Finding the root of UK's rebounding woes
Kentucky picked a lousy time to have their worst defensive rebounding game of the season last Saturday, as it no doubt contributed to their loss to Auburn. It wasn't entirely unexpected, as Auburn rates as the best offensive rebounding team UK has played this season by several metrics. As I'll demonstrate, the extent to which UK fell apart on the boards is surprising. The timing couldn't be any worse, as the Wildcats get ready on Tuesday to play a Mississippi State team that is one of the few teams even better at offensive rebounding than Auburn. In today's article, I take a long look at UK's defensive rebounding struggles and how the Cats might turn it around.
How bad was the Auburn performance? Even worse than you think..
Kentucky had a defensive rebound rate against Auburn, a season low (click here for a note on how rebound rate is calculated). Their previous low was 62% against South Carolina, and Kentucky averages 72% for the season, so 55% is very bad. Auburn ranks 22nd in the country, allowing opponents to get only 65% of their defensive rebound opportunities, so you'd expect UK to do poorly. But UK got 15% fewer defensive rebounds than you'd expect, given the number of missed shots by Auburn; that's the worst differential UK has had this season. UK got 10% fewer defensive rebounds against Evansville and Lamar, but have actually averaged collecting about 2% more defensive rebounds than their opponents usually allow. So, this performance was even worse than you would expect.
UK's defensive rebound rate has been much worse over the last month than earlier in the season
Since January 1, UK has a 69% defensive rebounding rate. Prior to January 1, UK had a 72% defensive rebounding rate.
It's not just driven by UK playing teams who are better at rebounding, either. Adjusting for the opponent's offensive rebounding and the number of shots missed, UK has collected 2% fewer defensive rebounds than expected since January 1, after collecting 5% more than expected prior to January 1st. In the 9 games since January 1, UK has posted a worse defensive rebounding rate than the opponent usually allows 67% of the time, but prior to January 1 UK did so only 42% of the time.
UK played some good rebounding teams before Jan 1 in addition to afterwards. Michigan State, UAB, and Louisville all rank in the top 50 nationally in offensive rebounding, each allowing opponents less than 68% defensive rebounding rate. UK had a 72% defensive rebound rate against them, and outperformed what all 3 usually allow. Georgia, South Carolina, and Auburn all also rank in the top 56 nationally, but UK had a defensive rebound rate of 65% against them and only outperformed what they allow in the first Georgia game.
UK's defensive rebounding is more opponent-dependent now, and has a bigger impact on their overall performance
Prior to January 1, UK's defensive rebound rate had a low correlation (0.15) with what the opponent allowed. This means that UK could have strong rebounding games against teams that normally prevent it. But since January 1, UK's defensive rebounding has had a much higher correlation (0.70) with what the opponent usually allows. This means that UK has not been able to rebound well against teams who prevent it.
UK's defensive rebounding rate was largely insignificant in predicting their overall performance before January 1. The correlation between UK's rate and their adjusted margin per possession (which measures scoring margin adjusted for opponent) was -0.05, indicating no predictive relationship. If you knew UK would be bad or good at defensive rebounding in a game, it wouldn't tell you anything about whether they would play at a high level. Since January 1, the correlation between UK's defensive rebound rate and adjusted margin is 0.78, indicating that there is a very close relationship. Over these 9 games, UK has played well when they get defensive rebounds (Arkansas, for example) and not well when they don't (South Carolina, Auburn).
The difference boils down to the performance of one duo
I looked at every combination of UK big men to see how their rebounding changed pre- and post-Jan 1, controlling for the opponents they played. The evidence points to the combo of Nick Richards and Nate Sestina as the biggest factor:
The combination of Richards and Sestina has become the most common big man pairing for UK, but their defensive rebounding has cratered. While they had grabbed 4% more defensive rebounds than you'd expect before Jan 1 (based on opponents), they have collected 12% fewer since Jan 1. Every big man combo with Sestina in it has seen a similar decline, as has Richards by himself (although those have received much less playing time). The primary issue is the big men themselves, as they combined for a 34% defensive rebound rate pre-Jan 1 and are at 24% since.
Looking at Richards + Sestina's splits pre- and post-Jan 1 compared to the rest of UK's lineups, you see how they've weighed down UK lately:
Pre-Jan 1, Richards + Sestina had a 76% defensive rebound rate and were +82 in 248 possessions (+26 pts/100 poss adjusted for opp)
Since Jan 1, they have a 63% defensive rebound rate and are -24 in 211 possessions (+1 pts/100 poss adjusted)
Pre-Jan 1, all other UK lineups had a 75% defensive rebound rate and were +70 in 616 possessions (+12 pts/100 poss adjusted)
Post-Jan 1, all other UK lineups have a 71% defensive rebound rate and are +70 in 463 possessions (+16 pts/100 poss adjusted)
This culminated in the Auburn performance, where UK was -19 in 27 possessions where Richards + Sestina teamed up. The Cats collected only 38% of the available defensive rebounds.
The issue seems fixable...maybe
I went back and watched film of a couple games where Richards + Sestina did well on the defensive glass (EKU, at Georgia) and two where they did poorly (at Texas Tech, at Auburn). There isn't one magic bullet, but a few things showed up:
In their better performances both Sestina and Richards were consistently boxing their man out; this wasn't as present when they struggled
Sestina had a lot of trouble when 1) Richards went to block a shot but missed, 2) the man he was guarding was far from the basket, or 3) he tried to provide help defense
Against EKU, Richards and Sestina consistently put a body on their man. Very few other EKU players crashed the glass, and UK got a huge portion of the available rebounds. The same was true against Georgia. But against Auburn and Texas Tech, Richards went for a lot of blocks (and missed most), Sestina was often caught far from the basket, and the opponent was more aggressive at sending more players to get rebounds.
The effort in boxing out can get fixed, and I think it largely will with some occasional lapses. UK does need to adjust their philosophical approach on shot blocking with Richards, however. When he misses on a block attempt, the offensive glass is usually wide open. Montgomery is athletic enough to help make up for this, but Sestina isn't. They may want to consider having Richards reign in his aggressive ways, honestly.
Sestina isn't going to become a much better athlete, and he's at a disadvantage when trying to box out players away from the basket. He misses a fair number of these, allowing a free run at an offensive board. I'm not sure that is going to get fixed this season.
Kentucky's guards have helped out a lot on the defensive boards this season; they have the highest combined rebounding rate for any trio of perimeter players under Calipari at Kentucky. They have combined for a 26% defensive rebound rate this season, but only 22% when playing alongside Sestina and Richards. What's more, they had a 24% defensive rebound rate pre-Jan 1 with Sestina and Richards, but only 20% since then. It won't solve everything, but it'd be nice to get more commitment from UK's guards when these two are in the game.
UK's rebounding struggles are a recent phenomenon, and it's hurting the team's overall performance. It's most traceable to the combo of Richards + Sestina, and the primary culprit seems to be their technique. Better help from UK's backcourt can reverse some of this, but UK needs to really prioritize the defensive glass with these two in the game together.
I want to thank you for subscribing to my newsletter. At Hoops Insight, I hope to use data to find insights that reveal things you didn't realize about the Kentucky Wildcats. I'm using play-by-play data to track what's happening, who's doing it, and who is in the game, in order to show you things the box score can't. Check out my past newsletters in the Hoops Insight Archive, and read about the stats I like to use in my stats glossary.
If you have any questions about things I'm saying, the data behind it, or if you just want to debate a point, feel free to contact me on Twitter at @hoopsinsights or email at sean@hoopsinsight.com. I'd love your feedback on the newsletter and how I can improve. Thanks, and I hope you enjoy my work.