The Rotation Change UofL Needs to Make Now
For the second season in a row, UofL's coaching staff made dramatic changes to the team's starting lineup in January. Last season, they didn't take those changes far enough and failed to maximize the team's performance. This season, they haven't yet recognized a major rotation weakness associated with that new lineup and it continues to put an artificial ceiling on the Cardinals' performance. In this edition of Hoops Insight, I make the case for UofL to change the rotation and eliminate a backcourt pairing that has fallen short over and over.
Perry and Kimble have been a weak link
Beginning with the Florida State game on January 4th, UofL switched up the starting lineup and inserted Fresh Kimble in place of Ryan McMahon. While they lost that first game, they ripped off 10 wins in a row before stumbling twice last week. Seems like that move worked, right? A closer look reveals that UofL has been winning in spite of the Perry/Kimble pairing:
During the 10 game win streak, Perry & Kimble played about 30% of the game together; they had a positive +/-together 3 times in 10 games
During the 10 game win streak, UofL had a positive +/- during the time when Perry & Kimble weren't together in every single game
UofL was winning because of how they performed when Perry & Kimble weren't together. Because there's been a clamor lately for UofL to adjust the starting lineup, I wanted to look at Perry & Kimble's performance across various game segments. What I found is that UofL shouldn't just change the starters...they should stop playing Perry & Kimble together at any point:
Adjusted margin per 100 possessions translates +/- into a number that can be compared across games by adjusting for opponent strength and playing time. UofL has improved as the game goes on, largely because they reduce the playing time given to the Perry & Kimble pairing. All other UofL lineups are remarkably consistent through the game, and vastly outperform Perry & Kimble.
Key Problems Have Been 2 point Shooting and Offensive Rebounding
I looked at which stats fall of the most when Kimble & Perry are paired together, and two stood out:
Louisville's 2 point shooting percentage falls from 49% when they don't play together to just 32% when they do
Louisville rebounds 39% of their own misses when other lineups play, but just 22% when Perry & Kimble are together
Most of Perry & Kimble's time together has been alongside Dwayne Sutton, Jordan Nwora, and either Steven Enoch or Malik Williams. I looked at their individual shooting and offensive rebounding to see where the problem is.
Perry & Kimble hit 21% of their twos when playing together, but 41% when they are apart
Sutton & Nwora hit 38% of their twos when playing with Perry & Kimble, but 49% otherwise
Williams & Enoch hit 34% of their twos when playing with Perry & Kimble, but 54% otherwise
Literally every player in the rotation with Perry & Kimble struggles to score inside the arc. Oddly enough, there's no dropoff in 3 point shooting...in fact, the opposite. When Perry & Kimble are together, UofL hits 47% of threes compared to 38% otherwise. But the drop in 2 point shooting more than erases any improvement from the outside.
Watching the film of some of their games, you see a pretty clear pattern with the offense when Perry & Kimble play together. There is very little dribble creation by them, and the offense largely runs through post-ups or midrange isolations (mostly by Nwora). The defense frequently leaves Perry and Kimble on the outside to double team the post or midrange, and most shots are contested. This may explain the strong 3 point shooting, as defenders frequently leave those two open.
Offensive rebounding is also an issue, although it's easier to trace the source of the dropoff:
Sutton & Nwora actually improve their offensive rebounding with Perry & Kimble, getting 13% of UofL's misses vs 9% otherwise
Williams & Enoch fall off significantly, getting 2.5% of UofL misses vs 12% otherwise
Perry & Kimble do not contribute a single offensive rebound when playing together, but in all other lineups UofL's guards collect 9% of UofL's misses
Because much of the offense is running through the post, it is harder for Williams & Enoch to get offensive rebounds since they are shooting. Nwora and Sutton make up for some of this, but not all. The help from the guards is non-existent, while it's actually a big component in other lineups.
Playing David Johnson More Can Help
The biggest reason Louisville gets more offensive rebounds when Perry & Kimble don't play together is David Johnson. He actually has the 3rd highest offensive rebound rate on the team, behind only Malik Williams and Dwayne Sutton, getting almost 9% of UofL's misses when he plays. His length and athleticism are a huge help here.
Johnson also helps with the shooting woes, as his dribble penetration helps open up shots for others. He also has the highest steal rate on the team, and UofL's 3 point defense is even better with him in. His length and athleticism at the guard spot is a huge factor in why UofL has been playing well with him and poorly without him lately.
Flashback to Last Season's Rotation Issues
As I mentioned at the top of this article, UofL had some issues with managing their rotation last season as well. When they inserted a new starting lineup featuring Malik Williams and Khwan Fore, that group dominated. That group wasn't given much time outside of starting halves, however, and I believe that was a major factor in UofL's struggles toward the end of games last season. Take a look at their adjusted margin impact by game segment last season:
Again you see Louisville making some mistakes in rotation management. Whereas last season they failed to play a lineup that was clearly strong, this season they are failing to bench lineups which are clearly ineffective.
Louisville's coaching staff needs to do more than re-evaluate the starting lineup. They need to completely overhaul their guard rotation and stop playing Perry and Kimble together. David Johnson is a better option, and even Ryan McMahon can create spacing via the threat of his shooting. Perry and Kimble need to be alongside one of these players at all times, or else the Cards are going to keep wasting minutes where they could be building a lead.
Thanks for reading my newsletter. For a primer on the stats I like to use, click here: http://hoopsinsight.com/stats-glossary. If you'd like to take a look at previous newsletters I've sent, check out my archive. 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.