UK vs UofL Game Recap
This is my recap of the 2017 UK vs UofL game. To read my preview editions, click on the links below:
Part 1: The Season So Far
Part 2: Strengths and Weaknesses
Part 3: How Each Team Can Win, and my prediction
UK won big...how did that happen?
UK's margin of victory was a huge surprise, although I had the Wildcats as a slight favorite in my prediction of the game. In part 3 of my preview, I thought there were some key factors that would help UK win:
UK would likely get offensive rebounds and turn it over, and needed to shoot well and draw free throws; lineups with Diallo & Gilgeous-Alexander were their best bet to do that
UK defends well without fouling, and lately has been able to force turnovers; lineups with Gabriel and Washington in the frontcourt should do best
UK's starting lineup has been more productive than UofL's
How did these hold up? Did they make a difference? Let's take a look at the advanced data:
The Diallo/SGA backcourt was a huge difference maker
Before the game, I highlighted the Diallo/SGA backcourt combo as making a difference for UK. Both players are strong passers, which helps UK shoot better, and both players are effective at drawing fouls. These factors tend to correlate with strong UK performances, and that held up in this game. For the game, UK's effective FG% was 53% (well above the 43% UofL usually allows), and they drew fouls on 32% of their possessions (above the 25% UofL usually allows). When Diallo and SGA were in the game together, UK's eFG% was even better at 54%, and UK drew fouls on a ridiculous 44% of possessions. UK outscored UofL by 33 points when Diallo and Gilgeous-Alexander shared the court, and were outscored by 4 points when one or both was out of the game. That prediction was definitely fulfilled.
For a team that won by 29 points, UK didn't play great defense
It's odd to say when a team wins by 29 points, but I don't think UK was all that great on defense. Compare most of their defensive stats to their season averages:
They allowed UofL to shoot 48% on 2's; UK allows 47% for the season
UK fouled on 25% of their possessions; their season average is 24%
UK forced turnovers on 14% of possessions; their season average is 19%
UK rebounded 71% of UofL's misses; UK's season average is 71%
UofL was at or worse than their season averages in nearly every significant defensive stat. The lone exception was 3 point shooting; UofL shot 12%, while UK usually allows 31%. UofL also shot threes a little less frequently than UK opponents, although by a very small amount. But some very smart analysis (like here) suggests that defenses don't impact three point accuracy as much as three point frequency. The thought is, teams just won't shoot defended threes, so they show up as fewer attempts, not as misses. UofL just had an awful shooting day, which makes UK's defense look good at first blush. Looking deeper, however, you see that UK didn't do much more than their average defensive day.
My expectation was that lineups with Gabriel and Washington up front would be able to force turnovers and challenge shots. That was true, as UofL shot 42% on twos and turned it over 17% of the time when Gabriel and Washington played together, compared to 48% and 14% for the game. That duo outscored UofL by 8 points in 23 possessions; UK was +21 in 54 possessions when those two weren't together. Those are pretty much the same scoring margins per possession, so I'd say that prediction was mostly fulfilled. UK wasn't amazing with them in, but was effective enough, and in the same ways I expected.
UK's starters did outperform UofL's...but UK was willing to switch it up
I noted in my preview that UofL's starters have had trouble outscoring opponents on the season, but that lineup also gets much more time together than any other. I worried that UofL would have stretches with that lineup where they would give up leads to UK's starters, and that UK has been more willing than UofL to mix up rotations to play more effective lineups.
That's pretty much what happened. UofL's starters played 19 of the 72 possessions together, and were outscored by 6 points. They did start the game competitively, but gave up ground to start the second half. To be fair, UofL wasn't any more competitive when the starters were on the bench, but it's another case where UofL's starters have struggled to outscore opponents.
UK's starters didn't exactly set the world on fire; they only played 5 possessions together (the first 5 of the game) before Sacha Killeya-Jones took over for Nick Richards. UK's starters had a scoring margin of +0, so they played UofL to a draw. Killeya-Jones started the 2nd half in place of Richards, as UK showed more willingness to mix up their rotation. In contrast to UofL, who played their starters 19 possessions out of 72, UK didn't play any lineup more than 10 possessions, and the two lineups that played the most against UofL (Green/SGA/Knox/Washington/Killeya-Jones, Green/Diallo/Knox/Washington/Killeya-Jones) have played the 10th and 8th most time together on the full season, respectively. UK lived up to my expectations, and a bit more, in willingness to mix up their rotation to get results.
So, several of my predictions about what would work for UK came true. But what about my predictions for UofL? I thought the following would help UofL:
Undersized lineups (with only 1 big man) would force turnovers and defend well
Bigger lineups (2 bigs, and bigger wings like Sutton and Adel) could get offensive rebounds and shoot reasonably well
I was worried that UofL would have good offensive or good defensive lineups, but maybe not many that would do both. How did that play out?
UofL didn't get to play their best small lineups
Foul trouble for Anas Mahmoud really hurt UofL and their rotation. The undersized lineup I thought UofL could really use is Snider/King/Sutton/Adel/Mahmoud; it's been one of their best defensive lineups this year. That lineup didn't play a second together against UK. Sutton only played 5 minutes, and the combo of Snider/King/Sutton/Adel never played together. However, UofL did play several smaller lineups, mostly with Jordan Nwora at the 4 or 5
Snider/King/Adel/Nwora/Spalding played 12 possessions and outscored UK by 2 points
A super-small Snider/Perry/McMahon/Adel/Nwora played 4 possessions with a +0 margin
Snider/King/Adel/Nwora/Williams played 4 possessions with a -2 margin
These smaller lineups played 20 possessions, and played UK to a draw. The other 52 possessions, UofL got outscored by 29 point. These lineups did a solid job on defense, forcing 4 turnovers and holding UK to 22 points in those 20 possessions. UofL didn't play the smaller lineups I thought could be successful, but the ones they did play had some moderate success.
UofL didn't get to play many of their best big lineups either
Foul trouble and an over-reliance on the starters really limited UofL's ability to play more effective lineups. The only lineup featuring 2 of Mahmoud/Spalding/Williams and 2 of King/Adel/Sutton that played much (other than the starters) was McMahon/Sutton/Adel/Spalding/Mahmoud; that lineup was +4 in 8 possessions, and killed UK on the boards. Limited minutes for Sutton and Mahmoud meant a lot of lineups with 2 of Snider/Perry/McMahon, or a lot of Nwora or Thomas up front.
UofL wasn't able to play the lineups I thought could be successful. Only 4 lineups had a positive scoring margin for UofL against UK, but they all fit the descriptions I predicted for good lineups. Three were undersized upfront:
Snider/King/Adel/Nwora/Spalding, +2
Snider/McMahon/King/Nwora/Spalding, +1
McMahon/King/Adel/Nwora/Williams, +1
And 1 featured 2 of UofL's core bigs, and 2 bigger wing players:
McMahon/Sutton/Adel/Spalding/Mahmoud, +4
So, the few lineups that had success were along the lines of what I expected, and they generally succeeded in the ways I thought they would.
Takeaways for UofL fans
It sounds weird to say after a 29 point loss, but I wouldn't be terribly concerned about this result. Two big things went wrong that defined this game: UofL had a disastrous night shooting threes, and UofL fouled way too much. A 3-25 shooting night just isn't going to happen very often; this will probably be the worst 3 point shooting game all season for UofL. As I noted above, opponents don't usually affect three point shooting much...bad nights are more due to randomness than anything. I don't think there's anything to fix there.
UofL has had problems with fouling all year long, and UK is really good at drawing fouls and free throws. Luckily for UofL, most of the top ACC teams don't draw free throws at a high rate; only Virginia Tech, Wake Forest, Syracuse, Georgia Tech, and NC State. Those teams all have significant other weaknesses, so UofL's foul trouble shouldn't doom them in their tougher matchups.
Foul trouble really destroyed UofL's rotations, but I do think some tweaks are needed. A shakeup to the starting lineup might be in order, and a little more consistent reason to the types of lineups being played could help. But, the lineups played against UK just aren't typical of what usually succeeds for UofL...getting back to those should improve results.
As a positive, UofL was able to have success when they got into their offense quickly against UK. I've written about this before, but UofL has sped up their offensive tempo this season and takes more shots early in the shot clock. For the season, UofL takes 35% of their shots in the first 10 seconds of the clock, and has an effective FG% of 55%. Against UK, UofL took 41% of their shots within 10 seconds, with a 52% eFG%. That's a great indicator that UofL's revamped style was able to have success! The issue was when UofL took more time; they had an eFG% of 26% in those situations, mostly due to missed threes. But the transition baskets and quick drives were there, and led to success.
Takeaways for UK fans
Besides bragging rights, UK fans can feel good about the way the coaching staff made some tweaks to the rotation and playing time after the disastrous loss to UCLA. In my preview, I called for Nick Richards to play fewer minutes, as UK's rebounding was much better with him out of the game. Richards played 17 minutes, his lowest total in over a month. In his 17 minutes, UK was +4 with a 58% defensive rebound rate. In the other 23, UK was +25 with an 81% defensive rebound rate.
During the UCLA game, I tweeted that I didn't understand why UK would ever play a lineup with Hamidou Diallo and PJ Washington both on the bench. The passing ability of those two helps UK's offense tremendously, and on the season UK has been outscored by 33 points when those two both sit. Against UCLA, UK played 11 possessions with both on the bench; the Wildcats were outscored by 5 points in that short time. Against UofL, there was not a single second where both players were on the bench. That's a nice adjustment, and UK would be well served to keep that up.
As I mentioned above, I don't think this was a great defensive game by UK. There isn't a lot to point to other than UofL missing threes, but UK did well enough to win on that end. This was a very strong offensive performance, however, and showed improvement in a lot of areas. UK avoided turnovers much more effectively than in other games, and imposed their style on UofL's defense rather than the other way around. UK's ability to draw free throws has been a hallmark of strong Calipari teams, and that was fully realized in this game. Overall, I think UK fans should feel a little better about their team after this win, since it showcased what they are capable of. However, consistency has been an issue this year, so I wouldn't assume this sets some sort of new baseline.
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