MBB: Ohio State 71, Duke 66 Post-Game Thread

[kshepinthehouse]

Jones went out with an injury early in the Syracuse game at Cameron, going for a loose ball.

I remember that game. Syracuse was having a hard time even bringing the ball up court because of his defense.

[dukelifer]

Oh, from history Duke cannot throw stones. I was there for the infamous Herman Veal game in 1984. Veal was accused of rape prior to the game. The entire student section (which at the time was the entire lower level, including behind the benches), chanted "Raaaa-pist!" at Veal throughout the game, including, of course, when he was at the line. We (yes, including me), changed the "D-U-K-E" cheer that went to each side of the court to "R-A-P-E". People threw women's underwear and condoms at him during warmups and introductions. It was all caught TV, Raycomm and ESPN. We thought it was funny at the time, but it was definitely the worst mass crowd verbal behavior I've ever witnessed. And it ticked off Veal, who scored big points (20? 30?) in the win over Duke. The allegations were later dropped, so Veal was never formally charged. There was a lot of questions if the accuser was pushed to do so, but that's beside the point of this post. Exceedingly bad verbal crowd behavior by Duke.

Duke got chastised by K, Terry Sanford (Duke president at the time, who sent the famous "An Avuncular Letter", since we called him Uncle Terry), the local media, the Washington Post (sure, they were homers, but they were right) and the national media in general. The next home game was against North Carolina. We dressed up as angels, put up signs like "A Hearty Welcome to Dean Smith" and started the now famous chant of "We beg to differ!" instead of "Bullsh-t!". But we lost that game, too. Bad mojo for some time...

9F

That was the Wild West of Cameron. Things have changed in a big way.

[moonpie23]

I'm generally opposed to any negative discussion of refs, because it is what rhe absolute worst fan bases revert to regularly.

.

on the UK board, its the constitution of their fanship...uke had never lost a game fairly, and duke has never won a game fairly...

[Nugget]

Ok, I dug into the data a bit. First, I agree with you that this is not going to be super valuable because of 1) small sample size, as you indicated; 2) the volatility of post season results (ie the Zion team was one bucket away from going to the FF where it had a good chance to cut down the nets, but was also a bucket away from losing to UCF and to VT), and 3) the definition of post season success, which is subjective (ie many programs are happy with sweet sixteen but many Duke fans think anything other than a FF is a let down; also people value consistency differently (would you rather go the EE every year, or miss the tournament one year and go the FF the next?)).

All that said, here is the relative rank of every Duke team's 3p% going back to 1993 (as far back as I could find; data from of sports-reference.com), leaving out 2020 since there was no tournament:

Year	3p%	Rank	Seed    Result
2021	35.2	110	Missed	Missed
2019	30.8	328	1	Elite Eight
2018	37.2	71	2	Elite Eight
2017	37.8	60	2	Second Round Loss
2016	38.5	25	4	Sweet 16
2015	38.7	29	1	Champs
2014	39.5	15	3	First Round Loss
2013	39.9	6	2	Elite Eight
2012	37.1	56	2	First Round Loss
2011	37.4	46	1	Sweet 16
2010	38.5	27	1	Champs
2009	34.9	130	2	Sweet 16
2008	37.7	63	2	Second Round Loss
2007	38.1	49	6	First Round Loss
2006	38.6	31	1	Sweet 16
2005	38	35	1	Sweet 16
2004	36.4	83	1	Final Four
2003	36.3	91	3	Sweet 16
2002	36.3	91	1	Sweet 16
2001	38.5	27	1	Champs
2000	38.3	23	1	Sweet 16
1999	39.6	7	1	Runner Up
1998	36.9	62	1	Elite Eight
1997	38.9	13	2	Second Round Loss
1996	37.6	49	8	First Round Loss
1995	38.1	36	Missed	Missed
1994	36.5	85	2	Runner Up
1993	39.7	27	3	Second Round Loss

As you noted, here are our bottom 10 3pt shooting teams by rank:

Year	3p%	Rank	Seed   Result
1998	36.9	62	1	Elite Eight	
2008	37.7	63	2	First Round Loss
2018	37.2	71	2	Elite Eight
2004	36.4	83	1	Final Four
1994	36.5	85	2	Runner Up
2003	36.3	91	3	Sweet 16
2002	36.3	91	1	Sweet 16
2021	35.2	110	n/a	Missed
2009	34.9	130	2	Sweet 16
2019	30.8	328	1	Elite Eight

In these 10 seasons, we had 2 Final Fours, 5 Elite 8s, 8 Sweet 16s, a second round exit, and a year we missed the tournament.

In contrast, here are our best 10 years according to 3p shooting rank:

Year	3p%	Rank	Seed    Result
2013	39.9	6	2	Elite Eight
1999	39.6	7	1	Runner Up
1997	38.9	13	2	First Round Loss
2014	39.5	15	3	Second Round Loss
2000	38.3	23	1	Sweet 16
2016	38.5	25	4	Sweet 16
2010	38.5	27	1	Champs
2001	38.5	27	1	Champs
1993	39.7	27	3	First Round Loss
2015	38.7	29	1	Champs

In these 10 seasons, we had 3 titles, 4 Final Fours, 5 Elite 8s, 7 Sweet 16s, 2 second round exits, and a first round exit.

Again, small sample size plus small differences in ranks plus volatility of a single elimination tournament, but this would say I'd much rather be a good shooting team come tournament time.

True on the conclusion that you’d prefer to be a better shooting team at tournament time. But what’s more interesting to me is your inclusion of the seed information, which seems to better measure the relative strength of the team than the small sample size/crapshoot nature of tournament results. And your data indicates essentially no difference in the average seeds of our highest-ranked 3 point shooting teams (1.9) and our lowest (1.7 for the 9 teams that made the tournament if you disregard last year or 2.7 if you assign last year’s team a 12 seed designation as one of the last few teams that would have been left out).

What also jumps out is that our 3 point percentages had been very consistently in the 36-39 percent range for the last 20 years, including in the one and done heavy period since 2013-2014, until falling off a cliff with Zion’s team and this year’s team (so far) - not really sure what would explain it.

[Dr. Maturin]

The best way to win the in the tournament is to have a really good team. I'm not sure there's any secret sauce beyond that. The tournament is just so random, and has such a small sample size. I've come to enjoy things more once I realized that the NCAAT is a ton of fun, but it doesn't "prove" anything especially. Every team but one is going to lose, and when that happens people will jump to the "fatal flaw" that they saw all along, etc. It's really hard to win a title, and it usually requires quite a bit of luck.

I do think improved 3-point shooting would make for a more balanced team that is harder to stop. Against Ohio State, it appeared to me that they had better success on D by tightening up on the inside. Instead of driving to kick out for open 3s (only 14 3 point attempts) we often settled for 2 point jumpers or something contested toward the basket. The result was 41% shooting from 2, which is losing basketball. Perhaps a bit better ball movement, and a few more made 3s, can loosen up those defenses.

[phaedrus]

True on the conclusion that you’d prefer to be a better shooting team at tournament time. But what’s more interesting to me is your inclusion of the seed information, which seems to better measure the relative strength of the team than the small sample size/crapshoot nature of tournament results. And your data indicates essentially no difference in the average seeds of our highest-ranked 3 point shooting teams (1.9) and our lowest (1.7 for the 9 teams that made the tournament if you disregard last year or 2.7 if you assign last year’s team a 12 seed designation as one of the last few teams that would have been left out).

What also jumps out is that our 3 point percentages had been very consistently in the 36-39 percent range for the last 20 years, including in the one and done heavy period since 2013-2014, until falling off a cliff with Zion’s team and this year’s team (so far) - not really sure what would explain it.

Pretty clear that we just need to be in the 38.5%-38.7% range. No higher, no lower.

[indy1duke]

Once Mark Williams (or our coaches) realized the OSU offensive game plan called for a low post ISO combined with Mark’s inability to block Zed Key’s shot it was time to switch Mark’s defensive strategy. He should have fronted Key. There would have been no way OSU could have fed the post. Mark would have had to work to get defensive rebounds but it was a strategy worth trying.

[kshepinthehouse]

The best way to win the in the tournament is to have a really good team. I'm not sure there's any secret sauce beyond that. The tournament is just so random, and has such a small sample size. I've come to enjoy things more once I realized that the NCAAT is a ton of fun, but it doesn't "prove" anything especially. Every team but one is going to lose, and when that happens people will jump to the "fatal flaw" that they saw all along, etc. It's really hard to win a title, and it usually requires quite a bit of luck.

I do think improved 3-point shooting would make for a more balanced team that is harder to stop. Against Ohio State, it appeared to me that they had better success on D by tightening up on the inside. Instead of driving to kick out for open 3s (only 14 3 point attempts) we often settled for 2 point jumpers or something contested toward the basket. The result was 41% shooting from 2, which is losing basketball. Perhaps a bit better ball movement, and a few more made 3s, can loosen up those defenses.

Probably need Kedsy's help but I did see a study on what NC winners had in common and I believe it was defense, rebounding, or both. Can't remember

[CDu]

Once Mark Williams (or our coaches) realized the OSU offensive game plan called for a low post ISO combined with Mark’s inability to block Zed Key’s shot it was time to switch Mark’s defensive strategy. He should have fronted Key. There would have been no way OSU could have fed the post. Mark would have had to work to get defensive rebounds but it was a strategy worth trying.

We did do some of that. But the challenge is if Key gets his body in position, it's hard to take it away. Also, fronting only works as largely a direct line-of-pass defense. If you move the ball quickly from one side of the court to the other, the fronting defender is no longer in front of his man. At best he is then beside his man.

Fronting is a perfectly reasonable strategy. But ultimately if a team wants to take fronting away, they can do so. We tried it, and sometimes it worked. But often OSU was patient enough to get Williams out of position.

[Kedsy]

Probably need Kedsy's help but I did see a study on what NC winners had in common and I believe it was defense, rebounding, or both. Can't remember

Just because it's you, kshep.

Here's a table showing champions going back to 2002 with Pomeroy's pre-tournament adjusted defensive rank and offensive rank, along with the champions unadjusted OR% rank and unadjusted DR% rank going back to 2010.

Note that I use pre-tournament numbers because we're talking about predictive value. Obviously a team like 2015 Duke (which ended up #11 in dRank) can have a so-so defense all season and then turn it up in the tournament, but it's pretty hard to predict that.

Year	Champ		dRank	oRank	OR rank	DR rank
2021	Baylor		44	3	8	235
2019	Virginia	5	2	127	47
2018	Villanova	22	1	168	90
2017	UNC		25	4	1	24
2016	Villanova	7	15	247	138
2015	Duke		37	3	36	142
2014	UConn		12	57	220	266
2013	Louisville	1	17	19	193
2012	Kentucky	6	2	23	107
2011	UConn		27	21	9	263
2010	Duke		5	4	8	162
2009	UNC		37	1		
2008	Kansas		3	1		
2007	Florida		14	2		
2006	Florida		18	13		
2005	UNC		6	3		
2004	UConn		10	11		
2003	Syracuse	31	18		
2002	Maryland	12	4

I don't think there's a silver bullet here. The most correlative of the four would seem to be offense (though even with offense, there are some notable exceptions), not defense. And the least correlative of the four would seem to be defensive rebounding.

Frankly, even overall excellence doesn't necessarily lead to a championship. KenPom's pre-tournament #1 team has only won the championship twice in 20 years (2008 Kansas and 2012 Kentucky). In the KenPom era, there have been 28 teams that were both top 10 offense and top 10 defense. Only 5 of them (18%) won the championship (2005 UNC; 2008 Kansas; 2010 Duke; 2012 Kentucky; 2019 Virginia). Of those champions, three of them were top 5 in both offense and defense (2008 Kansas; 2010 Duke; 2019 Virginia), but even that isn't a lock because five such teams did not win a championship (meaning teams in the top 5 in both offense and defense only won 37.5% of the time). For example, in 2002, Duke was #1 in offense and #2 in defense and couldn't get past the Sweet 16. Teams that were top 10 in both but not top 5 in both only won the championship 8.7% of the time. Teams that were top 10 in both but not top 6 in both have never won (at least since KenPom's ratings have existed).

[CDu]

Just because it's you, kshep.

Here's a table showing champions going back to 2002 with Pomeroy's pre-tournament adjusted defensive rank and offensive rank, along with the champions unadjusted OR% rank and unadjusted DR% rank going back to 2010.

Note that I use pre-tournament numbers because we're talking about predictive value. Obviously a team like 2015 Duke (which ended up #11 in dRank) can have a so-so defense all season and then turn it up in the tournament, but it's pretty hard to predict that.

Year	Champ		dRank	oRank	OR rank	DR rank
2021	Baylor		44	3	8	235
2019	Virginia	5	2	127	47
2018	Villanova	22	1	168	90
2017	UNC		25	4	1	24
2016	Villanova	7	15	247	138
2015	Duke		37	3	36	142
2014	UConn		12	57	220	266
2013	Louisville	1	17	19	193
2012	Kentucky	6	2	23	107
2011	UConn		27	21	9	263
2010	Duke		5	4	8	162
2009	UNC		37	1		
2008	Kansas		3	1		
2007	Florida		14	2		
2006	Florida		18	13		
2005	UNC		6	3		
2004	UConn		10	11		
2003	Syracuse	31	18		
2002	Maryland	12	4

I don't think there's a silver bullet here. The most correlative of the four would seem to be offense (though even with offense, there are some notable exceptions), not defense. And the least correlative of the four would seem to be defensive rebounding.

Frankly, even overall excellence doesn't necessarily lead to a championship. KenPom's pre-tournament #1 team has only won the championship twice in 20 years (2008 Kansas and 2012 Kentucky). In the KenPom era, there have been 28 teams that were both top 10 offense and top 10 defense. Only 5 of them (18%) won the championship (2005 UNC; 2008 Kansas; 2010 Duke; 2012 Kentucky; 2019 Virginia). Of those champions, three of them were top 5 in both offense and defense (2008 Kansas; 2010 Duke; 2019 Virginia), but even that isn't a lock because five such teams did not win a championship (meaning teams in the top 5 in both offense and defense only won 37.5% of the time). For example, in 2002, Duke was #1 in offense and #2 in defense and couldn't get past the Sweet 16. Teams that were top 10 in both but not top 5 in both only won the championship 8.7% of the time. Teams that were top 10 in both but not top 6 in both have never won (at least since KenPom's ratings have existed).

Which, of course, makes sense. In a 6-game, single-elimination tournament, there is just too much noise to predict the winner with any confidence. And no way to derive a meaningful trend among winners. Especially as the construct of the game changes enough over time that you can't even try to make up for the noise with a large enough sample (not that we'd be anywhere close to a large enough sample even with the ~35 data points of champions).

I mean, you don't even have to be particularly great to win a championship. See UConn twice in the 2010s, Villanova in the mid-80s. Usually it's a really good to great team that wins the title. But not always. There just isn't a magic formula for winning the title, other than playing well enough over six games in March/April. There isn't a magic formula for winning, just like there isn't a magic formula for losing other than having a bad day at the wrong time.

[Kedsy]

Which, of course, makes sense. In a 6-game, single-elimination tournament, there is just too much noise to predict the winner with any confidence. And no way to derive a meaningful trend among winners. Especially as the construct of the game changes enough over time that you can't even try to make up for the noise with a large enough sample (not that we'd be anywhere close to a large enough sample even with the ~35 data points of champions).

I mean, you don't even have to be particularly great to win a championship. See UConn twice in the 2010s, Villanova in the mid-80s. Usually it's a really good to great team that wins the title. But not always. There just isn't a magic formula for winning the title, other than playing well enough over six games in March/April. There isn't a magic formula for winning, just like there isn't a magic formula for losing other than having a bad day at the wrong time.

Exactly. And yet a lot of people spend a lot of time suggesting requirements for winning and weaknesses that will preclude winning. Ultimately there are no such things, either way. You don't need a "true" PG, you don't need lights-out shooters, you don't need a lockdown defense. All those things are nice to have, but none are necessary to win a championship.

[dukelifer]

Which, of course, makes sense. In a 6-game, single-elimination tournament, there is just too much noise to predict the winner with any confidence. And no way to derive a meaningful trend among winners. Especially as the construct of the game changes enough over time that you can't even try to make up for the noise with a large enough sample (not that we'd be anywhere close to a large enough sample even with the ~35 data points of champions).

I mean, you don't even have to be particularly great to win a championship. See UConn twice in the 2010s, Villanova in the mid-80s. Usually it's a really good to great team that wins the title. But not always. There just isn't a magic formula for winning the title, other than playing well enough over six games in March/April. There isn't a magic formula for winning, just like there isn't a magic formula for losing other than having a bad day at the wrong time.

A key part of the formula is to have your players healthy at the end of a season. It starts from there.

[kshepinthehouse]

A key part of the formula is to have your players healthy at the end of a season. It starts from there.

Actually starts with making the tournament

[CDu]

A key part of the formula is to have your players healthy at the end of a season. It starts from there.

And even that isn’t 100% critical, as Louisville showed in 2013. And to a lesser degree Duke showed in 2001 (though Boozer was at least mostly healthy by the Final Four).

[uh_no]

And even that isn’t 100% critical, as Louisville showed in 2013. And to a lesser degree Duke showed in 2001 (though Boozer was at least mostly healthy by the Final Four).

While I agree ultimately, Kevin Ware wasn't more than a role player. He was their 8th option. So for 2 games, not hard to cope with losing that. Maybe there is a potential situation where you'd need him, and fortunately they didn't run into it...but it would be a far cry from moore or banchero getting hurt for us this year.

[CDu]

While I agree ultimately, Kevin Ware wasn't more than a role player. He was their 8th option. So for 2 games, not hard to cope with losing that. Maybe there is a potential situation where you'd need him, and fortunately they didn't run into it...but it would be a far cry from moore or banchero getting hurt for us this year.

Oh by no means was he a star. But he average over 16 mpg that year, and was their only reserve guard. So Louisville was most certainly not fully healthy going into the Final Four. Fortunately for them, they were a really loaded team and could withstand the loss of a primary reserve.

[MChambers]

And even that isn’t 100% critical, as Louisville showed in 2013. And to a lesser degree Duke showed in 2001 (though Boozer was at least mostly healthy by the Final Four).

Brian Davis had a badly sprained ankle, suffered in the semi final against Indiana. He gutted out 15 or 20 minutes against Michigan, but he was severely hobbled. Luckily, Duke was a far better team than Michigan and was able to win without a key starter.

[uh_no]

Brian Davis had a badly sprained ankle, suffered in the semi final against Indiana. He gutted out 15 or 20 minutes against Michigan, but he was severely hobbled. Luckily, Duke was a far better team than Michigan and was able to win without a key starter.

while it brings up bad memories for most, okafor was dealing with back issues for most of the season, and it was questionable most game days whether he would have issues. he got plenty of rest in the Duke game.

[robed deity]

while it brings up bad memories for most, okafor was dealing with back issues for most of the season, and it was questionable most game days whether he would have issues. he got plenty of rest in the Duke game.

Ugh, in the most poorly officiated game I have ever seen