Fast Breaks....this year versus last year

[Duke76]

It just dawned on me that one of the biggest differences this year vs last are the number of fast breaks per game we have witnessed....Haven't looked at the numbers but intuitively it feels like night and day...Top 5 highlights on each game seem to include many more. Mason is getting out on the break at an alarmingly positive rate, Cook and the other guards are pushing much more...we are finishing at a high percentage...I'd say this a major reason we have gotten off to such a great start and can be a real differentiator in a drive to another championship.

Our more quantitative posters hopefully will see this as a worthy topic for rumination and fill in my many holes of analysis.

[Native]

It just dawned on me that one of the biggest differences this year vs last are the number of fast breaks per game we have witnessed....Haven't looked at the numbers but intuitively it feels like night and day...Top 5 highlights on each game seem to include many more. Mason is getting out on the break at an alarmingly positive rate, Cook and the other guards are pushing much more...we are finishing at a high percentage...I'd say this a major reason we have gotten off to such a great start and can be a real differentiator in a drive to another championship.

Our more quantitative posters hopefully will see this as a worthy topic for rumination and fill in my many holes of analysis.

I think the big difference here is twofold: the improvement of our perimeter defense and the improvement of our guards, particularly Sheed and Cook.

More steals and disruption happening on the defensive end. We now we have a full complement of guards (Cook and Sheed, especially) who know how to keep their heads up, push the tempo, control the ball in a fast break and look for easy passes to players like Mason, who runs the floor better than any other big man in college basketball IMO. Other guys are following the example and it's a fun team to watch.

Anybody have any stats on our ability to turn over the ball from last season to this year? That's what's night and day - our defense is miles better than last year and it's showing in all aspects of the game.

I'd also like to take this opportunity to remind the board at large that Carolina is an awful basketball team. Just thought I'd throw that in there. Go Duke!

[Bob Green]

I agree with Native the difference is defense. Fast break opportunities are created by turnovers and rebounds. In the 3-point era, long rebounds off missed bonusphere shots really create an opportunity.

[pfrduke]

It just dawned on me that one of the biggest differences this year vs last are the number of fast breaks per game we have witnessed....Haven't looked at the numbers but intuitively it feels like night and day...Top 5 highlights on each game seem to include many more. Mason is getting out on the break at an alarmingly positive rate, Cook and the other guards are pushing much more...we are finishing at a high percentage...I'd say this a major reason we have gotten off to such a great start and can be a real differentiator in a drive to another championship.

Our more quantitative posters hopefully will see this as a worthy topic for rumination and fill in my many holes of analysis.

This is certainly an incomplete measure, since it doesn't take into consideration opportunities, conversion rate, etc., but per the box scores, we have 69 fast break points through 8 games this season compared to 42 through 8 games* last season. So that's certainly consistent with your impressions. Interestingly, we've done better on the break against top competition - 12 against Kentucky, 13 against OSU, and 17 against Louisville, in contrast to 6 against Georgia State and 0 against Gulf-Coast.

*the box score for last year's Belmont game does not list fast break points, so the 42 points is from games 2-9 (Presbyterian through Colorado State).

[wilko]

Another thought to add in to the mix on the topic of an improved D..

How many Charges have we actually taken this year? (I don't know, I'm asking)
Sure doesn't seem like very many...

I think that the Perimeter D is good at staying in front of people so we dont HAVE to do that so much this year.
A comparison of 8 games in this yr vs: past yrs would be interesting...

[OZZIE4DUKE]

Another thought to add in to the mix on the topic of an improved D..

How many Charges have we actually taken this year? (I don't know, I'm asking)
Sure doesn't seem like very many...

I think that the Perimeter D is good at staying in front of people so we dont HAVE to do that so much this year.
A comparison of 8 games in this yr vs: past yrs would be interesting...

I think the refs are calling fewer charges this year - the flopping thing and all... Except when they call one of our players for a charge, and then the "other old" rules apply. :cool: Just ask devildeac, he'll tell ya!

[bedeviled]

...we have 69 fast break points through 8 games this season compared to 42 through 8 games* last season.
Ha! If I read that right, it equates to an additional 3.375 fast-break pts per game...or, greater than 1 additional fast-break basket per game than last year!! ;) Thus, it does not appear to be a big difference maker.

For the 1st 8 games of 2011 & 2012, our steals are 6.225 -> 6.875, and our turnovers forced are 14.75 -> 14.5. This doesn't seem very significant either. Looking at our defensive efficiency, defensive rebound %, and tempo so far this year, I do not suspect fast-breaks are considerably more frequent though I welcome dissenting opinions. My own suspicion is that the visual difference is produced by secondary break and early offense opportunities. We have played some teams with high defensive pressure which, when overcome, leads to the appearance of a break. At any rate, I would love if someone was willing to break down our secondary break schemes from this year vs last year. My gut is that they are similar as, in recollection, it seems Austin's unique role was predominantly in the half-court set.

This season, I have frequently wondered if we are playing with a different mentality of pacing or if our flow has simply been dictated by the teams we've played. Cook's direct comparison stats against opposing PGs is certainly favorable. However, I personally think Siva, Smith, and Craft outplayed him in directing the games, creating, and isolating mismatches....veteran qualities of the floor general. I don't think we really have an offensive identity yet as the style of games have largely been dictated by our opponents. It is remarkable and exciting that we are good enough to handle that...and kind of cool that the roles and MOTM switches frequently! I am looking forward to Cook (still quite young) growing during this part of the season when he probably won't be spurred on by the grit, heat, and rush that lit him up to date. I have thought that game awareness would flow through Seth or Ryan, but I'll be a happy man if Cook becomes, not just our team leader, but the floor leader.

[dbowen]

i'd also like to take this opportunity to remind the board at large that carolina is an awful basketball team. Just thought i'd throw that in there. Go duke!

i concur!

[Wander]

Ha! If I read that right, it equates to an additional 3.375 fast-break pts per game...or, greater than 1 additional fast-break basket per game than last year!! ;) Thus, it does not appear to be a big difference maker.


It's easy to underestimate the statistical significance of what appears to be small number in these kinds of contexts. I actually take the opposite conclusion: that 3.4 points per game in fast breaks is actually really significant. Consider that last season, Duke had seven games where the margin at the end of regulation was less than 3.4 points.

[MChambers]

It's easy to underestimate the statistical significance of what appears to be small number in these kinds of contexts. I actually take the opposite conclusion: that 3.4 points per game in fast breaks is actually really significant. Consider that last season, Duke had seven games where the margin at the end of regulation was less than 3.4 points.
The other thing to consider how tough Duke's opponents have been this year. It's a lot harder to get fast breaks against good teams, so I view the difference as significant.

[Oriole Way]

The #1 reason, far and away, is Quinn Cook is actually getting playing time. He was the only guard on our entire roster last season capable of running fast breaks, but he didn't get substantial playing time. Sulaimon is the second big reason. He also helps generate fast breaks because he is a great defender and can handle the ball in transition.

Last year we had Rivers, Thornton, and Curry handling the ball most of the time, at least in terms of initiating the offense. After forced turnovers or long defensive rebounds, a vast majority of the time they chose to slow down the tempo, deciding against fast breaking (or starting a secondary fast break), and instead they almost always set up the half court offense. Several of the few fast breaks we did have did not result in converted field goals because Thornton and Rivers simply weren't capable passers in the transition. Rivers in particular turned the ball over quite a few times and had poor judgment and average court vision at best. He frequently held onto the ball too long when he should have passed to an open teammate, frequently leading to an offensive charge, a poor shot that didn't fall, or a bad pass/turnover after turning back from the rim.

I honestly remember just a handful of successful fast breaks last year, and Cook led most of them. Since he's starting and getting big minutes this year, we've been able to generate a lot more breaks, and we've been able to convert on a majority of them thanks to his passing ability and solid IQ running an offense (in addition to the great play of Sulaimon starting and finishing breaks).

[TruBlu]

In addition to the fast breaks mentioned, our ball movement seems crisper, with less individual dribbling and more passing.

Through 8 games, against a somewhat comparable schedule (remember we had a tough schedule early last year also), here are the assists averages/game:

Assists last year - 11.6/game

Assists this year - 15.6/game


I haven't looked up the stats (and don't know how), but another general impression for me is that we seem to be getting better shots, earlier in the shot clock. Last year it seemed that we had many more "hurried" shots taken at or near the end of the shot clock.

[bedeviled]

3.4 points per game in fast breaks is actually really significant. Consider that last season, Duke had seven games where the margin at the end of regulation was less than 3.4 points.

Good point. I agree with the importance of points themselves. To factor that in, I looked back at the total number of points for 8 games each season (using the games designated by pfrduke). We scored 629 and 630 in 2011 and 2012, respectively. Thus, it is not like we are getting an extra 3.4 points per game; the points are just coming from somewhere else. Is the shift in fast-break vs somewhere else significant? I don't think so as it is 1-2 baskets per game. It's less than 3% of our possessions (Our tempo was 68.5 in 2011 (8games) and 69.25 in 2012).

It would be fun to be run-and-gun, and we may very well eventually be that team. Right now, though, I don't think the numbers support a change to a fast-breaking team being "a big difference" between 2011 and 2012. After all, with essentially the same number of steals, forced turnovers, tempo, and a not-so-glorious rate of defensive rebounds, it's hard for me to justify where an increase in fast breaks would fit in (so pfrduke's #s make sense to me). As I stated, I suspect the pace and breaks we see visually are predominantly in secondary and early offense when we have broken the pressure defenses we've faced.


The other thing to consider how tough Duke's opponents have been this year. It's a lot harder to get fast breaks against good teams, so I view the difference as significant. I quoted this, just to "like" it as a nervous fan. Of course, it is hard to score by any means against the teams we've faced. Thus scoring 2 easy lay-ups is soooo much nicer than having to fight for those buckets in half-court.

[Duvall]

http://www.youtube.com/watch?v=FFRoQpv43LA&feature=g-u

[Duke76]

http://www.youtube.com/watch?v=FFRoQpv43LA&feature=g-u

Thanks for the Collins video and it appears there is some validity to what most of us are seeing.....fast breaks also tire the other team out leading to pressure and turnovers...another added benefit...wonder what the opponents turnover average is this yr compared to last? It all kinda works together....bottomline I am stoked about the quality of our team this yr

[Duke76]

In addition to the fast breaks mentioned, our ball movement seems crisper, with less individual dribbling and more passing.

Through 8 games, against a somewhat comparable schedule (remember we had a tough schedule early last year also), here are the assists averages/game:

Assists last year - 11.6/game

Assists this year - 15.6/game


I haven't looked up the stats (and don't know how), but another general impression for me is that we seem to be getting better shots, earlier in the shot clock. Last year it seemed that we had many more "hurried" shots taken at or near the end of the shot clock.

Agree with all of the above as well...basketball is some much fun to watch when played the way the greats have always taught and played it

[roywhite]

From Duke Blue Planet and Coach Chris Collins:

Blue Print: Transition Offense (http://www.youtube.com/watch?v=FFRoQpv43LA&feature=g-u)

Good stuff, and a key area of improvement.

[gep]

I think the refs are calling fewer charges this year - the flopping thing and all... Except when they call one of our players for a charge, and then the "other old" rules apply. :cool: Just ask devildeac, he'll tell ya!

I don't think I've seen much of Duke's trying to "take the charge"... is that accurate? But, I do remember seeing a lot of Duke "taking the charge" and winning the call... :cool: Seems like they are "picking their spots" ...

[bedeviled]

How many Charges have we actually taken this year?

The best I can piece together from play-by-play is that we averaged drawing 2 offensive fouls (not necessarily charges) per game in the first 8 games last year, 2 per game in the last 8 games last year, and 2.25 per game so far this year. It's shoddy data with no context, but no one answered with bettter stuff :D

[bedeviled]

Thanks for the C.Collins video and response. Now I know why I felt like a jerk trying to figure out why the fast break stats didn’t reflect the real-time visual experience…it’s those darn semantics!! My understanding is that the majority of the clips in the video are Duke “getting out on the break” or “getting out in transition” but are not “fast break” plays. Thus, they would not show up in the official fast break stats. Nit-picking, I know. The bottom line is that Duke is pushing the ball in transition. (My error, though, was more than semantics. I had suspected our visual speed was due to breaking pressure, but the clips provided are from possessions where the pressure was never set).

If you’re bored by the semantics, I suggest you stop reading this post now ;) But, I will continue on in case other people share my interests and can help me learn more about game analysis. I’m just a casual/lay fan, so I’d really appreciate if someone corrects me if I’m wrong (Please imagine that every sentence starts with “from what I understand”)! Here’s my conception of transition offense…

FAST BREAK: My impression is that a fast break (fb) is when the O is out in front of the D, creating a man-advantage. Given that we have not played lazy defenses this year, it would be hard to accomplish an increase in fbs without an increase in steals/turnovers. Unless, of course, we turned into Loyola Marymount and started shooting outlet passes to sprinting lanes. I think the relative emphasis in fast break offense is on the sprinter's path so that the gunner knows where to fire the outlet and keep the man-advantage toward the basket.

I spent too much time in vain this morning trying to find the official criteria for a fast break (so, if there is a score keeper in the house, please speak up). I know it must be super easy to find, but I couldn’t do it. The NCAA basketball rulebook (http://www.wbca.org/includes/media/docs/2010-11WBBPlayRulesHandbook.pdf) refers to “establishing a position in any outnumbering fast break situation” when referring to blocking fouls, but that’s all I found. For the NBA, I found a Milwaukee Journal Sentinel link (http://news.google.com/newspapers?nid=1683&dat=20051202&id=onY0AAAAIBAJ&sjid=Po4EAAAAIBAJ&pg=4696,1775977) that states the following:
“The fast-break points that appear in the official box score are determined by a cut-and-dried definition that is used by the league’s scorekeepers. The Bucks, though, chart fast-break points in a different way, using a bit more leeway and interpretation to determine what qualifies as fast-break points. The Bucks also take into consideration points that are scored off the “secondary” break, when there is a pass or two involved before the basket is score."
This quote suggests that the NBA has additional criteria besides man-advantage.

SECONDARY BREAK: If no fb, or “primary break,” is available, an O transitions into the “secondary break.” Here, they are attacking a defense that is present but not set. Obviously, this can have a different strategy than a 2-on-1 or 3-on-2 scenario though they are often of the same flow. Also, a secondary break (sb) is more likely to include all 5 O team members making their way downcourt. Many sb begin with putting the ball in the PG’s hands to initiate O. The emphasis for the runners becomes more weighted toward the spots they are to occupy. This is seen in Florida’s sb (http://www.youtube.com/watch?feature=player_embedded&v=8wt_JQc_oFs) where Noah is more intent on getting to his block than looking for the pass. The Chris Collins video details Duke’s initial spots. Unfortunately, for us armchair PGs, he only shows easy baskets without breaking down how our set evolves from there. (Here’s a bonus link (http://www.youtube.com/watch?v=-IgWLWdO3cI) to videos of Chris running transition drills. I had heard rumor that Izzo incorporated helmets & pads in practices, and it turns out we use blocking pads, too!)

EARLY OFFENSE: Secondary break systems very often include “quick hitter” or “early offense” plays while the lumbering giants make their way downcourt and/or the defense is trying to get its bearings. There is lots of stuff on how UNC’s transition offense (http://www.breakthroughbasketball.com/offense/fast-break-offense-carolina.html) evolves from fast break to early offense online. Finally, the O transitions fully from the early offense into its half-court set.

Sorry about the length. I hope the content is valid.

[bedeviled]

As soon as I considered it, I knew my last post was going to be laborious. So, before I posted, I did a little homework to thank everyone for putting up with my junk. This post is my apology gift.

TruBlu relayed the impression that “we seem to be getting better shots, earlier in the shot clock. Last year it seemed that we had many more "hurried" shots taken at or near the end of the shot clock.” To get more of a sense of our transition game, I went through the play-by-play of the first 8 games of 2011 and 2012 to determine the shot-clock use during transition play.

I did not count plays where the defense could get set prior to the offense initiation - fouls, OOB (except for made baskets), subs.
I did not include the last 5 minutes of the game where subs & stall tactics cloud the picture.
I did not include offense after an offensive rebound.

Caveats:

The data was hand gathered and calculated…I’d be shocked if it was error-free.
The data does not include fouls/free throws…it doesn’t say anything about FGs missed due to a foul or additional point from FTs.
The data does not give a full picture of transition efficiency as possessions/turnovers/OR are not included. Unfortunately, it involved too much interpretation (and work!) of the raw play-by-play to include lots of variables.

Hopefully these flaws cancel each other out when comparing values. Anyway, take it for what it is worth….a crude way to look at when we shot in transition and if they were quality shots.



Year
Time off clock
FG
FGA
3P
3PA


2011
0-7 sec
13
24
2
7


2011
8-14 sec
38
70
11
22


2011
15-21 sec
43
87
14
31


2011
22-28 sec
15
36
4
11


2011
29-35 sec
13
22
4
7


2011
total
122
239
35
78


2012
0-7 sec
27
50
5
16


2012
8-14 sec
34
75
13
31


2012
15-21 sec
31
74
7
22


2012
22-28 sec
20
43
6
12


2012
29-35 sec
4
12
2
7


2012
total
116
254
33
88



300130023003

[dragoneye776]

Wow this is great data. Thanks for putting all of this together. So the evidence does support that we are taking shots earlier in the shot clock this year. However, i find it an odd anomaly that in 2011, would field goal percentage late in the shot clock was so much higher than it was this year. Could you specify on what that was from? Perhaps, when our plays were well defended and we just had to chuck up a 3, Austin and Andre were better at that than our shooters from this year?

[toooskies]

Small sample size.

[bedeviled]

... field goal percentage late in the shot clock was so much higher than it was this year. Could you specify on what that was from?
Short answer: No. ;)
1. The play-by-play is schematic enough, but this is worsened because I was trying to plow through the numbers and didn’t record “extra” info. All I have left is scattered numbers and spinning eyes.
2. toooskies is right - poor sample size.

Late shot-clock eFG% due to skilled 3P specialists?
If you look at the table, there is only a small sample size for the last part of the shot clock. Our 3P shots in 2011 and 2012 were 4/7 and 2/7, respectively. I don’t think too much can be deduced from those numbers…especially because one of the 4/7 was Thornton’s Kansas prayer. With the equal, low sample size, I don’t think this subset of data makes any statement about which year had better marksmen.

Late shot-clock eFG% due to clutch 3P specialists?
Well, perhaps our 2011 snipers were just cool under pressure due to their comfort level with the dwindling shot clock. The low number of shots in this stage suggests that this may not be the case (Also, the distribution of attempts across the seconds of the final stage was not significantly different between years). Actually, one of the surprising things to me is that, in 2011, our shots are still weighted toward the earlier end of the shot clock. We apparently were not holding the ball forever; we just weren’t pushing the early transition.

Late shot-clock eFG% simply a trend of better eFG% overall?
Another surprising thing to me is the eFG% throughout the possession last year. It suggests that we are not getting better looks than last year…or at least not finishing them as well.
Good throughout the possession and low number of late possessions aside, though, you are right - we shot exceptionally well at the end of the shot clock last year…including 2Ps which has a greater sample size. I don’t know what to tell you; I don’t think the answer is in this data.



End of Shot-Clock FG%


Year
2P%
3P%


2011
0.6
0.571429


2012
0.4
0.285714



So, you can’t tell me anything?
Hmmmm. Well, maybe there’s something of interest to you. Late in the possession in 2011, we got more points from 2Ps than from 3Ps as we took more than double the number of 2Ps than 3Ps. In fact, this was a pretty consistent across our possession last year.



Shot selection


Time off clock
3P/FG%


0-7 sec
0.291667


8-14 sec
0.314286


15-21 sec
0.356322


22-28 sec
0.305556


29-35 sec
0.318182


average
0.32636


0-7 sec
0.32


8-14 sec
0.413333


15-21 sec
0.297297


22-28 sec
0.27907


29-35 sec
0.583333


average
0.346457



It appears that our basic shot selection was stable throughout the possession in 2011. Perhaps our offense continued as usual in the late seconds, allowing those late shots to come within our normal rhythm (whatever that may have been, lol)? IDK.

Maybe it would be interesting to see what in the 2012 transition offense leads to our taking a considerably greater percentage of 3s in the 8-14 second range. And, yes, maybe we are just chunking up desperation junk at the end of the shot clock this year.