hated by win based computers. Sagarin has Maryland at 45 overall, but with giant splits between the win based ELO Chess (80) and the margin based Pure Points Predictor (21). Their RPI is between 110-120 depending on who you ask. If they don't get a win over Villanova or Duke in the rematch, and finish with 10 or so ACC wins, they will be an interesting case for the committee.

Agreed. Personally, I'm a believer in the margin based predictions, but well to get INTO the Tournament, you need to get good wins.

And they're starting 0-2, with one of those being a bad home loss (BC's a decent team, ,but MD should've won at home). Not an ideal situation. They should beat Wake easily, but Nova, VTech, and Clemson form a 3 game stretch where they need to at least go 1-2, if not 2-1.

MChambers
01-16-2011, 04:19 PM
I know this seems crazy, but Duke is still #1 in Pomeroy. Duke is also #1 in the Sagarin Predictor (but #4 in his blended rating).

I think this shows a few things. First, although Duke's going through a bit of a disappointing stretch, we're still darned good. Second, Kansas and Ohio State are also struggling right now, although they haven't lost yet. Third, the three teams had really separated themselves from the other teams earlier this season. They're coming back to the pack somewhat now.

loran16
01-16-2011, 05:24 PM
I know this seems crazy, but Duke is still #1 in Pomeroy. Duke is also #1 in the Sagarin Predictor (but #4 in his blended rating).

I think this shows a few things. First, although Duke's going through a bit of a disappointing stretch, we're still darned good. Second, Kansas and Ohio State are also struggling right now, although they haven't lost yet. Third, the three teams had really separated themselves from the other teams earlier this season. They're coming back to the pack somewhat now.

Duke actually dropped to #2, but OSU's close game against PSU dropped em back.

cptnflash
01-16-2011, 06:08 PM
I know this seems crazy, but Duke is still #1 in Pomeroy. Duke is also #1 in the Sagarin Predictor (but #4 in his blended rating).

I think this shows a few things. First, although Duke's going through a bit of a disappointing stretch, we're still darned good. Second, Kansas and Ohio State are also struggling right now, although they haven't lost yet. Third, the three teams had really separated themselves from the other teams earlier this season. They're coming back to the pack somewhat now.

Keep in mind that Pomeroy does not adjust for changes in personnel. So the games that Kyrie played count towards our ranking as if he were still on the team.

loran16
01-16-2011, 08:13 PM
Keep in mind that Pomeroy does not adjust for changes in personnel. So the games that Kyrie played count towards our ranking as if he were still on the team.

Correct, But Duke's lead over OSU/KU has basically been the same since Kyrie went down. Which means the drop hasn't made us not one of the top 3 really.

loran16
01-20-2011, 01:49 PM
Duke fell to #3 today for the first time all year. We gave up more points than expected to NC State and OSU had a huge game vs Iowa.

Of course, the Top 3 (Duke-KU-OSU) are still really close (Seperated by .0013 of pythag) and still have a decent lead on #4 Pitt.

superdave
01-22-2011, 12:44 PM
Anyone else watching Cuse vs. Nova? KenPom's got them at #9 Cuse vs. #11 Nova, however the coaches' poll has them inflated to #3 Cuse vs. #7 Nova.

Nova is presently jump-shot-ing Cuse to death at 40-28 with a few minutes left in the half. They have been pressing full court and knocked down 8-9 threes. Nova can get you off your game, but dont seem to have the big men to be really good. I think Cuse is hanging around still and starting to be able to get to the rim a little better. I'm not sure either team is elite - no players really scare me.

The Big East is currently the #2 conference behind the Big 11 ( I mean 10).

hurleyfor3
01-23-2011, 02:15 PM
Kansas' loss just barely caused us to switch places with them in Pomeroy. We're now #2 with KU #3.

HCheek37
01-25-2011, 08:55 AM
any update coming here or are we dumping Sagarin and KenPom?

94duke
01-25-2011, 09:14 AM
any update coming here or are we dumping Sagarin and KenPom?

Kenpom's last update was Sunday.

Delaware
02-02-2011, 09:53 AM
Duke is now 4 in KenPom.

And for the first time all year predicted to lose the away game at the Hill.

Not that anyone believes in this black magic.

:cool:

COYS
02-02-2011, 11:33 AM
Duke is now 4 in KenPom.

And for the first time all year predicted to lose the away game at the Hill.

Not that anyone believes in this black magic.

:cool:

Yeah, the St. John's game trashed our defensive rating. Our offensive rating has suffered majorly due to the relatively underwhelming performances of some of our former opponents (Mich. St, K-State, FSU, previously our strongest opponents, have all been dropping). A big game in College Park can turn things around pretty quickly, though. Anybody recall what kind of a hit we took in Pomeroy's ratings last year after the Georgetown and NCState games?

JasonEvans
02-03-2011, 10:15 AM
The Maryland game gave Duke a decent little boost in the KenPom numbers.

Prior to the game, Duke was 4th in the rankings with a score of .9675. We were a somewhat distant 4th, more than .0100 behind 3rd place. The top three were Ohio St (.9809), Texas (.9773), and Kansas (.9767).

Duke was 9th in offense with a 118.0 and 10th in defense with a 87.9. Our defensive numbers had been crushed by St. John's whupping us. Pom predicted Duke would win at Maryland only 51% of the time and he predicted a final score of 75-74.

Following Duke's 80-62 victory, Duke remains 4th in the Pom standings, but has moved a lot closer to the top 3. Our overall score is now .9721, a significant increase of .0046. Our offense has moved up to #4 with a rating of 120.0. Our defense moved down a bit (we committed too many fouls and gave up too many offensive rebounds) to 12th at 88.1 (lower is better in this defensive metric).

Anyway, Pom now projects we will go 14-2 in the ACC. He sees the ACC standings coming out like this--


Duke 14-2
UNC 12-4
Maryland 10-6
Va Tech 10-6
FSU 10-6
Clemson 9-7 (they really blew it losing to UVA last night)
Ga Tech 7-9
BC 7-9
Miami 6-10
NC St 5-11 (buh-bye, Sid!)
Virginia 5-11
Wake 2-14

-Jason "I love being a Dork!" Evans

MChambers
02-03-2011, 10:38 AM
The Maryland game gave Duke a decent little boost in the KenPom numbers.

Prior to the game, Duke was 4th in the rankings with a score of .9675. We were a somewhat distant 4th, more than .0100 behind 3rd place. The top three were Ohio St (.9809), Texas (.9773), and Kansas (.9767).

Duke was 9th in offense with a 118.0 and 10th in defense with a 87.9. Our defensive numbers had been crushed by St. John's whupping us. Pom predicted Duke would win at Maryland only 51% of the time and he predicted a final score of 75-74.

Following Duke's 80-62 victory, Duke remains 4th in the Pom standings, but has moved a lot closer to the top 3. Our overall score is now .9721, a significant increase of .0046. Our offense has moved up to #4 with a rating of 120.0. Our defense moved down a bit (we committed too many fouls and gave up too many offensive rebounds) to 12th at 88.1 (lower is better in this defensive metric).

Anyway, Pom now projects we will go 14-2 in the ACC. He sees the ACC standings coming out like this--


Duke 14-2
UNC 12-4
Maryland 10-6
Va Tech 10-6
FSU 10-6
Clemson 9-7 (they really blew it losing to UVA last night)
Ga Tech 7-9
BC 7-9
Miami 6-10
NC St 5-11 (buh-bye, Sid!)
Virginia 5-11
Wake 2-14

-Jason "I love being a Dork!" Evans

Thanks for the update, Jason. Really wish our defense was moving in the other direction. 12th is subpar for Duke. Nice to be able say that having the 12th best defense in the country is disappointing, I guess, but I want better.

camion
02-03-2011, 10:42 AM
Kenpom now predicts that Duke will lose 1 more conference game and 2 more games overall, ending the regular season 14-2 and 27-4.

But wait. Duke has only one non-conference game remaining, Temple at home, and we're shown with a 90% chance of winning that one. Where is that second loss going to come from?

Note: I know the math can work out like that, but it is an interesting quirk in the predictions.

gw67
02-03-2011, 10:47 AM
Jason,

Thanks for posting Pomeroy's projections. The projections for Duke, UNC, Florida State and Virginia Tech as well as the bottom feeders appear right to me. I haven't gone through the remaining schedules in detail but I think a 10-6 record for Maryland is a reach. They don't have to play Duke twice during the second half of the season but they play UNC and FSU instead.

gw67

hurleyfor3
02-03-2011, 10:47 AM
Anyway, Pom now projects we will go 14-2 in the ACC. He sees the ACC standings coming out like this--


Duke 14-2
UNC 12-4
Maryland 10-6
Va Tech 10-6
FSU 10-6
Clemson 9-7 (they really blew it losing to UVA last night)
Ga Tech 7-9
BC 7-9
Miami 6-10
NC St 5-11 (buh-bye, Sid!)
Virginia 5-11
Wake 2-14

Our projected ACC record has been pretty stable, but unc's has been gradually rising. Weren't they predicted at under .500 at one point?

Everyone thinks we're a terrible conference this year, but I believe the middle is decent, and have to believe anyone at 10-6 is going to get into the NCAAs. The extra four bids this year may even get a 9-7 team in. So that's at least five bids for the ACC. We've done worse.

pfrduke
02-03-2011, 11:05 AM
The Maryland game gave Duke a decent little boost in the KenPom numbers.

It also knocked Maryland down a peg, taking them from 13 all the way to 21.

DevilBen02
02-03-2011, 11:09 AM
The Maryland game gave Duke a decent little boost in the KenPom numbers.

It is also fun to note that Duke's offensive performance last night dropped Maryland from #2 in defensive efficiency (at least that's what I thought I saw them at before the update) down to #4. In fact, Duke's performance was the most efficient offensive game against Maryland this season (if I'm reading the "Game Plan" correctly). Let's hope that this is a sign that of things to come against solid D's.

gw67
02-04-2011, 10:35 AM
Anyway, Pom now projects we will go 14-2 in the ACC. He sees the ACC standings coming out like this--


Duke 14-2
UNC 12-4
Maryland 10-6
Va Tech 10-6
FSU 10-6
Clemson 9-7 (they really blew it losing to UVA last night)
Ga Tech 7-9
BC 7-9
Miami 6-10
NC St 5-11 (buh-bye, Sid!)
Virginia 5-11
Wake 2-14

-Jason "I love being a Dork!" Evans

I looked at the remaining schedules for the top 7 teams. The toughest schedule by far belongs to the Heels. They play 7 of their remaining 9 games against the top 7 teams including 4 games against Duke and Florida State. The Noles have the easiest remaining schedule. Besides playing the Heels home and away, they play at Maryland. Based on my chrystal ball (very cloudy as usual), I predict the final standings for the top 7 teams:

Duke 14-2
UNC 13-3
FSU 12-4
VT 11-5
Md 10-6
Clem 8-8
BC 7-9

When I initially did this, I had Md at 9-7 and BC at 8-8 but being a fan of the Terps I figure that they are owed one for the initial loss to BC.

If the above plays out, four teams are locks for NCAAT and the Terps could probably play their way in with a win in the ACCT.

gw67

hurleyfor3
02-04-2011, 11:17 AM
Duke 14-2
UNC 13-3
FSU 12-4
VT 11-5
Md 10-6
Clem 8-8
BC 7-9

When I initially did this, I had Md at 9-7 and BC at 8-8 but being a fan of the Terps I figure that they are owed one for the initial loss to BC.

If the above plays out, four teams are locks for NCAAT and the Terps could probably play their way in with a win in the ACCT.


I assume that's pre-Drew transferring. :) Can anyone solidify their bid hopes by picking an extra win off the Heels?

gw67
02-04-2011, 11:23 AM
I've watched the Heels play a few times and, honestly, they did not play well except for their last game. I figured that they would split with Duke and Florida State and defeat all the rest. I don't see where Drew leaving affects this prediction unless this is a sign that there is some discontent on that team.

gw67

ns7
02-04-2011, 05:57 PM
Thanks for the update, Jason. Really wish our defense was moving in the other direction. 12th is subpar for Duke. Nice to be able say that having the 12th best defense in the country is disappointing, I guess, but I want better.

I'm really worried about the defense--starting with the FSU game in January. It's been a combination of the other team either shooting well or getting a lot of free throws, and the last two games are particularly worrying on the FT aspect. I wonder if a potential Irving comeback would help here, maybe on defensive breakdowns due to dribble penetration but a lot of it is due to fouls on opposing post players.

The other thing on defense that's troubling is the 3pt defense. This is typically a Duke strength, e.g., last year's team was #2 at defending three point shots.

loran16
02-04-2011, 06:47 PM
I'm really worried about the defense--starting with the FSU game in January. It's been a combination of the other team either shooting well or getting a lot of free throws, and the last two games are particularly worrying on the FT aspect. I wonder if a potential Irving comeback would help here, maybe on defensive breakdowns due to dribble penetration but a lot of it is due to fouls on opposing post players.

The other thing on defense that's troubling is the 3pt defense. This is typically a Duke strength, e.g., last year's team was #2 at defending three point shots.

I wouldn't be too worried about the D right now....Duke is #2 in the ACC in defense in conference play...it's just that Pomeroy expects more out of us due to how the offenses we've faced have been rated (FSU in particular hurt a bit).

ns7
02-05-2011, 01:50 PM
I wouldn't be too worried about the D right now....Duke is #2 in the ACC in defense in conference play...it's just that Pomeroy expects more out of us due to how the offenses we've faced have been rated (FSU in particular hurt a bit).

Well, let's see how the NC State and UNC games go. If we hold both under 1 PPP then I'll feel better.

superdave
02-05-2011, 05:35 PM
I looked at the remaining schedules for the top 7 teams. The toughest schedule by far belongs to the Heels. They play 7 of their remaining 9 games against the top 7 teams including 4 games against Duke and Florida State. The Noles have the easiest remaining schedule. Besides playing the Heels home and away, they play at Maryland. Based on my chrystal ball (very cloudy as usual), I predict the final standings for the top 7 teams:

Duke 14-2
UNC 13-3
FSU 12-4
VT 11-5
Md 10-6
Clem 8-8
BC 7-9

When I initially did this, I had Md at 9-7 and BC at 8-8 but being a fan of the Terps I figure that they are owed one for the initial loss to BC.

If the above plays out, four teams are locks for NCAAT and the Terps could probably play their way in with a win in the ACCT.

gw67

Hmmmmm....dont know about these records. It's certainly not far fetched to think Duke only drops one more league game. But ACC teams the last 5-6 years almost always drop a few frustrating games. No one seems to seize the moment (maybe MD last year?) except for the 1-2 best teams. I'd drop another loss or two on everyone except Duke. My expectations for the league are low.

I forget who it was, but a few weeks ago after losing to Harvard, someone tweeted "Hey BC - either stop losing to Harvard or stop beating ACC teams. Thanks!" Thought that was pretty accurate.

JasonEvans
02-06-2011, 09:07 AM
I noticed some disturbing symmetry while looking at Pom this morning.


The odds of FSU beating Carolina at Carolina today -- 18%
The odds of Carolina beating Duke at Duke Wednesday -- 18%


--Jason "(today) c'mon 18%... (Weds) booo, 18%" Evans

gw67
02-06-2011, 09:53 AM
Hmmmmm....dont know about these records. It's certainly not far fetched to think Duke only drops one more league game. But ACC teams the last 5-6 years almost always drop a few frustrating games. No one seems to seize the moment (maybe MD last year?) except for the 1-2 best teams. I'd drop another loss or two on everyone except Duke. My expectations for the league are low.

I forget who it was, but a few weeks ago after losing to Harvard, someone tweeted "Hey BC - either stop losing to Harvard or stop beating ACC teams. Thanks!" Thought that was pretty accurate.

Super (Isn't there also a Supadave?),

Obviously, I assumed that the top 7 teams would mostly beat up on the lowly 5 teams. Quite frankly, I see only two teams from the ACC making it out of the first weekend of the NCAAT (Duke and UNC) regardless of how many are invited. Can anyone really see FSU, VT or BC actually winning two games in the tourney? Not I.

gw67

MB in MD
02-06-2011, 10:08 AM
After whupping State our AdjD is a more respectable 8 on Kenpom, just behind Maryland's 7, though I still think it needs to rise for us to be competitive in the tourney. But scroll down a bit the the team ranked 16th overall and you''ll find Carowhina at 6!!!? Huh?

tieguy
02-06-2011, 11:42 AM
Carolina always does fairly well in Pom's defensive standings. Their very fast pace is misleading- teams score a lot of points against them, but on a per-possession basis, they are pretty good stoppers- tall guys, and usually aggressive perimeter defenders backed up by those tall, often mobile, guys.

hurleyfor3
02-10-2011, 10:03 AM
Once we get the first unc game over with I start to get more analytical about this stuff.

Unc moves up to #10 Pomeroy by losing by "only" six. We're still #4.

Predicted ACC standings after the unc game:

Duke 14-2
Unc 12-4
Fsu/Maryland/Vpi 10-6
Clemson 9-7
Bc 8-8
Miami/GIT 6-10
UVa/Ncsu 5-11
Wake 2-14

There's not a whole lot there to disagree with, although I think one team in the 10-6/9-7 group might break out and sort of challenge unc for second place, especially if said breakout team beats unc. Unc has @clemson and @fsu in addition to home against Maryland. No game against vpi.

Our two toughest remaining games are @vpi and @unc. Ken thinks the chance we will lose at least one of these games is 65% (1-(.52*.68)).

-jk
02-15-2011, 10:05 AM
So, with K St spanking KU last night, Duke moves back into second place in Pom's (http://kenpom.com/) rating. Now if we can just win the six games we're projected to win in his system...

And don't pay attention to his projected records! I'll see what I want to see. Or, as some like to say, "perception is reality" - and I perceive Pom projecting wins in six more games, so that's my reality. It worked for the final six games last year! :D </Channeling my inner Ozzie>

-jk

ACCBBallFan
02-15-2011, 07:23 PM
I think the way Pomperoy determines league recoird is simply adding all the partial percentages for the remaining games and relying on law of large numbers.

http://kenpom.com/team.php?team=Duke

Latest:

14-2 Duke
13-3 UNC
11-5 FSU but does not factor in impact of Singleton injury
10-6 VA Tech

9-7 MD and Clemson
8-8 BC

6-10 Miami
5-11 GA T; NC St and UVA
1-15 Wake

camion
02-15-2011, 09:00 PM
Remember the Simon and Garfunkel commentary on statistics.

"All lies and jest, still a man hears what he wants to hear
And disregards the rest, hmmmm."

I remain hopeful that the prediction will prove true. :)

Bob Green
02-15-2011, 10:47 PM
We've heard all year how this is a down year for the ACC, and the league's performance in OOC games supports the theory. However, I wonder if the league is better now than it was the first couple of months of the season. Using Ken Pomeroy's numbers, the ACC has four teams in the Top 25, which ties us with the Big 10 for second place behind the Big East:

Big East 7
ACC 4
Big10 4
Big12 3
MWC 2
Pac10 2
CAA 1
SEC 1
WAC 1

Perhaps the ACC isn't as down as conventional wisdom indicates.

Wander
02-16-2011, 12:13 AM
However, I wonder if the league is better now than it was the first couple of months of the season.


You mean now that it gets to play ACC teams all the time? ;)

uh_no
02-16-2011, 01:25 AM
You mean now that it gets to play ACC teams all the time? ;)

Fun fact: the combined record of all the acc teams during the acc season is .500 always....

Fun fact #2: this is true of all conferences

:P

superdave
02-16-2011, 09:48 AM
You mean now that it gets to play ACC teams all the time? ;)

What is the ACC's best out of conference win this year? Unc over Kentucky?

Duke beat Michigan State, K State, Butler and Marquette (highest in Pom at 34).

Bob Green
02-16-2011, 10:03 AM
You mean now that it gets to play ACC teams all the time? ;)

I guess I set myself up and deserve that response. :) However, it isn't what I meant. UNC is obviously playing better now than in December, Virginia Tech has developed into a better team after surviving early turmoil with a rash of injuries, and Duke has adapted to life without Kyrie Irving and is a better team. So, IMO, the ACC is setting itself up to make some noise in the NCAAT. Chris Singleton being out at Florida State is a blow to the Seminoles so we will have to wait and see how quickly (if at all) they can adapt to a new line-up.

Perhaps my optimistic nature is getting the best of me, but I'm starting to doubt the whole "ACC is down" theory.

JRH1010
02-16-2011, 10:03 AM
What is the ACC's best out of conference win this year? Unc over Kentucky?

Duke beat Michigan State, K State, Butler and Marquette (highest in Pom at 34).

I would say that was it. The ACC is not as down as the mainstream would like you to believe. Like I say every year in March. The rest of the nation does not want to see a middle ACC team on their schedule in the first or second round.

toooskies
02-16-2011, 11:14 AM
The other thing to take into account is while the ACC has lost a lot of non-conference games, a lot of those games we lost were close. UNC lost to Texas by 2 points, and only has a single loss by more than 10. VT took Purdue to overtime. Maryland only lost by single digits to Pitt, Illinois, and Villanova; the only teams they lost by more than 10 to were VT and Duke. Duke is ranked highly because in the vast majority of games this year, we not only are winning, but winning comfortably. The UNC game is the major exception.

Also: when saying Duke doesn't have any "good wins" at this point, remember that Kansas State and Michigan State were putting a lot of their offseason focus into winning at Duke, their first big game. Their seasons could have gone drastically differently with a big win against us. KSU lost by 14 and we demolished a pre-season All-American who hasn't played like one until the Kansas game. MSU played at perhaps their best and still lost by 5.

You could definitely argue that UNC and VT are playing above their level, as their rankings have been steadily increasing all year. And there's good non-statistical reasons for that, with UNC improving once they found their best PG and VT has started overcoming their early injury issues.

Point being, there's no reason to believe that the ACC can't compete in the NCAAs, and at least Duke and UNC are poised for deep runs. Plenty of other teams (VT, Maryland, possibly FSU, possibly Clemson) could pull off a surprise win. Then again, they could also lose more close games.

TwiceDuke
02-16-2011, 11:43 AM
I have no clue if this has been noticed earlier, or in other threads, but I wonder if the perception of the ACC having a down year is aided by the fact that currently, four ACC teams (Florida State, North Carolina, Maryland, and Duke) are in the top eight in terms of defensive efficiency this year.

Looking back at the data available on Pomeroy's site, this is a change in trend (which may be subject to change by the end of the season). In 2010, only Duke and Florida State finished in the top ten on defense. Going back

2009 - Florida State was the top entry, at # 12
2008 - Three in the top 15
2007 - Three in the top 10
2006 - Duke was the top entry, at # 13
2005 - Three in the top 10
2004 - Three in the top 10
2003 - Two in the top 20


Certainly, the ACC is having a down year; I don't think anyone can paint over that fact. The out-of-conference record speaks for itself. I also believe that the computer numbers reveal the state of the ACC.

But I also think that popular perception and the media can (and do) overlook computer numbers. At times, I think the assessment that a team is a "good team" can rely solely upon it's ability to put points on the board. And I get the sense - though I cannot point to any particular pundit or article - that part of the knock on the ACC this year has been the lack of explosive offenses.

If the defenses in the ACC are having a particularly good year, perhaps ACC offenses are not as bad as they have been pronounced to be (not by Pomeroy's numbers - which I understand are adjusted, but by critics).

Or, maybe, I'm trying to defend a league that just simply doesn't cut it this year.

pfrduke
02-16-2011, 11:50 AM
I have no clue if this has been noticed earlier, or in other threads, but I wonder if the perception of the ACC having a down year is aided by the fact that currently, four ACC teams (Florida State, North Carolina, Maryland, and Duke) are in the top eight in terms of defensive efficiency this year.

Looking back at the data available on Pomeroy's site, this is a change in trend (which may be subject to change by the end of the season). In 2010, only Duke and Florida State finished in the top ten on defense. Going back

2009 - Florida State was the top entry, at # 12
2008 - Three in the top 15
2007 - Three in the top 10
2006 - Duke was the top entry, at # 13
2005 - Three in the top 10
2004 - Three in the top 10
2003 - Two in the top 20


Certainly, the ACC is having a down year; I don't think anyone can paint over that fact. The out-of-conference record speaks for itself. I also believe that the computer numbers reveal the state of the ACC.

But I also think that popular perception and the media can (and do) overlook computer numbers. At times, I think the assessment that a team is a "good team" can rely solely upon it's ability to put points on the board. And I get the sense - though I cannot point to any particular pundit or article - that part of the knock on the ACC this year has been the lack of explosive offenses.

If the defenses in the ACC are having a particularly good year, perhaps ACC offenses are not as bad as they have been pronounced to be (not by Pomeroy's numbers - which I understand are adjusted, but by critics).

Or, maybe, I'm trying to defend a league that just simply doesn't cut it this year.

It's an interesting theory, but I think much of the criticism has derived from the lack of out-of-conference success (under .500 against the BCS conferences, losses to several small conference teams that any ACC team should be beating). If FSU had beaten Florida and Ohio State, for example, it wouldn't matter that their offense is sub-par - they'd be thought of much more highly. You could say the same for Virginia Tech w/r/t Kansas State, UNLV, and Purdue; Maryland w/r/t Illinois, Pitt, Temple, and Villanova, etc., etc. If the conference collectively turned 12 non-conference losses into wins - just 1 for each team - we'd be thought of more highly.

uh_no
02-16-2011, 11:52 AM
The rest of the nation does not want to see a middle ACC team on their schedule in the first or second round.

Yes, but that's true of a middle big east team, a middle big 10 team, or a middle big 12 team as well, so it doesn't really say much about the respective strengths. And honestly, I'd rather face a clemson than a syracuse...

Duvall
02-16-2011, 12:00 PM
Yes, but that's true of a middle big east team, a middle big 10 team, or a middle big 12 team as well, so it doesn't really say much about the respective strengths.

If you're facing a team from the middle of the Big 12 in the second round, it's probably a sign that you're in the NIT. So no one wants that.

gumbomoop
02-16-2011, 12:03 PM
Duke moves up to #2 in KenPm.

http://kenpom.com/

NSDukeFan
02-16-2011, 12:40 PM
I guess I set myself up and deserve that response. :) However, it isn't what I meant. UNC is obviously playing better now than in December, Virginia Tech has developed into a better team after surviving early turmoil with a rash of injuries, and Duke has adapted to life without Kyrie Irving and is a better team. So, IMO, the ACC is setting itself up to make some noise in the NCAAT. Chris Singleton being out at Florida State is a blow to the Seminoles so we will have to wait and see how quickly (if at all) they can adapt to a new line-up.

Perhaps my optimistic nature is getting the best of me, but I'm starting to doubt the whole "ACC is down" theory.

I think a down year for the ACC is solid compared to most conferences and the league isn't as weak as some would say. On the other hand, I don't know if I would expect any ACC team to win a single game in the NCAAT besides Duke and UNC. I don't think I would pick Virginia Tech, FSU (w/o Singleton), or Clemson vs. most 5 to 9 seeds. I hope you're right to be optimistic, but it would be a major upset to have 3 ACC teams through the first weekend based on how teams have played to date. This can certainly change in the next three weeks, though.

superdave
02-16-2011, 12:48 PM
I have no clue if this has been noticed earlier, or in other threads, but I wonder if the perception of the ACC having a down year is aided by the fact that currently, four ACC teams (Florida State, North Carolina, Maryland, and Duke) are in the top eight in terms of defensive efficiency this year.

Looking back at the data available on Pomeroy's site, this is a change in trend (which may be subject to change by the end of the season). In 2010, only Duke and Florida State finished in the top ten on defense. Going back

2009 - Florida State was the top entry, at # 12
2008 - Three in the top 15
2007 - Three in the top 10
2006 - Duke was the top entry, at # 13
2005 - Three in the top 10
2004 - Three in the top 10
2003 - Two in the top 20


Certainly, the ACC is having a down year; I don't think anyone can paint over that fact. The out-of-conference record speaks for itself. I also believe that the computer numbers reveal the state of the ACC.

But I also think that popular perception and the media can (and do) overlook computer numbers. At times, I think the assessment that a team is a "good team" can rely solely upon it's ability to put points on the board. And I get the sense - though I cannot point to any particular pundit or article - that part of the knock on the ACC this year has been the lack of explosive offenses.

If the defenses in the ACC are having a particularly good year, perhaps ACC offenses are not as bad as they have been pronounced to be (not by Pomeroy's numbers - which I understand are adjusted, but by critics).

Or, maybe, I'm trying to defend a league that just simply doesn't cut it this year.

Are ACC defenses that good or are crappy ACC offenses boosting that ranking? I see a lot of weak PG play in the ACC this year and that is a major factor in how efficient offenses are.

MChambers
02-16-2011, 12:57 PM
Are ACC defenses that good or are crappy ACC offenses boosting that ranking? I see a lot of weak PG play in the ACC this year and that is a major factor in how efficient offenses are.
The Pomeroy efficiency numbers are adjusted to take into account the strengths of the offenses and defenses you play against. The ACC offenses looked quite poor even before conference play started, so that should not affect the defensive ranking. In fact, I believe that earlier in the year some of the ACC teams had very high defensive ratings, including Maryland.

uh_no
02-16-2011, 01:06 PM
If you're facing a team from the middle of the Big 12 in the second round, it's probably a sign that you're in the NIT. So no one wants that.

ZING

loran16
02-16-2011, 01:52 PM
Are ACC defenses that good or are crappy ACC offenses boosting that ranking? I see a lot of weak PG play in the ACC this year and that is a major factor in how efficient offenses are.

The former. Last year, the ACC had ridiculous Defenses in the year, which was why Duke had the #1 Offense according to Pomeroy despite some lower raw numbers....it was insanely tough to score in the ACC.

The ACC's overall D is down this year (Thanks BC and Wake!), but some remnants of the awesome Ds remain.

toooskies
02-16-2011, 02:01 PM
The KenPom defensive numbers in-conference have been strong. I attribute this more to good coaching and great talent. K, Roy, and Gary have all been coaching basketball a long time and can turn any bunch of athletes into a good defensive team, and great athletes into a great defensive team. Also, you can sacrifice offense to get better defense (i.e. crashing the boards instead of running the fast break).

Even the best defenses have weak points, and a mediocre offense won't get you very far. The 2007 team was the 5th ranked KenPom defense, but that didn't translate to into wins. It translated into a lot of close losses.

kong123
02-16-2011, 03:06 PM
i would disagree with saying a coach can make great athletes into great defenders. I don't think anyone would deny that UNC has great athletes, but this doesn't always translate to the defensive end. I think proper technique and defensive strategy sets teams up to be great defenders.

I think UNC is defending well this year, but Roy will not change his strategy to guard the perimeter. Also, our athletes rely on help defense too much and are not able to stop ball penetration into the lane. On the flip side, Duke has less athletically gifted players at the guard positions, but they seem to be more capable of playing better, more intense defense. I believe it starts with the strategy and ends with the players willingness to commit to playing D. If you have the best athlete on your team have that desire to be a defensive stopper, then you have something. I believe John Henson has proved that this year while the guards seem to have a tough time stopping the ball and making shots.

loran16
02-16-2011, 03:29 PM
i would disagree with saying a coach can make great athletes into great defenders. I don't think anyone would deny that UNC has great athletes, but this doesn't always translate to the defensive end. I think proper technique and defensive strategy sets teams up to be great defenders.


Err so you mean you agree with that statement.

(Incidentally if you don't believe coaching leads to great defense, check out Calipari's teams' defense every year, though for once they have a down year this year. And I wouldn't say he recruits based on D)

ns7
02-16-2011, 04:54 PM
I think a down year for the ACC is solid compared to most conferences and the league isn't as weak as some would say. On the other hand, I don't know if I would expect any ACC team to win a single game in the NCAAT besides Duke and UNC. I don't think I would pick Virginia Tech, FSU (w/o Singleton), or Clemson vs. most 5 to 9 seeds. I hope you're right to be optimistic, but it would be a major upset to have 3 ACC teams through the first weekend based on how teams have played to date. This can certainly change in the next three weeks, though.

I think the real problem with the ACC this year is just how bad the bottom of the conference is. Wake, NC State, and Georgia Tech (20 point win over UNC excluded) have been terrible so far. Going back to 2003, the ACC has only had two teams outside of the top 100: Clemson in 2003 and Virginia 2010. This year both Virginia and Wake (245!!!) are outside. Also Georgia Tech and NC State are both close, I wouldn't be surprised to see one of the two drop out by the end of the year.

The top half is comparable to the past, except there have been some close OOC losses this year by those teams and no real marquee wins (UNC-Texas, VaTech-Purdue, FSU-Florida, NCSU-Syracuse). Flip half of those and the conference suddenly has some pretty good wins.

JasonEvans
02-17-2011, 09:13 AM
For folks who have been tracking KenPom's attempt to mathematically find a player of the year, there is an interesting update (http://kenpom.com/blog/index.php/weblog/the_kpoy_with_freshmen/).

At long last, the top 3 players in the KPOY match the three players that everyone agrees are the front-runners for the other POY awards.


1. Jimmer Fredette, Brigham Young (Rating of .542, last week: 2nd)
2. Jared Sullinger, Ohio St. (.523, LW: 1st)
3. Nolan Smith, Duke (.503, LW: 4th)
4. Jon Leuer, Wisconsin (.480, LW: 3rd)
5. Jordan Hamilton, Texas (.470, LW: 5th)

Nolan has been steadily climbing the past few weeks and may be in position to make a real run. I think last night will help him again, as he had a big statistical night in a game where he accounted for almost all of his team's offensive success.

--Jason "Ken also ranks Freshman this week-- and notes that HBarnes is now, finally, one of the top 5 freshmen in the country" Evans

JasonEvans
02-17-2011, 09:35 AM
I thought it would be interesting to look at Duke's efficiency numbers before and after that uuuugly game with Virginia to see how they were impacted.

We had a lot of turnovers (18) in the game, but we did shoot 50% from the field. On defense, our def FG% was boss. They shot a pathetic 28% from the field and we outrebounded them by a lot.

Anyway, Duke came into this game with a KenPom offensive rating of 120.5 and a defensive rating of 87.6. Our overall efficiency rating was .9751.

As you might expect, our offense took a bit of a hit, falling to a rating of 119.0 (down 1.5), but our D surged nicely to an 86.2 (an improvement of 1.4 points). Our new overall efficiency number is .9760, up .0009 from the previous day.

Now, I know that not all of this impact comes from just the UVA game. Other teams played and those games changed their ratings which, in turn can affect our ratings if they are a team we have played in the past. Still, I think it is sorta interesting that Duke seemed to improve slightly in the KP ratings despite playing what many consider to be a pretty sub-par game... especially on offense.

-Jason "yaaaay! We gained .0009!! What does that even mean? :) " Evans

MChambers
02-17-2011, 09:54 AM
I thought it would be interesting to look at Duke's efficiency numbers before and after that uuuugly game with Virginia to see how they were impacted.

We had a lot of turnovers (18) in the game, but we did shoot 50% from the field. On defense, our def FG% was boss. They shot a pathetic 28% from the field and we outrebounded them by a lot.

Anyway, Duke came into this game with a KenPom offensive rating of 120.5 and a defensive rating of 87.6. Our overall efficiency rating was .9751.

As you might expect, our offense took a bit of a hit, falling to a rating of 119.0 (down 1.5), but our D surged nicely to an 86.2 (an improvement of 1.4 points). Our new overall efficiency number is .9760, up .0009 from the previous day.

Now, I know that not all of this impact comes from just the UVA game. Other teams played and those games changed their ratings which, in turn can affect our ratings if they are a team we have played in the past. Still, I think it is sorta interesting that Duke seemed to improve slightly in the KP ratings despite playing what many consider to be a pretty sub-par game... especially on offense.

-Jason "yaaaay! We gained .0009!! What does that even mean? :) " Evans

What it means to me is that our defense improved, at least for one game. I hope it is a trend.

I'm not too worried about our offense – I'm sure Kyle will get his scoring touch back.

jaygdevil11
02-17-2011, 01:41 PM
What it means to me is that our defense improved, at least for one game. I hope it is a trend.

I'm not too worried about our offense – I'm sure Kyle will get his scoring touch back.

Last night moved us up to #3 in Kenpom defense and 2nd overall behind OSU. It seems like in the past I remember hearing that if you are top 10 in both offense and defense late in the year, you are in pretty good shape. That would leave us, OSU, and Kansas as the only teams in the top 10 in both categories. Although, last year Kansas finished the year in the top 10 and we all know what happened to them.

Reilly
02-17-2011, 01:57 PM
... I think it is sorta interesting that Duke seemed to improve slightly in the KP ratings despite playing what many consider to be a pretty sub-par game... especially on offense. ....

Duke has 11 ACC wins. Five of them were by greater margins than last night's 15. Five of them were by lesser margins than last night's.

However, last night, Duke scored 36% more points than our opponent: 15 points more than its 41.

The only ACC wins where Duke did better are when Duke scored 46% more than NCSU at home, and 41% more than WFU.

So, in one respect, and when considering offensive and defensive efficiency in tandem, this was Duke's third most-impressive ACC win this year. Its third biggest blowout. Felt that way, too, despite the 18 TOs and sloppiness ... that is, it never felt threatened, as expected during the third biggest blowout of the ACC season.

jaygdevil11
02-17-2011, 02:12 PM
Duke has 11 ACC wins. Five of them were by greater margins than last night's 15. Five of them were by lesser margins than last night's.

However, last night, Duke scored 36% more points than our opponent: 15 points more than its 41.

The only ACC wins where Duke did better are when Duke scored 46% more than NCSU at home, and 41% more than WFU.

So, in one respect, and when considering offensive and defensive efficiency in tandem, this was Duke's third most-impressive ACC win this year. Its third biggest blowout. Felt that way, too, despite the 18 TOs and sloppiness ... that is, it never felt threatened, as expected during the third biggest blowout of the ACC season.

Add the fact that Virginia is 3rd in scoring defense in ACC games thus far. The game looked ugly from a far but that is just the kind of game Virginia plays.

SCORING DEFENSE G Pts Avg/G
1. Duke 12 757 63.1
2. Clemson 11 695 63.2
3. Virginia 11 704 64.0
4. Florida State 11 707 64.3
5. Virginia Tech 11 729 66.3
6. North Carolina 11 743 67.5
7. Georgia Tech 11 771 70.1
8. Miami 11 776 70.5
9. Maryland 11 778 70.7
10. NC State 10 742 74.2
11. Boston College 11 817 74.3
12. Wake Forest 11 894 81.3

loran16
02-17-2011, 02:22 PM
Add the fact that Virginia is 3rd in scoring defense in ACC games thus far. The game looked ugly from a far but that is just the kind of game Virginia plays.

SCORING DEFENSE G Pts Avg/G
1. Duke 12 757 63.1
2. Clemson 11 695 63.2
3. Virginia 11 704 64.0
4. Florida State 11 707 64.3
5. Virginia Tech 11 729 66.3
6. North Carolina 11 743 67.5
7. Georgia Tech 11 771 70.1
8. Miami 11 776 70.5
9. Maryland 11 778 70.7
10. NC State 10 742 74.2
11. Boston College 11 817 74.3
12. Wake Forest 11 894 81.3

No they're not. Virginia's "defense" looks good becaue it has played by far the slowest of any ACC team.

Before this week (http://www.dukebasketballreport.com/forums/showthread.php?24343-ACC-Efficiency-Rankings-(Week-6)-State-s-not-11!&p=475145#post475145), Virginia was 9th out of 12 in Defensive efficiency, giving up 1.05 points per possession.

camion
02-20-2011, 10:43 AM
Here's a dork tidbit for Sunday morning, Experience.

Comparing the Big East and ACC in the kenpom experience category we find that only 4 of the 12 ACC teams, 33% are in the top half of the NCAA while 10 of the 16 Big East teams, 62.5% are in the the top half. I see this as one contributing factor to the ACC being down.


St. Johns 2.64 3
Notre Dame 2.40 18
W. Va 2.35 27
B.C. 2.24 33
Va Tech 2.21 37
Seton Hall 2.10 59
Clemson 2.00 83
Georgetown 1.99 88
Villanova 1.94 104
Cincinnati 1.94 106
Pitt 1.91 117
Marquette 1.86 134
Rutgers 1.82 149
Fla St. 1.74 177
--NCAA Avg-- 1.73
So. Fla 1.72 181
Maryland 1.72 182
Duke 1.65 206
Louisville 1.56 230
Miami 1.43 263
DePaul 1.39 272
Syracuse 1.38 276
Virginia 1.37 280
Providence 1.14 314
UNC 1.14 315
Ga Tech 1.11 318
NCSU 1.08 322
Wake Forest 0.98 328
UCONN 0.95 330

sagegrouse
02-20-2011, 10:53 AM
Here's a dork tidbit for Sunday morning, Experience.

Comparing the Big East and ACC in the kenpom experience category we find that only 4 of the 12 ACC teams, 33% are in the top half of the NCAA while 10 of the 16 Big East teams, 62.5% are in the the top half. I see this as one contributing factor to the ACC being down.



I believe you, Camion, that there is some meaning here, but what-the-heck is this table?

sagegrouse

loran16
02-20-2011, 11:02 AM
I believe you, Camion, that there is some meaning here, but what-the-heck is this table?

sagegrouse

Pretty sure it's the effective experience table. The table measures the average experience in years of players on each team, weighting the number by how many minutes each player plays..

St. John's is near the top of all of NCAA as they have 10 seniors (A senior would have 3 years of experience if they played all 3 seasons.)

Olympic Fan
02-20-2011, 11:57 AM
Just one other note about Pomeroy and the perception that UNC is improving.

Just after Duke's come-from-behind win over the Heels in Cameron, Pomeroy listed the odds for the rematch as Duke 51 percent. Those numbers have been steadily widening for the last two weeks -- after Saturday's UNC-FSU game, he had it at 59 percent.

That's still very close to a tossup (and indeed, he predicts the final score as 76-74), but the rising odds give lie to the percention that UNC is rapidly improving. Indeed, they made a big jump when Marshall replaced Drew at point guard, but they've plateaued out as other teams have learned to play the "new" Heels.

loran16
02-21-2011, 01:23 AM
Duke just returned to #1 in the Pomeroy Rankings.

That said, don't expect Duke to be #1 in the polls....most people who've revealed their ballots still have Duke 4-5 for some strange reason.

ns7
02-21-2011, 09:05 AM
Duke just returned to #1 in the Pomeroy Rankings.

That said, don't expect Duke to be #1 in the polls....most people who've revealed their ballots still have Duke 4-5 for some strange reason.

Duke jumping to #1 in Pomeroy is pretty much a function of Ohio State dropping after losing by 13.

Though I've been quite pleased with the team's play since the second half of the UNC game. I just pray that there are no slow starts @VaTech and @UNC.

One last note... Pomeroy wrote a few years ago that teams with both offense and defense in the top 10 do well in the NCAA tournament. Right now that group includes Duke, Ohio State, and Kansas, with Purdue's offense just missing at #14 and BYU's defense missing at #15. Pitt, Texas, Kentucky, and Washington all have subpar defenses for top contenders.

NSDukeFan
02-21-2011, 09:57 AM
Duke jumping to #1 in Pomeroy is pretty much a function of Ohio State dropping after losing by 13.

Though I've been quite pleased with the team's play since the second half of the UNC game. I just pray that there are no slow starts @VaTech and @UNC.

One last note... Pomeroy wrote a few years ago that teams with both offense and defense in the top 10 do well in the NCAA tournament. Right now that group includes Duke, Ohio State, and Kansas, with Purdue's offense just missing at #14 and BYU's defense missing at #15. Pitt, Texas, Kentucky, and Washington all have subpar defenses for top contenders.

Texas actually has the top ranked defense in the country. It is their offense (#29) that is not in the top 10.

ns7
02-21-2011, 11:16 AM
Texas actually has the top ranked defense in the country. It is their offense (#29) that is not in the top 10.

Apologies for the mistake.

You're right that Texas' defense has been lights out. It's amazing to see what they've done to their opponents in Big 12 play. I watched a couple of their games and they have great man-to-man defenders.

However, it's interesting that they lost because their defense had an off night. I would have assumed that their offense did not show up on Saturday.

loran16
02-22-2011, 09:09 AM
Of Note: After Kansas' blowout win last night, the top 3 teams (Still #1 Duke, #2 OSU, #3 Kansas) are all essentially identical in ranking.

The difference between the three teams is .0003 (of Pythag winning percentage):
Duke: .9767
OSU: .9766
KU: .9764
Tex: .9709

All Three of the top 3 teams have offenses and defenses within the top 8 of the country, making them incredibly well balanced. After those 3 however, no team is amazingly balanced and all the other teams excel at one of O or D, while are only good at the other. That's not to say those teams can't win it all (Texas' Defense is so far ahead of the pack that it certainly makes them a threat, even if their offense is only 29th in the country) but it shows that these three teams are currently a step above from the competition.

So ignore the media's silliness over 7 contenders. Really, there are only 3 or 4.

NSDukeFan
02-22-2011, 09:28 AM
Of Note: After Kansas' blowout win last night, the top 3 teams (Still #1 Duke, #2 OSU, #3 Kansas) are all essentially identical in ranking.

The difference between the three teams is .0003 (of Pythag winning percentage):
Duke: .9767
OSU: .9766
KU: .9764
Tex: .9709

All Three of the top 3 teams have offenses and defenses within the top 8 of the country, making them incredibly well balanced. After those 3 however, no team is amazingly balanced and all the other teams excel at one of O or D, while are only good at the other. That's not to say those teams can't win it all (Texas' Defense is so far ahead of the pack that it certainly makes them a threat, even if their offense is only 29th in the country) but it shows that these three teams are currently a step above from the competition.

So ignore the media's silliness over 7 contenders. Really, there are only 3 or 4.

Great post, though I disagree with your last line. In statspeak, OSU, KU and Duke may be ahead of the pack and are currently the top contenders. But, I wouldn't call any of them dominant and would doubt that there is much greater than a 50% chance (we are on a dork/ stat thread afterall), if even that, that the NCAA champion will come from those 3. I would argue that there are at least 7, if not more, solid contenders this year. I realize this is mostly semantics, as you are looking at the top contenders.

Wander
02-22-2011, 09:45 AM
So ignore the media's silliness over 7 contenders. Really, there are only 3 or 4.

Tempo-free statistics are awesome and I look at them all the time, but you're taking them way, way too seriously if you look at them and decide Pitt or Purdue isn't a contender.

loran16
02-22-2011, 10:47 AM
Tempo-free statistics are awesome and I look at them all the time, but you're taking them way, way too seriously if you look at them and decide Pitt or Purdue isn't a contender.

I don't think that's the case, particularly not in the case of Pitt. Pitt's never played as efficiently as the top teams all years, and this is due to defense that isn't top 10 or even top 20.

Sure, in a tournament they can win, but it raises major red flags.

I'm more partial to Purdue because I'm more partial to teams who are top 10 in Defense (Purdue is #5) and whose "weakness" is on O (#15 for Purdue), but there's a red flag there as well.

Essentially my point is this:
The efficiency statistics all year have loved Duke, KU, and OSU. Texas is the only team that has moved up above any of those three. Two of the other "top teams" as ranked by the media, Pitt and especially SDSU, have never come close to these three. That raises questions about the conventional wisdom there.

BYU incidentally is ahead of both of those two teams but still behind the other 4 by some margin. But it's ranked way higher than SDSU, and has been ranked higher for nearly all of this year, raising special questions as to who the real power is in the MWC.

Wander
02-22-2011, 11:14 AM
Essentially my point is this:
The efficiency statistics all year have loved Duke, KU, and OSU. Texas is the only team that has moved up above any of those three. Two of the other "top teams" as ranked by the media, Pitt and especially SDSU, have never come close to these three. That raises questions about the conventional wisdom there.

BYU incidentally is ahead of both of those two teams but still behind the other 4 by some margin. But it's ranked way higher than SDSU, and has been ranked higher for nearly all of this year, raising special questions as to who the real power is in the MWC.

I think it's good to use this stuff to raise questions about the conventional wisdom, but efficiency rankings aren't the final word on everything, even if they are better than other systems. I'm not convinced that every stat should be possession-based, and Pomeroy's ratings have badly missed on teams in the past that conventional wisdom had a much better feel for: Duke 2007, Gonzaga a few times recently, Memphis in 2009, and most relevantly, BYU last year.

loran16
02-22-2011, 11:27 AM
I think it's good to use this stuff to raise questions about the conventional wisdom, but efficiency rankings aren't the final word on everything, even if they are better than other systems. I'm not convinced that every stat should be possession-based, and Pomeroy's ratings have badly missed on teams in the past that conventional wisdom had a much better feel for: Duke 2007, Gonzaga a few times recently, Memphis in 2009, and most relevantly, BYU last year.

I don't think that Pomeroy or anyone can predict the tournament precisely....in a one and done format, even unlikely probabilities can occur (see Cornell last year averaging 1.5+ points per offensive possession in the first two rounds).

That said, it does highlight for sure the most likely contenders, particularly among the teams it identifies as the tippy top. Those teams certainly can lose early, ala KU last year. No Question. But there's a reason one of the top 2 teams in his rankings has won the whole thing each and every year except for 2003 (Cuse).

In other words, being in the top tier doesn't guarantee you anything. But it makes you by far the most likely to go far in the tourney.

Outside of the 3, well let's say 4 teams, those other teams do not qualify. Which raises serious questions.

Jderf
02-22-2011, 11:31 AM
... and Pomeroy's ratings have badly missed on teams in the past that conventional wisdom had a much better feel for: Duke 2007, Gonzaga a few times recently, Memphis in 2009, and most relevantly, BYU last year.

In defense of kenpom, that's about five or six counterexamples (and I would still disagree on Memphis and BYU) in a sample size of 340 over the span of 8 years. That's actually pretty impressive.

Pomeroy's system isn't perfect. No system is. But it is a very powerful analytic tool, and his numbers can really help highlight aspects of teams that may have been missed otherwise.

Wander
02-22-2011, 11:49 AM
I don't think that Pomeroy or anyone can predict the tournament precisely....in a one and done format, even unlikely probabilities can occur (see Cornell last year averaging 1.5+ points per offensive possession in the first two rounds).


I'm not talking about the unlikely stuff that happens because of the one-and-done nature of the tournament. Kenpom had BYU last year and Memphis the year before as serious national title contenders. But conventional wisdom "knew" they weren't. All I'm saying is that the argument of "It's silly to say team X is/isn't a title contender because efficiency rankings says they aren't/are" isn't a valid argument by itself.



In defense of kenpom, that's about five or six counterexamples (and I would still disagree on Memphis and BYU) in a sample size of 340 over the span of 8 years. That's actually pretty impressive.


Well, that was just an off the top of my head kind of thing that only considered the top 10ish out of 340 teams each year. I'm sure there are plenty more. But it doesn't matter, we're in complete agreement anyway - it's a great, impressive, and imperfect tool.

davekay1971
02-22-2011, 12:17 PM
In perusing these rankings today it's interesting to how well the ACC represented in national defensive efficiency rankings. Duke is, of course, number 2 for defensive efficiency. We're closely followed by UNC, at number 3, and Florida State, at number 4. Also in the top 20 nationwide for defensive efficiency are Maryland (number 11) and Clemson (number 18). Lurking outside the top 20 is Va Tech (number 30 defensive efficiency).

To me this begs the question: is some of the unwatchable hoops we're seeing in the ACC this year due to the fact that half of the ACC teams are in the top 30 nationwide for defensive efficiency?

superdave
02-22-2011, 12:22 PM
In perusing these rankings today it's interesting to how well the ACC represented in national defensive efficiency rankings. Duke is, of course, number 2 for defensive efficiency. We're closely followed by UNC, at number 3, and Florida State, at number 4. Also in the top 20 nationwide for defensive efficiency are Maryland (number 11) and Clemson (number 18). Lurking outside the top 20 is Va Tech (number 30 defensive efficiency).

To me this begs the question: is some of the unwatchable hoops we're seeing in the ACC this year due to the fact that half of the ACC teams are in the top 30 nationwide for defensive efficiency?

Yes, I think this is a great point. I'd also take the further step of saying that bad point guard play in the ACC also contributes to defenses dominating offenses. The two best point guards are an off-guard who took over point duties and a freshman who is just now playing consistently well.

COYS
02-22-2011, 12:22 PM
I think it's good to use this stuff to raise questions about the conventional wisdom, but efficiency rankings aren't the final word on everything, even if they are better than other systems. I'm not convinced that every stat should be possession-based, and Pomeroy's ratings have badly missed on teams in the past that conventional wisdom had a much better feel for: Duke 2007, Gonzaga a few times recently, Memphis in 2009, and most relevantly, BYU last year.

Were these really bad misses, though? I don't think Kenpom is the end-all-be-all of rankings, by any means. However, it's not like BYU was in the top 4. They were a solid top 10 team, but they were not one of his picks to make the final four (based purely on the rankings). Memphis was number 1 heading into the 2009 tournament and might be the best example of a bad miss for Kenpom. However, they made the sweet 16 (better than number 2 Kansas did last year) and Pomeroy strongly cautioned against picking them as a favorite because of their lack of battle testing games as they utterly destroyed weak competition. Indeed, the team lacked the leadership to overcome adversity in the tournament. Duke 2007 is an odd choice for a miss, as well, in my opinion. The offense was very poor for a top 15 pomeroy team and was only buoyed by a top 5 defensive rating. The team had a very small margin for error (and indeed, lost the first round game by a very small margin). That team was definitely not a title contender. Remember that even the teams that are favorite usually have a very significant chance of losing (I wish Pomeroy kept his prediction percentages up so we could see how often he is right).

If there are any glaring misses, I'd point to Kansas last year. Despite the loss to Northern Iowa and the success of other teams that continued in the tournament, thus boosting their rankings, Kansas remained the number 2 team at the end of the season. Northern Iowa topped out at number 77 with neither an offense nor a defense that was even in the top 50 of Pomeroy's rankings. That was a huge upset. However, I don't think anyone would disagree that Kansas was a very good team last year and had as good a shot as any of the other top teams at the title. They just had a bad day. Even if they had an 80 or 90 percent chance of winning the game, that means that either 1 in 5 or 1 in 10 games played against a similarly over-matched opponent will result in an upset.

ns7
02-22-2011, 12:23 PM
I'm not talking about the unlikely stuff that happens because of the one-and-done nature of the tournament. Kenpom had BYU last year and Memphis the year before as serious national title contenders. But conventional wisdom "knew" they weren't.

#10 BYU lost to #7 Kansas State. If anything BYU was unlucky to have faced a strong #2. I imagine BYU could have knocked off #21 Villanova (a 2 seed). By the way, Pomeroy was spot on about Villanova being an overrated 2 seed.

#2 Memphis lost to #6 Missouri--again hardly a knock against the system. Battles between top 6 teams are essentially tossups.