PDA

View Full Version : Why/when did Duke MBB switch its identity from defense to offense?



Zeb
11-23-2013, 02:11 PM
In the first half of Coach K's tenure as head coach, most analysis of his teams and coaching style talked about fierce man-to-man pressure defense. But it seems like of late, Duke has become known more for offense. I went to Ken Pomeroy to see if the numbers back this up. He only has data from 2003 on--see the table below.

Observations:
1) We're pretty lucky to be fans of such a consistently successful program. Not sure, but I doubt any other team over past 12 years has a better average overall rank.
2) Duke's average offensive efficiency ranks 12 spots higher than defense, and only 3 of the 12 years has Duke's defense been better than its offense.
3) This switch is really a recent phenomenon. From 2003 to 2008, Duke's D average rank was 9. From 2009 thru today (understanding that 2014's rank is from just 5 games), Duke's average D rank is 37.
4) The correlation between defense rank and overall rank (r=.73) is much stronger than between offense and overall rank (r=.38). (Note: I hope I am doing that right. Statistics gurus are welcome to validate/refute.) So it appears that all other things being equal, Duke would be better served having a stronger defense than offense, but the opposite has been true, especially of late.
5) The final column lists tempo, and I threw it in there just out of curiosity. It seems to have no correlation with overall rank, which makes sense to me but was also interesting to confirm. I didn't put this in the table, but correlation between tempo rank and defense rank was very weak (r=-.12) and pretty weak for offense as well (r=.29).

So, looking at this raises a few questions for me:
1) Is this simply a function of personnel, or has something changed in K's coaching approach that has caused our defensive efficiency to decline?
2) What specific factors account for our defensive efficiency decline? Is it poorer rebounding? Difficulty stopping interior penetration? I don't subscribe to Pomeroy so I don't have access to any more detailed data behind the overall efficiency rankings.
3) Off topic: How the hell did JJ and Shelden not have more post-season success? Those teams were really good!



Year
Rank
AdjO
AdjD
AdjT


2014
8
4
50
70


2013
6
5
31
96


2012
21
10
81
96


2011
4
5
21
51


2010
1
1
8
249


2009
12
8
36
122


2008
10
19
8
16


2007
12
51
7
202


2006
2
4
18
30


2005
3
16
3
93


2004
1
3
4
83


2003
7
10
16
48


Avg
7.3
11.3
23.6
96.3


r

.38
.73
.01

gocanes0506
11-23-2013, 02:23 PM
My opinion would be its harder to get top recruits to buy into coming in and playing big time D. Guys want to score and score in bunches in order to show case themselves for the NBA. Winslow is only one of a handful that I can remember of the past 5 years in the Top 15 whose D was rated higher than their O.

Also, I think K knew the direction of the game. He knew the game was going to go more offensive like it is now.

Kedsy
11-23-2013, 02:51 PM
3) This switch is really a recent phenomenon. From 2003 to 2008, Duke's D average rank was 9. From 2009 thru today (understanding that 2014's rank is from just 5 games), Duke's average D rank is 37.

As was discussed in another thread, the numbers you're using are restated numbers using a brand new Pomeroy formula. Under the old formula, for example, our defense was ranked #8 in 2011, #4 in 2010 and #20 in 2009 (and #70 in 2012, still bad, but better than #81).

If you throw out the outlier 2012, when our defense was just plain bad, and don't count this year yet because that story is a long way from being told, then under the old formula our recent (post-2008) defensive rank has averaged 15 and in the earlier period our average rank was 8. Obviously 15 is worse than 8, but it's not so different as to suggest some sort of tectonic shift.

Is the new Pomeroy formula better or more "accurate" than the old? I have no idea. Why are our recent defensive numbers so different under the new formula but our earlier numbers much less different? Based on Pomeroy's description of the change, I'd assume we've been beating up inferior offenses by a lot and not faring so well against better offenses. So, maybe we're scheduling differently? I don't know. But if the difference you've noted is largely due to a formula change, it's hard to say our team identity is different.

Getting back to your original question, in four of the six seasons making up the earlier period, we had Shelden Williams erasing perimeter mistakes. And if you include 2012 in the five season later period, you're counting a wild outlier that would skew the average big-time in so few seasons. Also, our defensive efficiency in Pomeroy was top 10 last season before Ryan Kelly got hurt and also probably would have been a lot better in 2011 if Kyrie had played more than 11 games. So the difference may just be an anomaly based on freak occurrences and small sample size.

Or it may not. Coach K has always adapted to his personnel; maybe our personnel in recent years has been more offensively inclined. Personally, I doubt your observation truly reflects a change in philosophy or team identity, though.

-jk
11-23-2013, 04:35 PM
I think part of it is how physical defense has become the last ten years or so. It became less about skill and more about raw power.

I do hope they continue calling games closely - the teams with good coaching will adjust.

-jk

theAlaskanBear
11-23-2013, 05:56 PM
I am not sure I would classify it as a long term trend just yet, but it certainly could become one. For me, team experience is a large component of strong defense....quite simply, if this recruiting continues as it has, our players are not going to stay together 3-4 years. This does a couple of things...our lineups have less repetition together, and player conditioning is worse. 3-4 years of college strength and physical training, of game experience banging away, learning what you can and can't get away with re: fouls...this is something in which freshmen and sophs will be deficient.

Look back at great defensive teams, and they will always bee anchored by upperclassmen.

Gthoma2a
11-23-2013, 06:02 PM
I don't know that it is a difference in emphasis rather than a shift in ability/style of defense. The personality of every team is different. With turnover and varying levels of ability between teams/individuals, it is to be expected. I wouldn't even want us to TRY to play the 2010 style with this team.

cato
11-23-2013, 06:11 PM
In the first half of Coach K's tenure as head coach, most analysis of his teams and coaching style talked about fierce man-to-man pressure defense. But it seems like of late, Duke has become known more for offense.

K's greatest team (so far) came during the first half of his tenure. Would you say that Hurley, Hill and Laettner were better know for defense than offense? They got their share of stops and steals, but they dominated at the other end of the court.

I feel the same way about his other great teams ('01, '99).

jimsumner
11-23-2013, 06:43 PM
K's greatest team (so far) came during the first half of his tenure. Would you say that Hurley, Hill and Laettner were better know for defense than offense? They got their share of stops and steals, but they dominated at the other end of the court.

I feel the same way about his other great teams ('01, '99).

In 1991 and 1992, Grant Hill might have been better known for his defense. He barely averaged 11 ppg in 1991. And those teams had Thomas Hill, Brian Davis and Tony Lang, all of whom could defend at a decent level.

Those 1980s FF teams had guys like Mark Alarie, David Henderson, Tommy Amaker, Robert Brickey and Billy King, who could flat out shut you down.

I miss those days.

-jk
11-23-2013, 06:48 PM
In 1991 and 1992, Grant Hill might have been better known for his defense. He barely averaged 11 ppg in 1991. And those teams had Thomas Hill, Brian Davis and Tony Lang, all of whom could defend at a decent level.

Those 1980s FF teams had guys like Mark Alarie, David Henderson, Tommy Amaker, Robert Brickey and Billy King, who could flat out shut you down.

I miss those days.

I miss defense with feet. Looking forward to its comeback.

(Fingers crossed)

-jk

vick
11-23-2013, 08:12 PM
In 1991 and 1992, Grant Hill might have been better known for his defense. He barely averaged 11 ppg in 1991. And those teams had Thomas Hill, Brian Davis and Tony Lang, all of whom could defend at a decent level.

1992 certainly wasn't bad defensively, but I think pretty clearly it was a much better offensive than defensive team. To some degree they often had leads, but even so, there's a reason they gave up something like 1.33 points per possession to a team whose second and third leading scorers were John Pelphrey and Deron Feldhaus. They played at a fast pace, sure, but their results are littered with pretty high scores for opponents--84 for Georgia Tech, 85 for Boston University, 89 for Maryland, 97 for Clemson, 87 for Maryland, etc.

OldPhiKap
11-24-2013, 01:33 PM
I think the advent of the three point line -- and lots of kids who could hit them -- did two things. First, "the best defense is a good offense" -- we got players who could push and score threes, which put constant game pressure on the opponent to keep up. Second, we focused on defending the three more with the idea that 3>2 and we could stay out ahead.

Time was, Danny Ferry was a match-up nightmare because he was one of the first biggish guys who could hit the outside shot. Now, it is much more common. So you have to extend out the defense, allowing for more offensive drives and cuts.

In sum, defense has gotten harder to play and scoring points has become easier. I do not think K has changed his emphasis, just adjusted to a changing game.

jimsumner
11-24-2013, 02:00 PM
1992 certainly wasn't bad defensively, but I think pretty clearly it was a much better offensive than defensive team. To some degree they often had leads, but even so, there's a reason they gave up something like 1.33 points per possession to a team whose second and third leading scorers were John Pelphrey and Deron Feldhaus. They played at a fast pace, sure, but their results are littered with pretty high scores for opponents--84 for Georgia Tech, 85 for Boston University, 89 for Maryland, 97 for Clemson, 87 for Maryland, etc.

The 1992 and 1999 teams were so good, they just toyed with many of the teams on the schedule. They felt like they could put it in cruise control and then hit the accelerator whenever they wanted. And the 1992 team did push it pretty close to the edge on multiple occasions, a game at Clemson and a home game against Maryland being prime examples.

Back to defense. What I really miss are the days in the 1980s when opposing players would quake in their Nikes at the idea of trying to run their offense against a Duke defense that contested every shot, every pass, every dribble.

That may not be possible today. 90-85 sells more tickets and draws more viewers than 60-55 and I understand that. But I do miss the days when Duke's core identity started on the defensive end of things.

OldPhiKap
11-24-2013, 02:37 PM
The 1992 and 1999 teams were so good, they just toyed with many of the teams on the schedule. They felt like they could put it in cruise control and then hit the accelerator whenever they wanted. And the 1992 team did push it pretty close to the edge on multiple occasions, a game at Clemson and a home game against Maryland being prime examples.

Back to defense. What I really miss are the days in the 1980s when opposing players would quake in their Nikes at the idea of trying to run their offense against a Duke defense that contested every shot, every pass, every dribble.

That may not be possible today. 90-85 sells more tickets and draws more viewers than 60-55 and I understand that. But I do miss the days when Duke's core identity started on the defensive end of things.

Kenny Smith once said that the only player he ever had to just turn his back to, in order to just get the ball up court, was Tommy Amaker.

We all remember what Billy King did to Mark Macon in a game we were supposed to lose handily.

Those were fun times. Slap the floor and dig in.