UK's Losing Streak Isn't As Bad As It's Seemed
UK has lost 3 games in a row...how bad is it?
Kentucky's consecutive losses against Missouri, Tennessee, and Texas A&M have spun a web of uncertainty around their NCAA tournament aspirations. Huge comeback wins against West Virginia and Vanderbilt revived memories of the 1998 NCAA tournament run of the Comeback Cats, for whom no deficit was insurmountable. Those memories have been erased after watching Kentucky get outplayed in the first three game losing streak of the Calipari era. Now, most fans are wondering, how bad is this situation? Is there any hope?
In the recent losing streak, one can find a glimmer of optimism if you look at the results in just the right context. First, Tennessee is a really good team; they rank 11th in the nation per KenPom.com (UK is 34th in the same rankings). Losing to them by 2 points is not a bad outcome; if adjusting for pace of the game and quality of opponent, UK played +0.19 points per possession better than an average team in that game. By the same metric, UK's season-long rating is...+0.19 points per possession better than average. So, that result was right in line with what should have been expected,
The Missouri and Texas A&M losses did represent a dip in form, however. By the same metric, UK rated +0.04 points per possession in those games; those are tied for UK's 3rd worst game of the season behind only the losses against South Carolina and UCLA. The most optimistic way to look at these downturns are that they did come in road games, where teams tend to struggle more. Furthermore, UK has a recent history of losing road games in conference, even to subpar teams; the 2014 team lost at LSU and South Carolina, and the 2016 team lost at LSU, Auburn, and Tennessee. None of these opponents went on to make the NCAA tournament.
To set expectations, none of these losses was in any way unprecedented for a young UK team; the Tennessee loss doesn't even represent a particularly poor performance. That being said, it is troubling to see Kentucky struggling to win more games in conference than they lose. Let's take a deeper look into the stats behind these losses, and see if there are any common takeaways.
Fastbreaks and fouling were the story against Missouri
Kentucky had a few issues that doomed them against Missouri. First, UK posted a season-worst 33% effective FG%. Second, Kentucky tied their season-worst in foul rate, fouling on 41% of opponent possessions (tied with the South Carolina loss). Your opponent has to be severely overmatched for you to be able to overcome your worst shooting and fouling game; Missouri was absolutely good enough to take advantage of these struggles by UK. To make matters worse, Missouri posted an effective FG% of 55% themselves, the 4th highest mark by a UK opponent this season.
The gap in shooting performance was a major issue in this game, but it really boils down to have the two teams performed in transition. UK took 15 shots in transition and hit only 4; with 2 of the 4 makes being three pointers, this works out to a 33% eFG%. The distribution of these shots was a huge issue; only 2 of the 15 transition shots came at the rim, and 9 were three pointers. Based on UK's season patterns, you would have expected about 8 of these shots to be at the rim, and 4 to be three pointers. UK also shot poorly, hitting just 1 of 2 at the rim and 2 of 9 three pointers in transition. Conversely, Missouri only got 4 shots in transition, but hit their only shot at the rim, and 2 of 3 three pointers. If the two teams had similar transition shot distribution and shooting percentages as their season averages, there would have been a 10 point swing in the final score. That alone would have turned a 9 point UK loss into a 1 point win. To further make the point that this performance was unusual, UK's next 2 games saw them return to more normal shot distribution and shooting in transition, with 12 of 23 shots coming at the rim, and UK posting a 52% eFG% on all transition shots.
UK also allowed way too many free throws to Missouri; this was the 5th time all season UK allowed free throws at a higher rate than the Wildcats earned them (as measured by free throw attempts per field goal attempt). These 5 games were: UCLA, at Tennessee, vs Florida, at Missouri, and at Texas A&M. UK is 0-5 in those 5 games. However, there's very little evidence that opponents have much control over how often they draw free throws against UK. None of these 5 opponents rank in the top 50 nationally in drawing free throws, and Florida (167th) and A&M (298th) are below average. None rank in the top 50 in preventing opponent free throws, either, with Tennessee (254th) rating very poorly. There is nothing about these teams that would lead you to have predicted that UK would lose the free throw battle to these teams. Furthermore, among UK's key advanced stats this season, the two with the highest variance from game-to-game are the opponent's free throw rate and UK's free throw rate. Simply put, there is a ton of randomness from game-to-game in how often UK draws fouls, and how often they foul opponents. It's pretty close to random chance when UK allows more free throws than they take. These two stats also have the biggest impact of any advanced stats on UK's per-game performance; much of UK's terrible performance this season just boils to to the randomness of whether they can draw fouls, and avoid committing them. That's what happens when you rely on a lot of 18 year olds.
Tennessee is just a very good team, and they limited UK's shot attempts
As I stated above, the result vs Tennessee is actually pretty much right in line with UK and Tennesee's season-long ratings. However, that doesn't mean that UK didn't have weaknesses in this game. Tennessee did a great job of limiting Kentucky's field goal attempts in this game by forcing turnovers and snaring defensive rebounds. UK ended up taking 6 fewer shots than the Volunteers, partly due to committing 2 more turnovers and getting 1 fewer offensive rebound. While 3 additional shots isn't many, UK would have been expected to score 3 points on those 3 shots; that would have turned a 2 point loss into a 1 point win.
Interestingly enough, Tennessee appears to have UK's number in these two areas. The Volunteers also forced UK turnovers at a high rate in their earlier win against UK, and UK struggled to get offensive rebounds in that game. The big difference between the earlier 11 point loss in Knoxville and this 2 point loss in Lexington was that UK committed more fouls in the road loss; most of the other stats were shockingly similar. Tennessee appears to be a very good team this year, maybe even a top 10 team; you just have to play at a higher level to beat that kind of team, even at home. UK just hasn't played at that level much this year, so I'm not going to complain too much about losing this one.
UK lost to Texas A&M when they couldn't draw fouls
Texas A&M appears to be a pretty bad matchup for UK this year. In an earlier 1 point win in Lexington, UK allowed Texas A&M to shoot very well (57% eFG%). UK overcame this by shooting very well themselves (60% eFG%) and drawing fouls on 37% of their possessions (vs a 28% season average). In the rematch, UK again allowed A&M to shoot well (54% eFG%) and shot reasonably well themselves (50% eFG%). However, UK only drew fouls on 24% of their possessions this time, and suffered the loss. However, as I addressed in the Missouri game recap above, UK's rate of drawing fouls/free throws has a lot of variance from game-to-game. When UK struggles to draw fouls, or fouls a lot, that has a huge negative impact. A&M is pretty good at limiting opponent free throws, at 53rd nationally, but there's not much reason to think that they could consistently avoid fouling the Wildcats. If it feels like there's a high degree of randomness in UK's results this season, it's driven by the fact that the biggest factor in whether they play well (free throw rate) is also the stat with the most variance.
There was one very interesting development in the A&M game that got some publicity. Late in the game, UK made a bit of a run with a 3 guard lineup. More specifically, UK played a lineup of Shai Gilgeous-Alexander, Quade Green, Hamidou Diallo, Kevin Knox, and PJ Washington. This group outscored A&M by 8 points in 12 possessions, all in the second half. Even though all of these players have started a substantial number of games this year, this was only the 3rd game they played together at all, and the first since playing 1 possessions against Virginia Tech on December 16th. While this lineup pretty much defines "small sample size", it's interesting that they managed to put up such strong results. They did this mostly by shooting well (54% eFG%) and forcing turnovers (4 in 12 A&M possessions). However, UK's 3 guard lineups have not shown a tendency to shoot well or force turnovers at a high rate this season, so it's probably just a fluke.
Wenyen Gabriel and Jarred Vanderbilt do represent bright spots
UK's coaches singled out Wenyen Gabriel as being a bright spot, and praised his effort despite the losses. This matches what I'm seeing in the data, as he is UK's only played with a positive plus/minus in the last 3 games. UK is +4 in the 92 possessions he has played over these games, and -26 in the 119 possessions he has sat out. With Gabriel in the game, UK opponents have really struggled to shoot recently. UK has also rebounded well with Gabriel in the game, especially defensively; they are getting 78% of opponent misses with him in and 69% with him out. The pairing of Gabriel and Jarred Vanderbilt has been especially effective recently, holding opponents to a 37% eFG% and outscoring them by +9 points in 50 possessions. UK also gets more fast break opportunities with Vanderbilt in the game due to his ability to get defensive rebounds and make outlet passes. However, UK has had problems with fouling too much with these two in the game.
UK's 3 game losing streak does represent their most disappointing stretch of the season. However, there are some reasons to think that all is not lost. First, the loss to Tennessee was not a particularly bad performance by Kentucky; they just ran into a very tough team who was a bit better. Second, the loss to Missouri was largely driven by fluky performances in regards to transition shooting (bad for UK, good for Missouri) that are not backed up by anything those teams have done in other games. Lastly, the losses to Missouri and Texas A&M were driven largely by poor offensive and defensive free throw rates for UK. These two stats have the most game-to-game variance for UK this year, and have an outsized impact on their individual game performances. Simply put, there was a lot of bad randomness in these games for UK, and there's not much reason to expect UK to be this bad in the next few games.
Auburn is a really good team, and UK may lose to them to extend their losing streak to 4 games. Auburn tends to give up a ton of free throws, however, meaning that UK should expect to see some improvement in that area. While it may not result in a win, I expect UK to bounce back somewhat in this game, and I think they are just as likely to play well down the stretch as they are to continue their recent poor play. Have some hope, UK fans...all is not lost. Your're just rooting for a team with a lot of randomness this season.
Thanks for reading my newsletter. If you have any questions, want to argue a point, or have some feedback, feel free to reach out via email at sean@hoopsinsight.com, or on Twitter @hoopsinsights. If you liked this, let me know as well, and tell your friends to subscribe at www.hoopsinsight.com.
You can forward this to others, but please ask them to subscribe as well so I can keep track of who's enjoying my insight and analysis.