PDA

View Full Version : In this thread, we pretend Joe Lunardi's opinion matters



Eakane
03-02-2014, 03:29 PM
Is it me, or just Joe look for excuses not to put Duke into a #1 slot? Instead of switching us with Syracuse,which has now lost 3 of their last five, and barely won the previous two before that, Lunardi slates Kansas,which JUST LOST their 7th game in front of us, based on their "total body of work."

We have work to do. But if we win out, and KU does too, we'll have earned that coveted 4th spot ahead of KU.

Calling Texhawk: do you disagree?

wilko
03-02-2014, 03:38 PM
Still more ball to be played.
We have 3 top 5 losses. (SU, UA, KU)... and 1 win SU (at home)
Wouldn't hurt to win the ACC Tourney to make a stronger case. We need to TAKE it instead of stumbling into it.
U of M and UVa look like our next best wins.


Beating UNC and SU 1 more time would go alooooong way proving I think.

hurleyfor3
03-02-2014, 03:46 PM
We should have the Kardashians put out a bracket each week. They need things to do and it's not like it would be any less official than what Joe Lunardi thinks. It would be interesting teevee two or three times. Once the real bracket comes out they switch to NFL mock drafts.

DavidBenAkiva
03-02-2014, 03:47 PM
My hunch is Lunardi's probably right about this. The polls are about how teams appear at the moment. Right now, Duke looks like a lock for that #4 ranking in the AP and Coaches polls. Seeding for the tournament is based on total body of work.

Right now, Kansas has a better resume than Duke. KU beat Duke head-to-head, is the regular season champion of its conference (Duke will get 2nd at best and may end up anywhere from 2nd to 4th by the end of next week), and has fewer bad losses than Duke (Notre Dame and Clemson are both out of the RPI Top 50 while none of KU's losses have been to "bad" teams).

Take the example of one of the hottest teams in the country right now for another seeding vs. poll comparison: Virginia. With those 4 non-conference losses, Virginia just doesn't have enough of a resume beyond the victory over the slumping Orange to move up much in tournament seeding beyond a 3 seed right now. They may get a 2 seed, but they would have to make it to the ACC Tourney title game at the least. In the polls, they will probably find themselves in the 5-6 range this upcoming week.

If Duke hadn't lost at ND and Clemson, they'd be right there for a #1 seed with Florida, Wichita State, and Arizona. Unfortunately, Syracuse and Kansas can claim a total season resume as good as or maybe even better than Duke's. That's just how it is.

Now, if Duke wins out next week and claims the ACC Tournament title, it will have certainly put itself in position to earn a #1 seed.

burns15
03-02-2014, 03:54 PM
Still more ball to be played.
We have 3 top 5 losses. (SU, UA, KU)... and 1 win SU (at home)
Wouldn't hurt to win the ACC Tourney to make a stronger case. We need to TAKE it instead of stumbling into it.
U of M and UVa look like our next best wins.


Beating UNC and SU 1 more time would go alooooong way proving I think.

I'd say that those are pretty good "next best" wins considering we are talking about the ACC and B1G regular season champs.

bob blue devil
03-02-2014, 03:55 PM
to oversimplify,
reason #1 is kansas rpi = 2, duke rpi = 7 (rpi is a terrible measure, but the committee loves it)
reason #2 is kansas beat duke head to head - if you're looking for an excuse to leap duke over kansas this doesn't help your cause
joe lunardi is not voicing his opinion of how the bracket should be, only how he guesses it would be based on how the committee does things. surprisingly, not everyone is out to get duke.

of course, if duke wins out its resume gets much better - wins over unc, and possibly syracuse and virginia, which is better than what kansas could do - maybe texas and iowa state. so that would change things. i would be surprised if we don't get a #1 if we win out.

tommy
03-02-2014, 03:56 PM
Is it me, or just Joe look for excuses not to put Duke into a #1 slot? Instead of switching us with Syracuse,which has now lost 3 of their last five, and barely won the previous two before that, Lunardi slates Kansas,which JUST LOST their 7th game in front of us, based on their "total body of work."

We have work to do. But if we win out, and KU does too, we'll have earned that coveted 4th spot ahead of KU.

Calling Texhawk: do you disagree?

I don't think much of Lunardi in general, but Kansas is #2 in the RPI, has played an insane schedule, won its conference regular season title, and beat us head-to-head. Still some ball to be played, and we can earn our way past them if they falter -- especially if Embiid's back continues to be a problem -- but right now, I can't think of a great argument why we should be ahead of the Jayhawks.

Troublemaker
03-02-2014, 04:11 PM
But if we win out, and KU does too, we'll have earned that coveted 4th spot ahead of KU.

I'm going to stop coveting it because getting a good bracket is about matchups and avoiding underseeded teams anyway. Let's say Duke gets the 1 seed and then gets Michigan St as the 4 seed and Ohio St or Kentucky as the 8 seed. I don't really covet that at all.

We have two shots at MSG -- being a 1 seed or being the highest-ranked 2 seed. It's hard to say which one is preferable without knowing the matchups ahead of time. I think we could just as easily receive a poor bracket (or good bracket) from either spot.

All of this 1-seed talk and MSG talk (to which I've contributed way too many posts myself) is depressingly moot with just one more loss by Duke. The stat geeks probably peg Duke with a 15% chance or so of winning out.

Winning out is the hard part. The glory that comes with winning out (beating UNC at least once, ACC championship) will probably make us not even care too much where we're seeded by the time it's over. I can't imagine seeing our guys holding an ACC championship trophy (which Duke hasn't held since 2011) and caring too much about whether Duke indeeds gets the final 1 seed. And, as mentioned, getting a good bracket is about luck of the draw anyway.

brevity
03-02-2014, 04:14 PM
We should have the Kardashians put out a bracket each week. They need things to do and it's not like it would be any less official than what Joe Lunardi thinks. It would be interesting teevee two or three times...

Well, if we're talking about Kim, I imagine the bubble watch will be a lot more interesting.

hurleyfor3
03-02-2014, 04:24 PM
Well, if we're talking about Kim, I imagine the bubble watch will be a lot more interesting.

See, that's the thing. They already have the bubble watch going on.

In fact I like the Bubble Watch column on espn. I don't closely follow who's in fifth place in the Big XII or which Big 10 teams barfed in the gutter last week, and it's nice to have all that info in one place.

wilko
03-02-2014, 05:01 PM
I'd say that those are pretty good "next best" wins considering we are talking about the ACC and B1G regular season champs.

They are FINE wins.
But to lock up a one seed, I think we need more. A split with KU/UA would have helped the argument.
The option on the table is to keep winning and playing well.

weezie
03-02-2014, 05:23 PM
We should have the Kardashians put out a bracket each week. They need things to do and it's not like it would be any less official than what Joe Lunardi thinks. It would be interesting teevee two or three times. Once the real bracket comes out they switch to NFL mock drafts.

Excellent idea. I hope they will eat kale salads and examine their split ends while they discuss the brackets.

weezie
03-02-2014, 05:25 PM
I don't closely follow who's in fifth place in the Big XII or which Big 10 teams barfed in the gutter last week, and it's nice to have all that info in one place.

You make me so proud! :D:D:D

hurleyfor3
03-02-2014, 05:42 PM
I told you I'd use it! And really, when you picture a college kid literally barfing in the gutter, isn't it outside the Sig Ep house at Michigan State or next to a dumpster in State College?

BTW, "Lunardi's bracketology" is redundant. I thought of changing the title to "In this thread, we pretend that what Joe Lunardi thinks actually matters" but some other mod would get mad.

sagegrouse
03-02-2014, 06:25 PM
They are FINE wins.
But to lock up a one seed, I think we need more. A split with KU/UA would have helped the argument.
The option on the table is to keep winning and playing well.

Look, Lunardi's making a buck doing brackets during the season. Good for him! It's nonsense because the season isn't over, and there will be determinative results in the conference tournaments (as well as Duke-UNC).

Two questions that shape my thinking (or, if you wish, my not-so-humble opinions).

1. Will any conference get two #1 picks? I think, no.

2. Can any team get a #1 seed if it loses, aside from the case where most of the top seeds lose in their conference tournaments? Yes, Arizona, which is #1 in both RPI and KenPom. The Wildcats are a lock, barring two losses.

Then, we are playing for three open #1 positions. Here are the teams whose record is sufficiently strong to be deserving of #1 seeds, with the RPI and KenPom ranking:

Kansas (9/2)
Florida (4/3)
Duke (3/7)
Syracuse (11/2)
Virginia (2/10)
Wichita State (6/8)

I dont think that Louisville (5/31), Cincy (18/16), Creighton (7/9), Villanova (8/5) and UNC (18/22) will get strong consideration for #1, and winning their tournaments are likely to produce a #2 seed.

Six teams will vie for three open #1 seeds, and -- guess what --- at least two will lose in the ACC Tournament, which leads us to:

Kansas
Florida
Duke, Syracuse, or Virginia
Wichita State

To me, the problem is easy. No way does Wichita State, playing only one top team, deserve a #1 seed over the other conference champions. But, if politics decree otherwise, I would still favor Duke over Kansas if both win the rest of their games. Duke would then have a record of 5-2 against Va., Syra. and UNC, better than anything Kansas can muster, and Duke's RPI would be much better than the Jayhawks.' Moreover, the likelihood is that someone other than the teams listed will show up as a conference champion.

Kedsy
03-02-2014, 06:26 PM
Take the example of one of the hottest teams in the country right now for another seeding vs. poll comparison: Virginia. With those 4 non-conference losses, Virginia just doesn't have enough of a resume beyond the victory over the slumping Orange to move up much in tournament seeding beyond a 3 seed right now. They may get a 2 seed, but they would have to make it to the ACC Tourney title game at the least. In the polls, they will probably find themselves in the 5-6 range this upcoming week.

I disagree. As you say, UVa will probably be around 5th or 6th (maybe 7th or 8th, but definitely top 10) in the polls this week, along with a #10 RPI (with a SOS of #25), and #2 in Pomeroy. I don't care about non-conference losses that happened two months ago (and I suspect the committee only would care about them if UVa is "tied" for a spot with some other team with no bad losses), that's a #2 seed profile right there. If the Hoos win out, both their AP rank and their RPI ranking will get even better (maybe Pomeroy, too, but it's hard to get too much better than second).

So I think you're wrong. If Virginia wins out, they're a solid #2 and a threat for the last #1. If they get to the ACC final and lose, they're still almost a lock for a #2. A semifinal ACC loss to Syracuse or Duke and I still see Virginia as a likely #2 seed. In my mind, the only way they drop to a #3 is if they lose to Maryland and/or lose early in the ACCT, and even then it depends on what the other contenders do.

TexHawk
03-02-2014, 06:41 PM
Is it me, or just Joe look for excuses not to put Duke into a #1 slot? Instead of switching us with Syracuse,which has now lost 3 of their last five, and barely won the previous two before that, Lunardi slates Kansas,which JUST LOST their 7th game in front of us, based on their "total body of work."

We have work to do. But if we win out, and KU does too, we'll have earned that coveted 4th spot ahead of KU.

Calling Texhawk: do you disagree?

I do, but I'm not going to lose sleep over it if I'm wrong. KU has had plenty of opportunity to put this argument to bed, they just couldn't get over the hump against 'Nova (three pointer with 3 seconds left), Colorado (buzzer beater), SDSU (missed a FT down 1 with 20 seconds left), KSU (overtime). Make a play in one or two of those, and nobody is freaking out about a possible #1 seed with 7 losses.

But yea, on the whole, if a #1 RPI team** wins the highest rated conference regular season and conference tournament championships, I think it should get the nod over another team that does not (assuming similar resumes). I don't think the head-to-head matchup will matter much at all.

This is the least of my worries at this point, as the team just didn't want to respond to the despicable-ness of Marcus Smart last night. Injured Embiid or not. KU has a game in Morgantown next weekend that scares me too.

** I think the RPI difference between KU and Zona is like .002, so if KU wins out, that probably means a win or two over highly rated RPI teams, which would push them above Zona again (since they may not face a decent team at all in the next two weeks). At least that's what I'm told.

Kedsy
03-02-2014, 06:47 PM
2. Can any team get a #1 seed if it loses, aside from the case where most of the top seeds lose in their conference tournaments? Yes, Arizona, which is #1 in both RPI and KenPom. The Wildcats are a lock, barring two losses.

Considering the Ashley injury, I think there's a pretty good chance Zona gets those two losses. They play Stanford tonight, but assuming they get past that, they finish with the Oregon trip. Arizona's last five road games include one good win over Colorado, two losses (to Cal and Arizona State), a 3 point win over Stanford and a 4 point win over Utah. They could easily lose at Oregon, and then again in the Pac-12 tourney.


Syracuse (11/2)

Syracuse is currently #6 in the RPI (not #2), according to ESPN. If they lose before the ACC tournament final, Syracuse will be closer to dropping to a #3 seed than to getting the #1.


Six teams will vie for three open #1 seeds, and -- guess what --- at least two will lose in the ACC Tournament, which leads us to:

Kansas
Florida
Duke, Syracuse, or Virginia
Wichita State

To me, the problem is easy. No way does Wichita State, playing only one top team, deserve a #1 seed over the other conference champions. But, if politics decree otherwise, I would still favor Duke over Kansas if both win the rest of their games. Duke would then have a record of 5-2 against Va., Syra. and UNC, better than anything Kansas can muster, and Duke's RPI would be much better than the Jayhawks.' Moreover, the likelihood is that someone other than the teams listed will show up as a conference champion.

I don't think the bolded part of your quote is right. Kansas is currently #2 in the RPI (with the #1 schedule) and Duke is #7. If both Duke and Kansas win out, I can't see how Kansas' RPI could possibly drop lower than Duke's. The choice in that case could go either way, but since the committee appears to put more stock in the RPI than it does in Pomeroy, I'd probably have to give the edge to Kansas.

Also, if Wisconsin wins out, meaning they'd be Big 10 tournament champion with a top 4 RPI (with the #2 SOS), they have a shot at a #1.

DavidBenAkiva
03-02-2014, 08:56 PM
I don't care about non-conference losses that happened two months ago (and I suspect the committee only would care about them if UVa is "tied" for a spot with some other team with no bad losses)

The committee almost certainly does consider losses that happened two months ago. Remember that Wisconsin was one of the losses from months ago. If Wisconsin wins the B1G tournament, which is a possibility, they would vault themselves ahead of Virginia. The other thing about UVA is their absolute lack of "signature wins." They have played 3 top-tier teams (Duke, Syracuse, and Wisconsin) and gone 1-2 against them. Their resume is not all that strong compared to some of the other potential #1 and #2 seeds at the moment.

But I think we agree to a great extent - if UVA wins their final games and makes a strong run in the ACC Tournament, they will build a better resume than they have now. I just think other teams like Kansas, Wisconsin, Duke, Syracuse, and Louisville will have an ample opportunity to build a better resume when it's all said and done. UVA is closer to a #3 seed than a #1 seed, even with a good showing in the polls.

Kedsy
03-02-2014, 09:55 PM
The committee almost certainly does consider losses that happened two months ago. Remember that Wisconsin was one of the losses from months ago. If Wisconsin wins the B1G tournament, which is a possibility, they would vault themselves ahead of Virginia. The other thing about UVA is their absolute lack of "signature wins." They have played 3 top-tier teams (Duke, Syracuse, and Wisconsin) and gone 1-2 against them. Their resume is not all that strong compared to some of the other potential #1 and #2 seeds at the moment.

But I think we agree to a great extent - if UVA wins their final games and makes a strong run in the ACC Tournament, they will build a better resume than they have now. I just think other teams like Kansas, Wisconsin, Duke, Syracuse, and Louisville will have an ample opportunity to build a better resume when it's all said and done. UVA is closer to a #3 seed than a #1 seed, even with a good showing in the polls.

Obviously none of us know exactly what goes on behind the closed doors of the selection committee meetings, but despite the fact that they talk about such things in public, I doubt "signature wins" and "bad losses" are the major factors that many people outside the room make them out to be. I suspect instead that the committee makes its decisions based on body of work and only look at things like single game results (even head-to-head) when confronted with two or more teams the committee members feel are evenly matched, and then good wins/bad losses/etc. become sort of a tiebreaker (which is probably why they talk about those things in public, because they generally get asked about the close cases).

So while I agree that if Wisconsin wins out they have a good shot at a #1, probably a better shot than Virginia, it wouldn't be because Wisconsin beat Virginia head-to-head. I mean, yes, it would be because Wisconsin's resume is more impressive, but I assume the most important part of the resume is not signature wins and bad losses, but RPI, SOS (which shouldn't be a separate factor but appears to be), regular season championships and tournament championships, AP rank, W/L record, possibly alternative ranking systems like Pomeroy.

Because if Virginia really hadn't played anybody, they wouldn't have the #25 schedule in the country and their RPI (which is 75% based on schedule) wouldn't be as high as it is. In Wisconsin's and Virginia's case, you may be right that Wisconsin would get the seed ahead of Virginia, but if so it would in my opinion be primarily because Wisconsin's RPI is #4 (with the #2 SOS) while Virginia's is #10 (with the #25 SOS), and because the committee thinks winning the Big 10 tournament is more impressive than winning the ACC tournament.

Put another way, the reason the committee originally began relying on a rating system (unfortunately the RPI, but better than nothing) was to get away from making the decision solely based on win/loss record and who beat who. Those things obviously have some bearing, but I can't imagine they matter as much as some people believe.

Kedsy
03-02-2014, 10:16 PM
I just think other teams like Kansas, Wisconsin, Duke, Syracuse, and Louisville will have an ample opportunity to build a better resume when it's all said and done. UVA is closer to a #3 seed than a #1 seed, even with a good showing in the polls.

You talk about resume and then say Syracuse and Louisville are more impressive than Virginia, but I think if you look at the whole body of work, Virginia beats both of them hands down.

Comparing Virginia to Syracuse (assuming Virginia goes equally as far or farther in the ACC tournament than Syracuse does), Virginia beats Syracuse in most of the important things -- Virginia won the ACC regular season, at least two games (assuming UVa beats Maryland) ahead of Syracuse; Syracuse has a slightly better RPI (currently #6) but a worse SOS (#64, again with the caveat that SOS shouldn't be a component separate from RPI but does seem to be) compared with UVa's #10 RPI and #25 SOS. (Ironically, you seem to think Syracuse's schedule is a plus compared to Virginia's, but the computer systems seem to think otherwise.) In the polls, UVa and Syracuse would be similarly ranked, and Virginia has a much better Pomeroy ranking (#2 compared to Syracuse's #11). Plus, if the committee sees them as equal, UVa did win head-to-head as a tiebreaker, though as I said I doubt if it really would come down to that.

Compared to Louisville, it's not even that close:

RPI: UVa #10 (#25 SOS) vs. Louisville #31 (#110 SOS)
Championships: ACC more impressive than AAC
AP rank: I'm assuming UVa is going to vault up to somewhere between #5 and #8, presumably ahead of Louisville, which lost this weekend
W/L record: UVa (23-5) vs. Louisville (23-4), with (again, perhaps ironically given your argument) Virginia getting the edge because the Hoos have played a much harder schedule
Pomeroy: Virginia is #2 while Louisville is #5

Especially considering Louisville's mediocre RPI, in my opinion there's no way Louisville gets a seed ahead of Virginia except maybe if Louisville wins the AACT and UVa tanks in the first round of the ACCT.

I think Virginia vs. Duke is similar to Virginia vs. Syracuse, except that Duke's schedule strength is much better than Syracuse's. Ultimately I doubt it will matter so much because only one of the three teams can win the ACC tournament, but if both Duke and UVa go the same distance in the ACCT, I wouldn't be surprised to see UVa win the "seed battle" (despite Duke's head-to-head win, though I suppose if the committee thought the two teams were equal that might come into play) -- though I think Duke has a better shot of moving past UVa than Syracuse does.

Virginia vs. Kansas would probably tilt to Kansas (assuming both win out) because Kansas' RPI is #2 in the country (with the #1 SOS), much better than Virginia's #10 (and #25 SOS).

NYBri
03-02-2014, 11:56 PM
A month ago, we would've been happy to know that we would now be discussing a potential number one seed. Back then we looked like a six or seven.

This team is come along way.

CDu
03-03-2014, 09:06 AM
You talk about resume and then say Syracuse and Louisville are more impressive than Virginia, but I think if you look at the whole body of work, Virginia beats both of them hands down.

Comparing Virginia to Syracuse (assuming Virginia goes equally as far or farther in the ACC tournament than Syracuse does), Virginia beats Syracuse in most of the important things -- Virginia won the ACC regular season, at least two games (assuming UVa beats Maryland) ahead of Syracuse; Syracuse has a slightly better RPI (currently #6) but a worse SOS (#64, again with the caveat that SOS shouldn't be a component separate from RPI but does seem to be) compared with UVa's #10 RPI and #25 SOS. (Ironically, you seem to think Syracuse's schedule is a plus compared to Virginia's, but the computer systems seem to think otherwise.) In the polls, UVa and Syracuse would be similarly ranked, and Virginia has a much better Pomeroy ranking (#2 compared to Syracuse's #11). Plus, if the committee sees them as equal, UVa did win head-to-head as a tiebreaker, though as I said I doubt if it really would come down to that.

Compared to Louisville, it's not even that close:

RPI: UVa #10 (#25 SOS) vs. Louisville #31 (#110 SOS)
Championships: ACC more impressive than AAC
AP rank: I'm assuming UVa is going to vault up to somewhere between #5 and #8, presumably ahead of Louisville, which lost this weekend
W/L record: UVa (23-5) vs. Louisville (23-4), with (again, perhaps ironically given your argument) Virginia getting the edge because the Hoos have played a much harder schedule
Pomeroy: Virginia is #2 while Louisville is #5

Especially considering Louisville's mediocre RPI, in my opinion there's no way Louisville gets a seed ahead of Virginia except maybe if Louisville wins the AACT and UVa tanks in the first round of the ACCT.

I think Virginia vs. Duke is similar to Virginia vs. Syracuse, except that Duke's schedule strength is much better than Syracuse's. Ultimately I doubt it will matter so much because only one of the three teams can win the ACC tournament, but if both Duke and UVa go the same distance in the ACCT, I wouldn't be surprised to see UVa win the "seed battle" (despite Duke's head-to-head win, though I suppose if the committee thought the two teams were equal that might come into play) -- though I think Duke has a better shot of moving past UVa than Syracuse does.

Virginia vs. Kansas would probably tilt to Kansas (assuming both win out) because Kansas' RPI is #2 in the country (with the #1 SOS), much better than Virginia's #10 (and #25 SOS).

You're essentially overvaluing pre-season schedule by ignoring the results against relevant competition. Syracuse is 3-2 vs the top-25, 7-2 vs the top-50, and 14-2 vs the top-100. UVa? They're 2-3 vs the top-25, 4-4 vs the top-50, and 10-5 vs the top-100. Cuse has one bad loss (to BC) and UVa has one mediocre loss (to Green Bay). Frankly, unless UVa wins the ACC tourney or beats Syracuse along the way (unlikely without winning the ACC tourney) or unless Syracuse continues to flame out down the stretch, I think Syracuse has the better resume.

Duke and UVa are indeed a much closer case study. Duke is 4-4 vs the top-25, 5-4 against the top-50, and 10-5 vs the top-100. Duke has one mediocre loss (@Clemson) and one bad loss (@Notre Dame). I'd put Duke slightly ahead, but it's very close there. A win over UNC pushes us comfortably ahead of UVa, and a loss to UNC would basically make it a wash with UVa going into the ACC tourney.

UVa has Louisville beat. Louisville is 1-3 vs the top-25, 4-5 vs the top-50, and just 5-5 vs the top-100.

If I had to rate those four teams' resumes right now, it'd be:
1. Syracuse
2. Duke
3. UVa
4. Louisville

Kedsy
03-03-2014, 09:36 AM
You're essentially overvaluing pre-season schedule by ignoring the results against relevant competition. Syracuse is 3-2 vs the top-25, 7-2 vs the top-50, and 14-2 vs the top-100. UVa? They're 2-3 vs the top-25, 4-4 vs the top-50, and 10-5 vs the top-100. Cuse has one bad loss (to BC) and UVa has one mediocre loss (to Green Bay).

I'm not ignoring the results, the computer ranking system takes it all into account. If Virginia has a better ranking against a better schedule (as they do in Pomeroy, for example), what does it matter if they've lost 4 games vs. the top 50 instead of 2 games vs. the top 50 and one game to nowhere-near-the-top-100? While it's true that in the RPI, Syracuse has a slightly better ranking against a much worse schedule, the difference is slight, in which case I'd expect tiebreaker items like Virginia finishing two games ahead of Syracuse in the conference, beating Syracuse head-to-head, and Syracuse losing three of its last four games (while UVa has won 13 in a row) might come into play.

Put another way, if Syracuse had played as tough a pre-season schedule as Virginia, maybe Syracuse would have another loss (or two). The point of computer rankings is so humans don't have to parse whether a 5 point win against #62 St. John's is more or less impressive than a 3 point loss to #15 VCU.

Having said all that, I have no idea whether the committee ignores the computer rankings in favor of ordinal recitations of wins and losses or not. If they do, then maybe you're right about Syracuse having a better resume. They often say that they don't punish teams for playing a tough schedule, though, and it feels like that's what you're doing, especially since the only reason this is even a discussion is because people feel UVa lucked out in its ACC schedule and therefore its amazing ACC record somehow doesn't count.

ice-9
03-03-2014, 10:02 AM
Cuse has one bad loss (to BC) and UVa has one mediocre loss (to Green Bay).

This surprised the heck out of me, but Green Bay is actually a solid team; at least better rated than BC in the dork polls.

Green Bay: #51 on KenPom, #52 RPI

Boston College: #139 on KenPom, #179 RPI

So it's Syracuse with the mediocre loss and UVA with a bad loss. Whodathunk?

tommy
03-03-2014, 10:24 AM
This surprised the heck out of me, but Green Bay is actually a solid team; at least better rated than BC in the dork polls.

Green Bay: #51 on KenPom, #52 RPI

Boston College: #139 on KenPom, #179 RPI

So it's Syracuse with the mediocre loss and UVA with a bad loss. Whodathunk?

Yeah, Green Bay seems to be pretty good. They're being mentioned as a possible small/mid conference bubble team, meaning if they get upset in their conference tournament, they may merit a bid anyway.

UVA's losses to Green Bay and Tennessee are probably about equal. Both could end up on either side of the (artificial) top-50 divide, meaning the Wahoos could end up with two, one, or zero "bad" losses.

pfrduke
03-03-2014, 10:35 AM
Yeah, Green Bay seems to be pretty good. They're being mentioned as a possible small/mid conference bubble team, meaning if they get upset in their conference tournament, they may merit a bid anyway.

UVA's losses to Green Bay and Tennessee are probably about equal. Both could end up on either side of the (artificial) top-50 divide, meaning the Wahoos could end up with two, one, or zero "bad" losses.

Well, except that one was by 3 and one was by 35. It's not just who you lose to, it's also how you lose.

flyingdutchdevil
03-03-2014, 10:38 AM
Well, except that one was by 3 and one was by 35. It's not just who you lose to, it's also how you lose.

Question: does the Selection Committee care by how much you lose to a team? Or merely that you lost? I always thought that the level of beat down wasn't a factor, but I guess it's hard to remove this out of the ultimate bubble decisions.

Olympic Fan
03-03-2014, 11:35 AM
Lunardi just released his latest version this morning:

http://espn.go.com/mens-college-basketball/bracketology

Look at it closely, especially the Midwest Regional.

I've said all along that I'd rather be the No. 2 in Wichita State's bracket than No. 1 somewhere else, but to stack No. 1 Wichita and No. 2 Villanova together -- just wow!

Compare that to 1. Kansas, 2 Syracuse in the East; No. 1 Florida, No. 2 Duke in the South and No. 1 Arizona, No. 2 Wisconsin in the West.

That would be a great spot for No. 3 Virginia ...

BTW: Joe has Duke and Virginia playing in Raleigh with UNC in San Diego.

Let me just add that I don't take these brackets very seriously ... but I do think the committee has to respond to the Wichita situation (they HAVE to have a No. 1 seed, yet I do not believe they are one of the nation's top 10 teams) by paring them with the strongest No. 2 possible -- not the weakest (and frankly, I don't think Villanova is one of the nation's 10 toughest teams).

CDu
03-03-2014, 11:37 AM
I'm not ignoring the results, the computer ranking system takes it all into account. If Virginia has a better ranking against a better schedule (as they do in Pomeroy, for example), what does it matter if they've lost 4 games vs. the top 50 instead of 2 games vs. the top 50 and one game to nowhere-near-the-top-100? While it's true that in the RPI, Syracuse has a slightly better ranking against a much worse schedule, the difference is slight, in which case I'd expect tiebreaker items like Virginia finishing two games ahead of Syracuse in the conference, beating Syracuse head-to-head, and Syracuse losing three of its last four games (while UVa has won 13 in a row) might come into play.

Because the overall SOS takes into account nearly 1/3 of the schedule against patsies. And the difference in quality of said patsies makes a difference in the overall SOS. Every top-25 team has at least 8-10 games against teams they absolutely should not lose to. Syracuse just chooses to play some REALLY bad teams for those 8-10.

Against the top-100 teams in the country, Syracuse went 14-2. UVa went 10-5. Against the top-50, Syracuse went 7-2, UVa went 4-4. Against the top-25, Syracuse went 3-2, UVa went 1-3. So if UVa has a stronger SOS, that's based on playing more teams in the 100-150 than in the 200-300 range, because Syracuse has played more teams in the top-25 and as many teams in the top 25-100.

So if you have a team that has played more upper-tier teams (and has a substantially better record against such teams), are you really suggesting we should overlook that and talk about the quality of the bottom-feeders that they faced? Because that's what you're doing, intentionally or not.


Having said all that, I have no idea whether the committee ignores the computer rankings in favor of ordinal recitations of wins and losses or not. If they do, then maybe you're right about Syracuse having a better resume. They often say that they don't punish teams for playing a tough schedule, though, and it feels like that's what you're doing, especially since the only reason this is even a discussion is because people feel UVa lucked out in its ACC schedule and therefore its amazing ACC record somehow doesn't count.

I'm not suggesting that we penalize teams for playing tougher schedules. I'm suggesting that, when determining #1/#2 seeds, you actually have to WIN against that tough schedule. But the reality is that UVa LOST to all but one of the even remotely tough teams on their non-conference schedule. And then their in-conference schedule was a bit easier than Syracuse's in-conference schedule. I'm penalizing UVa for not getting any wins in those games.

Again, let's look at the actual makeup of that non-conference schedule. Syracuse played one top-10 team (Villanova) and two other top-50 teams (Minnesota (NS) and Baylor (NS)). They won all three. UVa played one top-10 team (Wisconsin), one top-20 team (VCU), and two other top-50 teams (Southern Methodist, and @Tennessee). They lost 3 of 4. Against teams in the 51-100 range, Syracuse had four matchups (@St John's, vs Cal (NS), Indiana, and Eastern Michigan). UVa had just two (@Green Bay, Missouri State (NS). Syracuse won them all. UVa went 1-1.

So again, where is the extra "meat" in UVa's stronger SOS coming from? I certainly don't see a clear edge in toughness at the top of the schedule for UVa. If anything, I'd say Syracuse's non-con schedule against the top-100 was a bit tougher, actually. So it must be coming from the bottom feeders (the teams in the 100+ range and beyond). Is that really a strong way to argue in favor of UVa over Syracuse?

pfrduke
03-03-2014, 11:47 AM
Let me just add that I don't take these brackets very seriously ... but I do think the committee has to respond to the Wichita situation (they HAVE to have a No. 1 seed, yet I do not believe they are one of the nation's top 10 teams) by paring them with the strongest No. 2 possible -- not the weakest (and frankly, I don't think Villanova is one of the nation's 10 toughest teams).

But that would (potentially) violate one of the seeding rules, which is geographic preference to top seeds. Rather than pair #5 overall with #4 overall, the committee pairs #5 overall with its geographically preferred region (even if it's also the geographically preferred region for #1 overall). They then (supposedly) balance it out with the 3 and 4 seeds to try to achieve relative parity among the top 4 seeds in each of the regions. The tidy way to handle this is if Kansas is a 2, since its preferred region and Wichita State's will be the same. But it may not shake out that way, and both Duke and Syracuse have alternative preferences (although if I was the Duke SID office right now, I'd be making sure the committee knows that Indianapolis (MW regional) is closer to home than Memphis (S regional) and may thus be choice #2 after NYC).

Wander
03-03-2014, 11:48 AM
You're essentially overvaluing pre-season schedule by ignoring the results against relevant competition.


Sorry for the nitpick but this has always sort of annoyed me: the only "preseason" UVA had were a couple of closed scrimmages. The non-conference games do - and should - count as much as any other games.



I've said all along that I'd rather be the No. 2 in Wichita State's bracket than No. 1 somewhere else, but to stack No. 1 Wichita and No. 2 Villanova together -- just wow!

Compare that to 1. Kansas, 2 Syracuse in the East; No. 1 Florida, No. 2 Duke in the South and No. 1 Arizona, No. 2 Wisconsin in the West.

Let me just add that I don't take these brackets very seriously ... but I do think the committee has to respond to the Wichita situation (they HAVE to have a No. 1 seed, yet I do not believe they are one of the nation's top 10 teams) by paring them with the strongest No. 2 possible -- not the weakest (and frankly, I don't think Villanova is one of the nation's 10 toughest teams).

There are no metrics that I'm aware of - either human poll, RPI, kenpom, sagarin - that have Villanova as the weakest of those four teams. Actually, kenpom and sagarin have them as the strongest.

pfrduke
03-03-2014, 11:50 AM
Sorry for the nitpick but this has always sort of annoyed me: the only "preseason" UVA had were a couple of closed scrimmages. The non-conference games do - and should - count as much as any other games.



There are no metrics that I'm aware of - either human poll, RPI, kenpom, sagarin - that have Villanova as the weakest of those four teams. Actually, kenpom and sagarin have them as the strongest.

Second strongest, behind Duke (for both).

Wander
03-03-2014, 11:53 AM
Second strongest, behind Duke (for both).

Oops, you're right, thanks - I was thinking Duke was a 1 seed and Kansas was a 2, as kenpom and sagarin have Villanova ahead of Kansas, but not Duke. Point still holds that they're not the weakest of the four 2 seeds by any measure I'm aware of.

TexHawk
03-03-2014, 11:55 AM
The tidy way to handle this is if Kansas is a 2, since its preferred region and Wichita State's will be the same.

Nitting picks, and it doesn't detract from your overall point, but KU's geographic preferred region is Memphis (slightly closer to Lawrence than Indy).

flyingdutchdevil
03-03-2014, 12:04 PM
With one week left of non-tourney ball, the picture is looking pretty clear. Assuming Florida continues to play well, it's nearly guaranteed a 1-seed (even with another loss). Wichita St will probably win out and get a 1-seed. 'Zona found its grove again and will probably get a 1-seed (Thank God. I don't wanna play in the West). And that leaves Duke, UVA, Kansas, 'Nova, and Syracuse hunting for that last 1-seed.

And I think the winner is pretty easy: if Kansas wins out, it's Kansas. If not, then it's whoever wins the ACC Tournament (and isn't named UNC). I don't think 'Nova has a shot unless Kansas, Syracuse, and Duke (and possibly UVA) all lose in the tournament and 'Nova wins.

However, what I'm curious about is the ACC tournament. Because we have 4 teams ranked in the AP top 15, does that mean that more weight will be given than the BIG 10? I assume the answer is yes, and then does the winner get that 1-seed?

Ideally, we'd like Kansas to lose, essentially making the winner of the ACC Tournament the automatic 1-seed.

CDu
03-03-2014, 12:25 PM
With one week left of non-tourney ball, the picture is looking pretty clear. Assuming Florida continues to play well, it's nearly guaranteed a 1-seed (even with another loss). Wichita St will probably win out and get a 1-seed. 'Zona found its grove again and will probably get a 1-seed (Thank God. I don't wanna play in the West). And that leaves Duke, UVA, Kansas, 'Nova, and Syracuse hunting for that last 1-seed.

And I think the winner is pretty easy: if Kansas wins out, it's Kansas. If not, then it's whoever wins the ACC Tournament (and isn't named UNC). I don't think 'Nova has a shot unless Kansas, Syracuse, and Duke (and possibly UVA) all lose in the tournament and 'Nova wins.

However, what I'm curious about is the ACC tournament. Because we have 4 teams ranked in the AP top 15, does that mean that more weight will be given than the BIG 10? I assume the answer is yes, and then does the winner get that 1-seed?

Ideally, we'd like Kansas to lose, essentially making the winner of the ACC Tournament the automatic 1-seed.

People keep omitting Wisconsin, and I'm not sure why. They are 5-2 against the RPI top-25 and 8-3 against the RPI top-50. They are #4 in RPI, #10 in Pomeroy, and #9 in BPI. They have faced the second toughest schedule in the country and are 24-5. And they have a win on the road against UVa. Lunardi has them currently as a #2 seed. Their resume looks every bit as good as ours or Syracuse's and much better than UVa's. If Wisconsin wins their conference tournament, I think they're as much in the mix for a #1 as any of the ACC teams.

Kedsy
03-03-2014, 12:42 PM
So again, where is the extra "meat" in UVa's stronger SOS coming from? I certainly don't see a clear edge in toughness at the top of the schedule for UVa. If anything, I'd say Syracuse's non-con schedule against the top-100 was a bit tougher, actually. So it must be coming from the bottom feeders (the teams in the 100+ range and beyond). Is that really a strong way to argue in favor of UVa over Syracuse?

Well, I suppose if I wanted a strong way to argue in favor of UVa over Syracuse, I could note that Virginia beat Syracuse by 19 just last week.

Putting that aside, though, I'd argue that if major computer systems calculate SOS in such a way that Virginia has a better SOS than Syracuse, then whoever invented the computer system thinks it's a strong way to determine schedule strength. And if most people thought that was off base, they wouldn't put any stock in that computer system -- but the Committee still supposedly relies on the RPI and most people consider Pomeroy to be a top notch rating system. And for me, personally, I don't think I have the ability to parse through a team's schedule and figure out relative schedule strength better than a computer can.

I would note in your favor that Sagarin ranks Syracuse's schedule slightly ahead of Virginia's (#46 to #40) and ESPN's BPI also ranks Syracuse's schedule ahead of Virginia's, but neither of them give Syracuse enough of a schedule edge to move Syracuse's overall rating ahead of Virginia's.

In fact, most published computer rating systems rank Virginia better than Syracuse. And Pomeroy (the system that seems to get the most love here at DBR) thinks Virginia is the 2nd best team in the country vs. Syracuse's #11. So if you want a strong argument, how about that?


P.S.: I agree with you about Wisconsin. They have as good a shot at a #1 as most of the big contenders.

CDu
03-03-2014, 12:48 PM
Well, I suppose if I wanted a strong way to argue in favor of UVa over Syracuse, I could note that Virginia beat Syracuse by 19 just last week.

Putting that aside, though, I'd argue that if major computer systems calculate SOS in such a way that Virginia has a better SOS than Syracuse, then whoever invented the computer system thinks it's a strong way to determine schedule strength. And if most people thought that was off base, they wouldn't put any stock in that computer system -- but the Committee still supposedly relies on the RPI and most people consider Pomeroy to be a top notch rating system. And for me, personally, I don't think I have the ability to parse through a team's schedule and figure out relative schedule strength better than a computer can.

I would note in your favor that Sagarin ranks Syracuse's schedule slightly ahead of Virginia's (#46 to #40) and ESPN's BPI also ranks Syracuse's schedule ahead of Virginia's, but neither of them give Syracuse enough of a schedule edge to move Syracuse's overall rating ahead of Virginia's.

In fact, most published computer rating systems rank Virginia better than Syracuse. And Pomeroy (the system that seems to get the most love here at DBR) thinks Virginia is the 2nd best team in the country vs. Syracuse's #11. So if you want a strong argument, how about that?

P.S.: I agree with you about Wisconsin. They have as good a shot at a #1 as most of the big contenders.

Firstly, you know single-game head-to-heads don't usually determine who gets seeded where. Especially when that single game is a home game for the team that won. And remember that Cuse had a 1-pt lead at the half but played the entire second half with just 5.5 players (do to Grant's injury).

So you don't disagree with the argument that the ratings advantage for UVa seems to be based on their body of work against teams below #100? And you're cool with that? Okay, got it. We'll just have to agree to disagree then.

Kedsy
03-03-2014, 12:51 PM
Firstly, you know single-game head-to-heads don't usually determine who gets seeded where. Especially when that single game is a home game for the team that won. And remember that Cuse had a 1-pt lead at the half but played the entire second half with just 5.5 players (do to Grant's injury).

I do know, which is why I said "[p]utting that aside." However, if your argument consists primarily of parsing single game matchups (which is what your argument seems to me), then it's hard to ignore a 19 point beatdown against the team you're supposedly better than. Especially when it happened a few days ago.


So you don't disagree with the argument that the ratings advantage for UVa seems to be based on their body of work against teams below #100? And you're cool with that? Okay, got it. We'll just have to agree to disagree then.

This is just like our argument earlier in the season about whether Vermont was a top 100 team. Either you trust the computer system or you don't. To me it seems wrong to quote computer rankings when you agree with them and ignore them when you don't.

So in that vein, no, I neither agree nor disagree with your argument on what Virginia's rating "seems to be based on," because it doesn't make sense to me to try to parse it that closely.

Troublemaker
03-03-2014, 01:09 PM
People keep omitting Wisconsin, and I'm not sure why. They are 5-2 against the RPI top-25 and 8-3 against the RPI top-50. They are #4 in RPI, #10 in Pomeroy, and #9 in BPI. They have faced the second toughest schedule in the country and are 24-5. And they have a win on the road against UVa. Lunardi has them currently as a #2 seed. Their resume looks every bit as good as ours or Syracuse's and much better than UVa's. If Wisconsin wins their conference tournament, I think they're as much in the mix for a #1 as any of the ACC teams.

Agreed. Wisconsin is one team that many, including me, have been overlooking. Their non-conference wins over Florida and UVA look terrific right now. If the Badgers can become tournament champions of the Big 10, which is generally considered one of the top 2 conferences in America, that along with their non-conference accomplishments should put them right there with Kansas and the ACC winner for that fourth #1 seed.

CDu
03-03-2014, 01:26 PM
I do know, which is why I said "[p]utting that aside." However, if your argument consists primarily of parsing single game matchups (which is what your argument seems to me), then it's hard to ignore a 19 point beatdown against the team you're supposedly better than. Especially when it happened a few days ago.

Is UVa not better than Tennessee? Clearly they appear to be. Would you argue the opposite based on the Vols' 30 point beatdown of UVa? Obviously not, right?




This is just like our argument earlier in the season about whether Vermont was a top 100 team. Either you trust the computer system or you don't. To me it seems wrong to quote computer rankings when you agree with them and ignore them when you don't.

So in that vein, no, I neither agree nor disagree with your argument on what Virginia's rating "seems to be based on," because it doesn't make sense to me to try to parse it that closely.

And like that earlier argument, I am in favor of trying to dig deeper when the numbers look anomalous, rather than just accepting overall output blindly. I am not saying that UVa's SOS is wrong. I am just noting that the game logs seem to strongly suggest their SOS strength is not based on the top portion of their schedule. At least not as it relates to Syracuse. That is not a critique of the formula but rather an attempt to analyze what is really going on.

I highly doubt that the committee is going to just look at overall RPI/Sagarin/BPI/Pomeroy. There is a reason ESPN posts stuff like record/results vs top-25, vs top-50, etc. And that reason is because people are going to look beyond overall numbers and get down to who you played and who you beat (or lost to). And against the relevant (top-150) part of the schedule, Syracuse seems to have a better resume.

There's a reason that Lunardi, Palm, USAToday, teamrankings, bleacherreport, etc. all have Syracuse as a #1 or #2 seed and UVa as a #3. And it's because they think Syracuse currently has a better body of work than UVa. That could of course change over the next two weeks.

TexHawk
03-03-2014, 01:28 PM
Agreed. Wisconsin is one team that many, including me, have been overlooking. Their non-conference wins over Florida and UVA look terrific right now. If the Badgers can become tournament champions of the Big 10, which is generally considered one of the top 2 conferences in America, that along with their non-conference accomplishments should put them right there with Kansas and the ACC winner for that fourth #1 seed.

Teamrankings, who did well in their predictions last year (I think), has Wiscy as the 4th #1 (http://www.teamrankings.com/ncaa-tournament/bracketology/). Followed by Nova, Cuse, KU, Creighton.

Olympic Fan
03-03-2014, 01:31 PM
There are no metrics that I'm aware of - either human poll, RPI, kenpom, sagarin - that have Villanova as the weakest of those four teams. Actually, kenpom and sagarin have them as the strongest.

Well, Ken Pom -- IMO the best of the metrics -- has Virginia, Duke, Louisville and Creighton ranked ahead of Nova. The RPI has Wisconsin ranked higher. That's not counting the three likely No. 1's -- Arizona, Florida and Wichita State.

I just don't think 'Nova is a particularly strong No. 2 seed ...

SirBlueDevil
03-03-2014, 01:32 PM
Although i'm as giddy as the next blue devil faithful of our new ranking that came out today, before i remotely engage in the current discussion of our possible seeding for the upcomming big dance i'd much rather look towards our team taking care of both wake and unc in the days to come!

Go Duke Blue!

tbyers11
03-03-2014, 02:01 PM
Teamrankings, who did well in their predictions last year (I think), has Wiscy as the 4th #1 (http://www.teamrankings.com/ncaa-tournament/bracketology/). Followed by Nova, Cuse, KU, Creighton.

TexHawk,

I agree with you that TeamRankings has a pretty good methodology for ranking teams. However, they didn't do very well last year in predicting the bracket. They finished 5th from the bottom out of 100+ sites (I lost count on the chart) at Bracket Project (http://bracketmatrix.com/matrix_2013.html). The scoring system gives 3 points for getting a team right, 3 points for getting its seed exactly right and 1 pt for being +/- 1 point on the seed. Team Rankings formula may be good at predicting winners (like KenPom) but that doesn't mean that it can predict a bracket.

They also didn't get all 4 #1's right last year picking Duke as the 4th #1 over Gonzaga.

TexHawk
03-03-2014, 02:19 PM
TexHawk,

I agree with you that TeamRankings has a pretty good methodology for ranking teams. However, they didn't do very well last year in predicting the bracket. They finished 5th from the bottom out of 100+ sites (I lost count on the chart) at Bracket Project (http://bracketmatrix.com/matrix_2013.html). The scoring system gives 3 points for getting a team right, 3 points for getting its seed exactly right and 1 pt for being +/- 1 point on the seed. Team Rankings formula may be good at predicting winners (like KenPom) but that doesn't mean that it can predict a bracket.

They also didn't get all 4 #1's right last year picking Duke as the 4th #1 over Gonzaga.

There I go trusting my memory again. ;) I really only remember watching them closely for the #1 argument, since Duke/KU/Miami were all there. I think I recall them keeping KU as a 1 even after the TCU loss, which I found crazy, but they also got it right. Once that happened, I probably put too much faith in them (and I also forgot the Duke/Gonzaga bit).

I have been following them a bit this year, and one thing I found interesting was their take on WSU. For most of the year, they had them down at a 3 or 4, even when they were 25-0. Then one day, they went from a 1% chance of a #1 to 60%. Maybe they tweaked their algorithm, or maybe the swing in possible outcomes for WSU was that large.

TruBlu
03-03-2014, 02:38 PM
Lunardi just released his latest version this morning:

http://espn.go.com/mens-college-basketball/bracketology


BTW: Joe has Duke and Virginia playing in Raleigh with UNC in San Diego.


California has very strict environmental laws. Would they allow the stench of the dirtiest program in the nation inside their state?

Hoo in Maryland
03-03-2014, 03:02 PM
Many, if not most, of you seem to agree that UVa’s body of work is at least as strong as Duke’s. But there seems to be a number of people out there who are convinced otherwise, apparently because of a few losses UVa had at the beginning of the season. Unquestionably, we lost games at the beginning of the year that I wish we had won.

We have our share of arguably legitimate excuses, including the fact that Perrantes was still being eased in at PG early on, which meant that Brogdon wasn’t getting an opportunity to step up as a scorer, and solidifying those spots has made all the difference in the world. As for the Tennessee game, it simply got away from us after they shot 8 of 11 from three point range in the first half. (I would note that Tennessee, like a number of teams that we have lost to, is a lot stronger than people think (#23 KenPom; same for Green Bay, which is #51 but is among the others receiving votes in the AP poll and lost by 3 to Wisconsin). In fact, it looks like each of our losses will be to tournament-bound teams, which is something that not many teams (including Duke) can say. Finally, for all the talk of our OOC failures, there’s not much mention of our win against #18(!) SMU.

Anyway, below are the rankings, according to Kenpom, for the schools that each school has beaten or lost to (for wins, I stopped at #50). You could obviously choose another ranking service/method, and although it might make a difference for some results, I suspect it all evens out in the end. And what it seems to show is that the two teams are pretty darn comparable, with Duke’s wins being better (primarily because it beat Virginia) and Virginia’s losses being better (primarily because Duke lost to Notre Dame). Unless I have missed something (and it’s possible that I did), it certainly doesn’t show that Duke’s body of work is appreciably better, and at least arguably shows the opposite:

WINS:
Duke has wins over #2(h), 11(h), 13(h), 15(a), 16(h) 36(h)
UVa has wins over #11(h), 15(a), 19(n), 22(h), 36(h), 36(a), 50(a)

LOSSES:
Duke has losses to #1(n), 9(n), 11(a), 22(a), 50(a), 97(a)
Virginia has lost to #3(a), 10(h), 17(h), 23(a), 51(a).

Yes, you beat us (with some help from a friendly crowd and an even friendlier rim). But we have beaten several teams that you lost to, several of them by a landslide (Syracuse, UNC, Notre Dame, as well as a close win @ Clemson).

CDu
03-03-2014, 03:17 PM
I don't think anyone is arguing that the resume for UVa isn't very close to the resume for Duke. I certainly wouldn't do so. I think the teams have nearly identical resumes, actually. Duke has one more "elite" win, but one more "bad" loss. I'd have no issue if UVa got preferential placement over Duke if things were to end today.

Of course, we'll have an opportunity to beat UNC and may get the opportunity to beat two elite teams in the ACC tournament. If we do that, UVa falls certainly behind us in the pecking order.

Dukehky
03-03-2014, 05:28 PM
I don't think anyone is arguing that the resume for UVa isn't very close to the resume for Duke. I certainly wouldn't do so. I think the teams have nearly identical resumes, actually. Duke has one more "elite" win, but one more "bad" loss. I'd have no issue if UVa got preferential placement over Duke if things were to end today.

Of course, we'll have an opportunity to beat UNC and may get the opportunity to beat two elite teams in the ACC tournament. If we do that, UVa falls certainly behind us in the pecking order.

I'd rather have the 3 seed in the Midwest, where we play in Raleigh first two rounds, then move to Indy, where we've had some pretty good results, then face Nova and Wichita State. That seems way better to me than having to play Iowa State and then Florida in the South bracket as the 2 seed, so I'll take that trade.

Of course, I'd rather just win out and get the 1 seed in the East.

This is all according to Lunardi's brackets, and he is wrong a lot.

OldPhiKap
03-03-2014, 06:17 PM
Hoo in Maryland:

First, excellent post. Hope you visit and contribute often.

Second, anyone who has to live in Terpland has my sympathies.

Third, I think a rematch in Greensboro would be a great game. I really like what Bennett has done with your team and like the players. No superstars, but a tight group of talented and smart kids. Hard not to be impressed by your season.

throatybeard
03-03-2014, 07:42 PM
BTW, "Lunardi's bracketology" is redundant. I thought of changing the title to "In this thread, we pretend that what Joe Lunardi thinks actually matters" but some other mod would get mad.

I, for one, fully support this move.

brevity
03-03-2014, 09:02 PM
BTW, "Lunardi's bracketology" is redundant. I thought of changing the title to "In this thread, we pretend that what Joe Lunardi thinks actually matters" but some other mod would get mad.


I, for one, fully support this move.

Too long, and I would know. Consider "Joe Lunardi's bracket racket".

wilson
03-03-2014, 09:30 PM
...when you picture a college kid literally barfing in the gutter, isn't it outside the Sig Ep house at Michigan State or next to a dumpster in State College?I may or may not have barfed in a Duke gutter or two.

Turk
03-03-2014, 09:50 PM
Wait a minute, I beg to differ. I think you guys are holding Lunardi to a double standard. I poked around on Bracket Project a little, and the first three I checked out just listed four teams on each of the sixteen seed lines, without any pairings. Making a list is half the work, and I politely suggest it's the easy half. Lunardi (and anyone else who generates a complete bracket) does the harder half of the work by trying to slot teams into regions and sub-regions according to the selection committee rules (or whichever rules the committee chooses to publicize in any given year).

As we know, a team can be moved up or down off their "true" seed to meet the slotting rules. So according to the Bracket Project scoring system, if you try and fill a bracket you'll lose points every time you have to change a seed to create a valid bracket. If I were Commissioner of Brackets, I would disqualify anyone who just turned in a list of 64 teams. Wait, that's harsh. I would score and track those lists separately in the junior varsity division. We could call it the NIT bracket project.

All Lunardi is trying to do is replicate or predict what the selection committee would do at the current moment in time. Right now, anyone who goes way way way out on a limb and says "Duke is a #2" is adding nothing to the conversation. Finish the job and say which region, which #1 and #3 seed will be in with the Devils, and why.

kAzE
03-04-2014, 01:43 AM
In my opinion, being a 1 seed isn't all that different from being a 2 seed. Either way, you probably won't face a top 10 team until you get to the Elite 8, which is already hard enough. That said, Give Lunardi a break. It's not even his "opinion," it's just his best guess at what the committee would do if the season ended today. He's not supposed to account for who will win the ACC tournament, because he has no way of knowing who's going to win it. If Duke, Syracuse, or Virginia win the ACC tourney, there's strong reason to believe that team will be a #1 seed. I agree with Sagegrouse, if we won out, we would have a better resume than Kansas, even if they won out. We've also been better recently, which I think factors into the decision making process.

Mtn.Devil.91.92.01.10.15
03-04-2014, 05:31 AM
to oversimplify,
reason #1 is kansas rpi = 2, duke rpi = 7 (rpi is a terrible measure, but the committee loves it)
reason #2 is kansas beat duke head to head - if you're looking for an excuse to leap duke over kansas this doesn't help your cause
joe lunardi is not voicing his opinion of how the bracket should be, only how he guesses it would be based on how the committee does things. surprisingly, not everyone is out to get duke.

of course, if duke wins out its resume gets much better - wins over unc, and possibly syracuse and virginia, which is better than what kansas could do - maybe texas and iowa state. so that would change things. i would be surprised if we don't get a #1 if we win out.

I like this oversimplification, and it raises a good point. Is Joe L giving us his version of where the top teams ought to be seeded, or what be believes the committee would do at any given time?

If it is his opinion, then sure let's tear it to shreds. Let's flood him with email about how Duke deserves a #1 seed over Kansas and tell him why.

If these in season brackets are a shot at what the NCAA would do at a moment in time, it becomes harder to criticize. If Joe says "NCAA would put Kansas above Duke because of RPI and head to head" it is sort of a tough thing to fight against. Rather unassailable because his facts are correct, and his conclusion will never be disproven (the NCAA will make their decision in a few weeks based on substantially more information and different facts).

I guess what I am saying is - to a degree, getting upset/excited about Joe's week to week brackets is like a Mighty Mouse v. Superman conversation.

Now... if you gave me info that said something like "within 2 weeks of selection Sunday, Joe L's seedings of the top four lines are 90 percent accurate" I guess I pay closer attention. As it stands, it ia a crappy way to fill in some down time between games,

Turk
03-04-2014, 06:38 AM
I like this oversimplification, and it raises a good point. Is Joe L giving us his version of where the top teams ought to be seeded, or what be believes the committee would do at any given time?



The latter - what the committee would do at a given time. And we've discussed the same data that the committee uses here: SOS, RPI, polls, good wins, bad losses, etc. Guys like Lunardi who have been doing brackets for a while also have some historical knowledge to draw comparisons between current teams and teams with similar profiles in prior years.

Sure, there is a certain level of folly for one human being to try and predict what one or more other human beings will do, but there are enough interesting permutations where it's fun until the real bracket comes out. And as the money guys on Wall St. are fond of reminding us, "past results do not guarantee future performance."

When we get to Selection Sunday, the committee will include at least one and sometimes two bubble teams that "consensus" says has no business being in the bracket, leaving a team that thought they should have been in howling at the world's injustice. One region will be considered "easy", another will be considered "hard", and the other two will make sense to most of us. Which one will Duke be in? We'll know in a couple weeks, and we can have an idea of how teams will move as the games and conference tournaments play themselves out.

P.S. Renaming the thread was unnecessary. I find the animosity toward Lunardi (and Bill Simmons, for that matter) on this board hard to understand.

Mtn.Devil.91.92.01.10.15
03-04-2014, 07:00 AM
The latter - what the committee would do at a given time. And we've discussed the same data that the committee uses here: SOS, RPI, polls, good wins, bad losses, etc. Guys like Lunardi who have been doing brackets for a while also have some historical knowledge to draw comparisons between current teams and teams with similar profiles in prior years.

Sure, there is a certain level of folly for one human being to try and predict what one or more other human beings will do, but there are enough interesting permutations where it's fun until the real bracket comes out. And as the money guys on Wall St. are fond of reminding us, "past results do not guarantee future performance."

When we get to Selection Sunday, the committee will include at least one and sometimes two bubble teams that "consensus" says has no business being in the bracket, leaving a team that thought they should have been in howling at the world's injustice. One region will be considered "easy", another will be considered "hard", and the other two will make sense to most of us. Which one will Duke be in? We'll know in a couple weeks, and we can have an idea of how teams will move as the games and conference tournaments play themselves out.

P.S. Renaming the thread was unnecessary. I find the animosity toward Lunardi (and Bill Simmons, for that matter) on this board hard to understand.

My point was not to denigrate Lunadri's process or results, but rather to inject a little perspective. By the time we get to an hour before official brackets are released, there is near consensus about the teams included, and few surprises regarding seeding. Those of us who follow the season and pay attention to coference tournaments would be nearly as accurate as Joe in predicting the brackets at that point.

My other point is that the earlier brackets are just a mental exercise with little or no relevance to what the brackets look like weeks or months later. In my opinion, it isn't objectionable or worthwhile to analyze these exercises. Clearly, some folks here and elsewhere feel otherwise.

The only part I do feel is interesting as we move through the season is the idea of "bubble watching." As Duke fans, we have had the privledge of following as other teams land inside or outside the predicted group. It must be more excruciating if your beloved squad is riding that edge through the season.

Turk
03-04-2014, 07:16 AM
My point was not to denigrate Lunadri's process or results, but rather to inject a little perspective. By the time we get to an hour before official brackets are released, there is near consensus about the teams included, and few surprises regarding seeding. Those of us who follow the season and pay attention to coference tournaments would be nearly as accurate as Joe in predicting the brackets at that point.

My other point is that the earlier brackets are just a mental exercise with little or no relevance to what the brackets look like weeks or months later. In my opinion, it isn't objectionable or worthwhile to analyze these exercises. Clearly, some folks here and elsewhere feel otherwise.

The only part I do feel is interesting as we move through the season is the idea of "bubble watching." As Duke fans, we have had the privledge of following as other teams land inside or outside the predicted group. It must be more excruciating if your beloved squad is riding that edge through the season.

Agreed. Brackets in January and early February are like baseball batting averages in April and May. I think we need to be halfway through the conference schedule, with teams having about 20 games under their belts, and then we can start paying attention. Seed lines are starting to firm up now, and it's all about which teams can win out, or which teams are sinking like a rock.

Alas, for this year, I am only casually following the bubble. The non-Duke teams that I like are either safely in, or else well out of the hunt.

Nugget
03-04-2014, 10:31 AM
I don't think the bolded part of your quote is right. Kansas is currently #2 in the RPI (with the #1 schedule) and Duke is #7. If both Duke and Kansas win out, I can't see how Kansas' RPI could possibly drop lower than Duke's. The choice in that case could go either way, but since the committee appears to put more stock in the RPI than it does in Pomeroy, I'd probably have to give the edge to Kansas1.

I think the Committee would favor KU over Duke if both win out, but not bc of RPI, which I think tends to matter more to the committee when picking among bubble teams than it does on seeding top teams. Rather, I think what would weigh heavily is that KU would have won the regular season title of the strongest conference.

gumbomoop
03-04-2014, 10:59 AM
Keeping with the thread theme, pretend for a moment that my opinion matters. [If I can pretend, so can you.]

IMO, if Duke's a 1, UK is an 8 [maybe a 5]. If Duke's a 2, UK's a 7. Should Duke slide to a 3, UK will be a 6. Same sub-region.

I realize there will be plenty of tough-enough 6-8 seeds, and that UK's recent play is awful. Still, I wouldn't be happy to see them as a probable/possible opponent for second game. But, again IMO, since a Duke-UK matchup most years would occur at earliest in regional semi, and more likely in EE, this year presents the Selection Committee with an unusual opportunity to set up an early super-marquee contest. IMO, they won't let this opportunity pass.

Wander
03-04-2014, 11:12 AM
I realize there will be plenty of tough-enough 6-8 seeds, and that UK's recent play is awful. Still, I wouldn't be happy to see them as a probable/possible opponent for second game. But, again IMO, since a Duke-UK matchup most years would occur at earliest in regional semi, and more likely in EE, this year presents the Selection Committee with an unusual opportunity to set up an early super-marquee contest. IMO, they won't let this opportunity pass.

Why does anyone believe the selection committee does things like this?

Mtn.Devil.91.92.01.10.15
03-04-2014, 11:19 AM
I think the more likely happenstance is that K and Duke have been so good for so long, you can find storylines anywhere. Fill in the blanks for the following minidramas in potential matchups:
Kentucky
Kansas
Stanford
Michigan
Arizona
VCU
Lehigh
Butler
UConn
Vermont
Harvard
Northwestern

And so on and so on and so on. I'll wager that when brackets come out, I can find 4 storylines for Duke matchups in their region without trying.

flyingdutchdevil
03-04-2014, 11:47 AM
I think the more likely happenstance is that K and Duke have been so good for so long, you can find storylines anywhere. Fill in the blanks for the following minidramas in potential matchups:
Kentucky
Kansas
Stanford
Michigan
Arizona
VCU
Lehigh
Butler
UConn
Vermont
Harvard
Northwestern

And so on and so on and so on. I'll wager that when brackets come out, I can find 4 storylines for Duke matchups in their region without trying.

Mich St, Syracuse, UNLV, SDSU (with Steve Fisher), to add a few more. You're right - there is bond to be a storyline with nearly every team we'll face in the Second Round onwards.

sagegrouse
03-04-2014, 11:57 AM
I think the Committee would favor KU over Duke if both win out, but not bc of RPI, which I think tends to matter more to the committee when picking among bubble teams than it does on seeding top teams. Rather, I think what would weigh heavily is that KU would have won the regular season title of the strongest conference.

"KU would have won the regular season title of the strongest conference."

Lessee, the ACC has teams ranked 4, 5, 7, and 14. The Big 12 (nine members) has teams ranked 8, 16, and 23. Advantage ACC.

Also, let's look at KenPom, wherer the ACC today has teams ranked 2, 3, 11, and 25. The Big 12 has teams ranked 9 and 24. Advantage ACC.

Or the RPI (ESPN version), where the ACC has teams ranked 6, 8, 10, and 18 and the Big 12 has teams ranked 2, 11, 20, and 24. Closer, but maybe a tiny edge to the ACC.

I dunno, Nugget, the ACC looks pretty strong by these measures.

CDu
03-04-2014, 12:18 PM
Mich St, Syracuse, UNLV, SDSU (with Steve Fisher), to add a few more. You're right - there is bond to be a storyline with nearly every team we'll face in the Second Round onwards.

Add Louisville (1986, 2013, Pitino) to the list as well.

TexHawk
03-04-2014, 12:19 PM
"KU would have won the regular season title of the strongest conference."

Lessee, the ACC has teams ranked 4, 5, 7, and 14. The Big 12 (nine members) has teams ranked 8, 16, and 23. Advantage ACC.

Also, let's look at KenPom, wherer the ACC today has teams ranked 2, 3, 11, and 25. The Big 12 has teams ranked 9 and 24. Advantage ACC.

Or the RPI (ESPN version), where the ACC has teams ranked 6, 8, 10, and 18 and the Big 12 has teams ranked 2, 11, 20, and 24. Closer, but maybe a tiny edge to the ACC.

I dunno, Nugget, the ACC looks pretty strong by these measures.

I don't think those are the measures people use when talking about this stuff. There is a pretty decent chance that 70% of the Big12 will be in the NCAA tournament. KU, for example, has TCU as the only awful opponent that they could conceivably relax against (2013 notwithstanding). Texas Tech, the 9th best team in the conference, has beaten OSU at home, OU on the road, and lost on a buzzer beater to KU. There are fewer "nights off" in the Big12, imo.

flyingdutchdevil
03-04-2014, 12:26 PM
I don't think those are the measures people use when talking about this stuff. There is a pretty decent chance that 70% of the Big12 will be in the NCAA tournament. KU, for example, has TCU as the only awful opponent that they could conceivably relax against (2013 notwithstanding). Texas Tech, the 9th best team in the conference, has beaten OSU at home, OU on the road, and lost on a buzzer beater to KU. There are fewer "nights off" in the Big12, imo.

Can't you do that with any team? Notre Dame, the #11 best team in the conference, has beaten Indiana at home, Duke at home, came within one possession to tOSU and UNC.

And if you think that most teams are cupcakes in the ACC, you haven't been watching the ACC.

TexHawk
03-04-2014, 12:49 PM
Can't you do that with any team? Notre Dame, the #11 best team in the conference, has beaten Indiana at home, Duke at home, came within one possession to tOSU and UNC.

And if you think that most teams are cupcakes in the ACC, you haven't been watching the ACC.

I absolutely don't think that *most* teams are cupcakes. But Wake, GTech, Boston College, VTech are all much worse than Texas Tech by Kenpom, for example. (I wouldn't put Notre Dame in that group.) That's 8 games that Duke/Cuse/UVA *should* win, and likely would win, even if their best players are having off nights.

Yes, I'm aware that Cuse lost to BC, but that's an upset on the same level as KU to TCU in 2013.

Kedsy
03-04-2014, 01:01 PM
That's 8 games that Duke/Cuse/UVA *should* win, and likely would win, even if their best players are having off nights.

Not 8 games at all. This season, Duke played only 6 games against the four teams you mention. Syracuse and Virginia played just 5 games against those teams.

TexHawk
03-04-2014, 01:20 PM
Not 8 games at all. This season, Duke played only 6 games against the four teams you mention. Syracuse and Virginia played just 5 games against those teams.

Whoops, completely forgot that. Round-robins are fun, you guys should try it!

hurleyfor3
03-04-2014, 01:23 PM
Whoops, completely forgot that. Round-robins are fun, you guys should try it!

Yeah, yeah, yeah, we know.

You know what's also fun? Having a number in your conference name that's the same as the number of teams. :p

sagegrouse
03-04-2014, 01:24 PM
I don't think those are the measures people use when talking about this stuff. There is a pretty decent chance that 70% of the Big12 will be in the NCAA tournament. KU, for example, has TCU as the only awful opponent that they could conceivably relax against (2013 notwithstanding). Texas Tech, the 9th best team in the conference, has beaten OSU at home, OU on the road, and lost on a buzzer beater to KU. There are fewer "nights off" in the Big12, imo.

Taking a somewhat different tack, the top four teams in the ACC (UVa, Syracuse, UNC, Duke) make this year seem like the late 1980s, when the ACC was at its best. Or like the ACC in the '50s and '60s, when the Big Four ruled the roost and were nationally ranked. And Pitt, Clemson, State, Maryland and FSU can be very tough. The remaining six have pulled some surprises.

InSpades
03-04-2014, 01:27 PM
The ACC averages .753 on KenPom. The B12 averages about .783. It's not a huge difference, but it's not insignificant. The B12 is a really tough league this year, even the teams that are usually bad are pretty good. Obviously not as top heavy as the ACC, but much stronger throughout.

This is what leads Kansas to have such an amazing SoS. They played a good non-conference schedule and then every game outside of TCU is a top 100 game within the conference.

sagegrouse
03-04-2014, 01:37 PM
The ACC averages .753 on KenPom. The B12 averages about .783. It's not a huge difference, but it's not insignificant. The B12 is a really tough league this year, even the teams that are usually bad are pretty good. Obviously not as top heavy as the ACC, but much stronger throughout.

This is what leads Kansas to have such an amazing SoS. They played a good non-conference schedule and then every game outside of TCU is a top 100 game within the conference.

Or, on the other hand, the teams may have progressed (or regressed) significantly since the end of December. No one in the ACC or Big 12 (or any major conference) has played out of conference since early January. The high conference ranking of the Big 12 probably reflects the results in the OOC schedule in November and December, which in some years has been a poor predictor of results in March.

I would argue that UVa and UNC have improved lightyears in the past two months.

CDu
03-04-2014, 01:53 PM
Or, on the other hand, the teams may have progressed (or regressed) significantly since the end of December. No one in the ACC or Big 12 (or any major conference) has played out of conference since early January. The high conference ranking of the Big 12 probably reflects the results in the OOC schedule in November and December, which in some years has been a poor predictor of results in March.

I would argue that UVa and UNC have improved lightyears in the past two months.

Unfortunately, that's simply impossible to prove. Did UVa and UNC get that much better, or did they start facing less impressive competition?

Also, you're not acknowledging that teams in other conferences might have improved by "light-years" as well. For example, Kansas (a VERY young team) was likely playing better in February than in November/December. Of course, now they have Embiid's balky back to deal with, but that's a separate issue.

At the end of the day, we have only the November/December results to compare conferences. Anything outside of that is just speculation.

TexHawk
03-04-2014, 01:54 PM
Or, on the other hand, the teams may have progressed (or regressed) significantly since the end of December. No one in the ACC or Big 12 (or any major conference) has played out of conference since early January. The high conference ranking of the Big 12 probably reflects the results in the OOC schedule in November and December, which in some years has been a poor predictor of results in March.

I would argue that UVa and UNC have improved lightyears in the past two months.

That's fine, and may turn out to be completely valid. But in context of where conferences are rated... Not sure what else we have to go off of on March 4. Iowa State beat Michigan and Iowa out of conference. That little tidbit might mean something, it might not, but it's *something*. The Big12 could go 0-7 in the NCAA tournament and the conference rankings on this date still won't be wrong.

Unless you somehow want to give extra weight to UNC for improving light years, and not give any to someone like K-State, who lost to Georgetown by 30 before Christmas.

CDu
03-04-2014, 01:58 PM
That's fine, and may turn out to be completely valid. But in context of where conferences are rated... Not sure what else we have to go off of on March 4. Iowa State beat Michigan and Iowa out of conference. That little tidbit might mean something, it might not, but it's *something*. The Big12 could go 0-7 in the NCAA tournament and the conference rankings on this date still won't be wrong.

Unless you somehow want to give extra weight to UNC for improving light years, and not give any to someone like K-State, who lost to Georgetown by 30 before Christmas.

I completely agree.

The other thing I'd note is that I'm not entirely convinced that UNC has really improved all that much. They were beating top-25 teams in November/December. Right now, they're mostly just feasting (well, sort of feasting, they are actually squeaking by in a lot of games) on the weaker ACC teams. They have not faced a decent team away from home since they got blown out by UVa in late-January.

I will say that UVa has seemingly improved a lot since their blowout loss to Tennessee. But you could certainly make an argument that any number of teams in all of the conferences are playing better now than in November/December.

Olympic Fan
03-04-2014, 02:10 PM
Just one comment about the idea that Kansas as the No. 2 RPI team will be -- or should be -- seeded ahead of RPI No. 7 Duke.

It doesn't work that way -- witness a year ago, when No. 1 RPI Duke was given a 2-seed. Louisville, the No. 2 RPI seed, got the No. 1 seed in the tournament -- even though Duke beat them head-to-head on a neutral court and had the better RPI and SOS.

I've said it before, the rankings are better predictors for the top seeded teams ... not perfect, but better than RPI. Right now, Duke is the No. 4 team in the rankings -- if the Devils stay in the top four, then have a heck of a chance for a No. 1 seed.

If fact, I'll go so far as to make a bet with anyone who wants it -- if Duke wins out (regular season and tournament), the Blue Devils will absolutely, positively be a No. 1 seed.

If Duke loses one more game -- either this week or in the tournament -- the Devils absolutely, positively will NOT be a No. 2 seed.

Any takers?

flyingdutchdevil
03-04-2014, 02:15 PM
Just one comment about the idea that Kansas as the No. 2 RPI team will be -- or should be -- seeded ahead of RPI No. 7 Duke.

It doesn't work that way -- witness a year ago, when No. 1 RPI Duke was given a 2-seed. Louisville, the No. 2 RPI seed, got the No. 1 seed in the tournament -- even though Duke beat them head-to-head on a neutral court and had the better RPI and SOS.

I've said it before, the rankings are better predictors for the top seeded teams ... not perfect, but better than RPI. Right now, Duke is the No. 4 team in the rankings -- if the Devils stay in the top four, then have a heck of a chance for a No. 1 seed.

If fact, I'll go so far as to make a bet with anyone who wants it -- if Duke wins out (regular season and tournament), the Blue Devils will absolutely, positively be a No. 1 seed.

If Duke loses one more game -- either this week or in the tournament -- the Devils absolutely, positively will NOT be a No. 2 seed.

Any takers?

I will happily take that bet. I assume you mean 1 seed? If so, I'm not taking that bet.

TexHawk
03-04-2014, 02:23 PM
If fact, I'll go so far as to make a bet with anyone who wants it -- if Duke wins out (regular season and tournament), the Blue Devils will absolutely, positively be a No. 1 seed.

Of course, I'm not going to bet on anything do with the selection committee, which seems to pick random reasons to do thing every year. But I do find your hypothetical interesting. Let's say Duke and KU both win out. UVA/Cuse will suffer at least one more loss, so KU would jump them in the polls, to at least 6. They would probably stay there unless Nova loses.

KU would have the higher RPI and be the regular season and tournament champion of the #1 rated RPI conference.
Duke would have a tournament championship, and be two spots ahead of KU in the polls (possibly only 1).

Are you really THAT confident in that scenario?

** Again, I feel the need to mention, these are all hypotheticals. I am not confident in KU's play at all right now, and would not be surprised by a loss at WVU without Embiid on Saturday.

Olympic Fan
03-04-2014, 02:39 PM
Absolutely, I am that confident ... Duke's higher ranking in that scenario will trump Kansas' higher RPI.

(And, mea culpa, flying dutchdevil is right -- I got too wound up. I did mean to say that if Duke loses one more, it will absolutely, positively not be a No. 1 seed.

Okay, I don't think anybody is going to take the second half of my bet -- if Duke suffers another loss, it would take a bizarre and unlikely series of events to make Duke a No. 1. I don't think anybody argues that.

But, Texhawk, do you want to take me up on my first statement -- That if Duke wins out, the Devils will be a No. 1 seed? Not saying anything about Kansas, Wichita State, Florida or Arizona -- I'm saying that if Duke wins out, I'm betting that they are a No. 1 seed.

I think we should play for high stakes -- the loser should make a very public acknowledgement that the winner has a superior knowledge of the NCAA selection process.

(Obviously, the most likely scenario is that Duke loses another game ... in which case any bet would be moot).

So, are you on?

jay
03-04-2014, 03:00 PM
I think what's most entertaining to me this time of year is not just the speculating on seeding, but seeing how much the last week of conference play and conference tournaments can turn that speculation entirely on its head.

I mean, here you have Syracuse, which is in contention for a #1 seed, but seems to me at least very vulnerable right now. They could very well lose at FSU. Combine that with a quick ACCT exit, and you're talking about a 3 seed.

Or Duke....a win over UNC would have at the beginning of the conference season been not that big of a deal, but they're a borderline top-10 team now.

So Duke's pendulum could swing wildly based on how games shake out over the next week and a half. That goes for a lot of teams right now.

Nugget
03-04-2014, 03:16 PM
"KU would have won the regular season title of the strongest conference."

Lessee, the ACC has teams ranked 4, 5, 7, and 14. The Big 12 (nine members) has teams ranked 8, 16, and 23. Advantage ACC.

Also, let's look at KenPom, wherer the ACC today has teams ranked 2, 3, 11, and 25. The Big 12 has teams ranked 9 and 24. Advantage ACC.

Or the RPI (ESPN version), where the ACC has teams ranked 6, 8, 10, and 18 and the Big 12 has teams ranked 2, 11, 20, and 24. Closer, but maybe a tiny edge to the ACC.

I dunno, Nugget, the ACC looks pretty strong by these measures.

I'm looking at the Conference RPI rankings, which has the Big 12 at #1 and the ACC at #5. Counting only the good teams at the top of the league misses the point of how bad the bottom of the ACC is, and how watered down the 15 team ACC is compared to the 10 team (and full round-robin) Big 12.

Troublemaker
03-04-2014, 03:55 PM
But, Texhawk, do you want to take me up on my first statement -- That if Duke wins out, the Devils will be a No. 1 seed? Not saying anything about Kansas, Wichita State, Florida or Arizona -- I'm saying that if Duke wins out, I'm betting that they are a No. 1 seed.

I think that is likely true, but it has to do with the likelihood that if Duke wins out, other teams will have suffered a loss, thus moving Duke ahead of them.

The question seems to be if BOTH Kansas and Duke win out, who would get the 1 seed? Kansas or Duke? I think that's the bet TexHawk and you should take.

Of course, the statistical probability of both teams winning out is probably < 5% since it requires the teams to go a combined 10-0. So if you guys take that bet instead (since it more directly addresses the difference between TexHawk's and Oly Fan's opinions), there's a > 95% chance that the bet won't play and neither of you guys get to win, eat or crow about it.

InSpades
03-04-2014, 04:04 PM
I think that is likely true, but it has to do with the likelihood that if Duke wins out, other teams will have suffered a loss, thus moving Duke ahead of them.

The question seems to be if BOTH Kansas and Duke win out, who would get the 1 seed? Kansas or Duke? I think that's the bet TexHawk and you should take.

Of course, the statistical probability of both teams winning out is probably < 5% since it requires the teams to go a combined 10-0. So if you guys take that bet instead (since it more directly addresses the difference between TexHawk's and Oly Fan's opinions), there's a > 95% chance that the bet won't play and neither of you guys get to win, eat or crow about it.

Of course Duke winning out would likely mean beating UNC, Syracuse and possibly UVA (or UNC x2). This is where the top-heaviness of the ACC will help. Kansas will face an easier road to their championship. Those last couple of top echelon wins could push Duke ahead of Kansas (if Duke is behind Kansas, which isn't a given).

TexHawk
03-04-2014, 04:13 PM
Of course Duke winning out would likely mean beating UNC, Syracuse and possibly UVA (or UNC x2). This is where the top-heaviness of the ACC will help. Kansas will face an easier road to their championship. Those last couple of top echelon wins could push Duke ahead of Kansas (if Duke is behind Kansas, which isn't a given).

The UNC game is definitely tougher than anything left on KU's regular season schedule. But the tournament road isn't overwhelming in favor of Duke, at least in terms of RPI. Of course, there is a lot to come with tournament seeding, but a road that includes Iowa State (#11) and Oklahoma (#18), isn't awful. ISU is worse in RPI than Cuse/UVA by a few spots, but OU is ahead of UNC.

TexHawk
03-04-2014, 04:17 PM
I think that is likely true, but it has to do with the likelihood that if Duke wins out, other teams will have suffered a loss, thus moving Duke ahead of them.

The question seems to be if BOTH Kansas and Duke win out, who would get the 1 seed? Kansas or Duke? I think that's the bet TexHawk and you should take.

Of course, the statistical probability of both teams winning out is probably < 5% since it requires the teams to go a combined 10-0. So if you guys take that bet instead (since it more directly addresses the difference between TexHawk's and Oly Fan's opinions), there's a > 95% chance that the bet won't play and neither of you guys get to win, eat or crow about it.

I will take the bet because I think it's an interesting hypothesis. I have absolutely no confidence in my ability to pick these things, and I wouldn't be overly upset if it shook out that way. I will do what is required, but to be fair I haven't exactly crowed about anything. I just think it's a close call.

But in the interest of fun, why not?

sagegrouse
03-04-2014, 04:21 PM
I think that is likely true, but it has to do with the likelihood that if Duke wins out, other teams will have suffered a loss, thus moving Duke ahead of them.

The question seems to be if BOTH Kansas and Duke win out, who would get the 1 seed? Kansas or Duke? I think that's the bet TexHawk and you should take.
Of course, the statistical probability of both teams winning out is probably < 5% since it requires the teams to go a combined 10-0. So if you guys take that bet instead (since it more directly addresses the difference between TexHawk's and Oly Fan's opinions), there's a > 95% chance that the bet won't play and neither of you guys get to win, eat or crow about it.

Let me humbly suggest that, if Kansas and one of Duke, Virginia or Syracuse win out, there is ample room for Wichita State on the second seed line.

Troublemaker
03-04-2014, 04:26 PM
Let me humbly suggest that, if Kansas and one of Duke, Virginia or Syracuse win out, there is ample room for Wichita State on the second seed line.

Quite true! Or maybe Wisconsin wins out and knocks both KU and Duke to the 2nd line. (I wouldn't think so but who knows what the crazy committee could do.) Double winner and double loser scenarios exist.

InSpades
03-04-2014, 04:27 PM
The UNC game is definitely tougher than anything left on KU's regular season schedule. But the tournament road isn't overwhelming in favor of Duke, at least in terms of RPI. Of course, there is a lot to come with tournament seeding, but a road that includes Iowa State (#11) and Oklahoma (#18), isn't awful. ISU is worse in RPI than Cuse/UVA by a few spots, but OU is ahead of UNC.

Not overwhelming but certainly in favor of Duke. If they play the toughest schedule available to them they will face 2 top 10 RPI teams in the conference tournament. Also with the relative lack of parity in the ACC, they are more likely to face those teams. Throw in the UNC game as a season finale and Duke has a lot more room to impress than Kansas does.

-jk
03-04-2014, 04:51 PM
Or, on the other hand, the teams may have progressed (or regressed) significantly since the end of December. No one in the ACC or Big 12 (or any major conference) has played out of conference since early January. The high conference ranking of the Big 12 probably reflects the results in the OOC schedule in November and December, which in some years has been a poor predictor of results in March.

I would argue that UVa and UNC have improved lightyears in the past two months.

Wouldn't it be cool if the NCAA reserved the last week of Feb for a couple out-of-conference games for everyone? Let's see just how the conferences are doing somewhat later than all the holiday tournaments. An idea to explore...

-jk

Bob Green
03-04-2014, 04:56 PM
Wouldn't it be cool if the NCAA reserved the last week of Feb for a couple out-of-conference games for everyone? Let's see just how the conferences are doing somewhat later than all the holiday tournaments. An idea to explore...

-jk

Basically, Bracket Buster Weekend with the Big Boys instead of the Mid-Majors. Yeah, I agree, that's a great idea to explore. Unfortunately, with conferences getting bigger, there is less opportunity for non-conference scheduling flexibility.

sagegrouse
03-04-2014, 04:57 PM
Wouldn't it be cool if the NCAA reserved the last week of Feb for a couple out-of-conference games for everyone? Let's see just how the conferences are doing somewhat later than all the holiday tournaments. An idea to explore...

-jk

It's about the only workable system to fairly rank the conferences. As you know, I have viewed the job of the Tournament Selection Committee as a "fool's errand," in that all games BETWEEN conferences stop around January 1, making it impossible to fairly rank across conferences in a world where many teams in March are far better (or weaker) than in December.

tommy
03-04-2014, 10:58 PM
Why does anyone believe the selection committee does things like this?

For two reasons. One, because we might be tempted to do that if we sat on the committee. And two, because those kinds of juicy matchups move the meter in terms of TV ratings. The ratings for Duke-Kentucky in the second round would dwarf those of Duke-UMass or Duke-Arizona State.

TexHawk
03-05-2014, 10:51 AM
For two reasons. One, because we might be tempted to do that if we sat on the committee. And two, because those kinds of juicy matchups move the meter in terms of TV ratings. The ratings for Duke-Kentucky in the second round would dwarf those of Duke-UMass or Duke-Arizona State.

I don't think they are immune to them, but at least for the first few rounds, they are forced to work with a bunch of rules about geography, conference affiliations, re-matches, etc. first. If it came down to UK or UMass, and all other rules are satisfied, sure, I could see it.

Are there examples of the committee being swayed by TV ratings? It wouldn't be impossible to look up, but certainly kind of annoying. There are plenty of historical brackets out there. I recall KU being put up against UNC last year (juicy!), but they also had the option of putting Missouri in there instead, another juicy one (maybe not nationally though). UK and Duke played two historic games in the 90s, and both have been to just about every NCAA tournament since then. How many times were they placed in the same region, when they both weren't one seeds? I am sure it is > 0, but that doesn't necessarily prove anything. It may just be how the brackets shook out after all the rules were applied. If it's like 80% of the time, yea, I would buy it.

hurleyfor3
03-05-2014, 10:59 AM
Are there examples of the committee being swayed by TV ratings?

The whole teevee argument... the tournament has already been paid for and cbs' checks don't bounce.


UK and Duke played two historic games in the 90s, and both have been to just about every NCAA tournament since then. How many times were they placed in the same region, when they both weren't one seeds? I am sure it is > 0, but that doesn't necessarily prove anything. It may just be how the brackets shook out after all the rules were applied.

2001 and 2012. There was a potential Duke/Kentucky Sweet 16 matchup in 1994 (Duke a 2; Ky a 3), which I'm glad didn't happen. I agree, it's more a symptom of both teams being good.

Edit: Just remembered 2005, Kentucky was our 2. So since The Greatest College Basketball Game Ever Played, at least five times it could've happened if everything had gone chalk, but only one time it did.

throatybeard
03-05-2014, 02:45 PM
Lunardi now has Duke and Carolina in the same region. Uh, yeah, right.

Mtn.Devil.91.92.01.10.15
03-05-2014, 03:07 PM
Lunardi now has Duke and Carolina in the same region. Uh, yeah, right.

Now see, there's a potential storyline I didn't even bother to list.