Hoops Insight: Louisville Cardinals Newsletter #1
UofL's Same-Four Plus/Minus stats through the Jan 24th game vs Pittsburgh. Adel and Mitchell are having a huge positive impact, while VJ King seems to be hurting the team.
The table above is a little tough for first-timers to understand, so let me explain what it's showing. As an example, let's look at the data for Quentin Snider. He has played 979 possessions so far this year. If you take all of the 4-man combinations with which he has played, and add up all the possessions they have played with someone else in Snider's place, you get 470 possessions. During the 979 possessions Snider has played, UofL has scored 0.12 fewer points per possession than the 470 where he has been replaced (but other players have stayed the same). When comparing points allowed per possession, UofL has allowed .03 fewer with Snider in than when he has been replaced. Add those impacts together, and UofL has been .09 points per possession worse when Snider has played than when he has been replaced. If you go through all of the possessions in each group and account for the quality of opponent (as measured by KenPom.com), we add .02 points per possession to the raw +/- difference, because the possessions he has played have been against a little stronger opponents...so we shouldn't penalize him as much. This yields a Same-Four Plus/Minus of -.07 points per possession, meaning the data says UofL is worse with Snider than with the players UofL has used to replace him.
One caveat with Snider's data is that it is highly affected by the fact that UofL has blown out Clemson and Pittsburgh in 2 of the 3 games Snider has missed due to injury. Given that those were UofL's two best games this year (adjusting for opponent quality), the data reflects that UofL appears to get way better without Snider. That may be a bit misleading, because his Same-Four Plus/Minus was +.05 before his injury...so he had a positive impact for most of the year. However, we shouldn't ignore the fact that UofL has adjusted fairly well to his injury. I doubt Snider makes UofL worse, but there isn't much data this year to indicate that UofL falls apart without him.
Looking at the Same-Four Plus/Minus is the far right column, You'll see that Deng Adel and Donovan Mitchell are by far the biggest positive difference makers for UofL. Given that Adel plays about 50 possessions per game, he improves UofL by 50 x .23 = 11.5 points per game. On the flipside, UofL has really struggled when VJ King is in the game. Given that he tends to fill in for Deng Adel, it makes some sense that their ratings are mirror images. About 80% of King's minutes come with Adel out, so he suffers by being compared to UofL's most valuable player (maybe?). Clearly, there is a large gap between those two.
Let's take a deeper look into some of UofL's stats with and withou Adel and Mitchell, and see if we can pinpoint what makes them valuable.
Deng Adel
Our Same-Four Plus/Minus stats indicated that Deng Adel has a significant positive impact on UofL, both offensively and defensively, when compared to other options. By looking at more detailed data above when Adel is in vs out, let's see if we can identify how he's making such an impact.
First, let's look at his offensive impact. UofL shoots a little better when he's in the game, and shoots 3 pointers more often (although effective FG% accounts for that impact). UofL also takes a few more free throws, which are almost always good signs for an offense. The biggest impact, however, is that UofL turns the ball over much less often with Adel in the game. A big reason for that is that Adel himself has a very low turnover rate; in fact, he has the 2nd lowest on the team, per KenPom.com. When Adel leaves the game, not only is he replaced by someone who turns it over more frequently, but the players who take on a larger role are much more turnover-prone. In particular, Tony Hicks and Anas Mahmoud are involved about 20% more when Adel leaves the game, and they have 2 of the 3 highest turnover rates on the team. So, turnover-prone players take on a larger role of the offense when Adel sits. Quentin Snider and Donovan Mitchell are two options who could take on a larger role without Adel and commit few turnovers.
Now let's look at Adel's impact on defense. From the table above, we can immediately see that opponents shoot a much lower percentage when Adel is in the game. They take a few more three pointers with him in, but also take more free throws. Turnover rates are not impacted. So, it looks like Adel's positive impact is driven by limiting opponent shooting. We can break down opponent's shots by type of possession by looking at how long it took to get the shot; putbacks and shots taken under 10 seconds tend to be high percentage, since they are either very close to the basket or fast breaks. Shots taken deeper into the shot clock tend to be more difficult, since it means the defense has had time to set up. When Adel is in the game, opponent shoot a little worse on shots deeper in the shot clock, but take just as many whether he's in or out; that's causing some of the difference but not all of it. The real key is on putbacks; opponents shoot 63% on them when Adel sits, but only 38% when he's in the game! Given that these shots are right around the basket, it's impressive that UofL holds opponents to such a low percentage. Opponents also shoot 53% when Adel sits on shots taken within 10 seconds (mostly fast breaks), but 45% when he's in. Adel's ability to recover defensively on easy putbacks or fast breaks is making a huge difference, and causes opponents to shoot about 5% worse when he's in the game.
In summary, Deng Adel helps UofL offensively by committing few turnovers, and limiting the reliance on more turnover prone players; he helps defensively by being able to challenge opponents' easy shots. Combining these efforts, he has a tremendous positive impact on UofL's ability to win games.
Donovan Mitchell
According to Same-Four Plus/Minus data, Donovan Mitchell has little impact for UofL on the offensive end relative to his replacements, but has a significant positive impact on the defensive end. Given that most fans would think of Mitchell as an offensive standout, this may be a little surprising. Let's take a look into more detailed data in the table above to see where exactly he's impacting the game.
Right away, we see that UofL shoots a better percentage from the field when Mitchell leaves the game. Mitchell himself isn't a very accurate shooter, shooting 44% on twos and 35% on threes. When Mitchell leaves the game, UofL shoots fewer threes (25.8 3PA.FGA when out vs 33.1 when in), indicating that UofL relies more on frontcourt players; nearly all of UofL's frontcourt players shoot a higher percentage than Mitchell. UofL does draw more free throws with Mitchell in, and turn the ball over a bit less; these offset the worse shooting, and Mitchell's offensive impact is effectively a wash.
On the defensive side of the ball, however, Mitchell's impact shows up. UofL opponents shoot a lower percentage from the field when Mitchell is in the game, and draw fewer free throws. Free throws are a great source of offensive efficiency, so limiting opponent's attempts is a great way to defend better. When Mitchell is in, UofL's opponents shoot much worse on possessions using 11 or more seconds of the shot clock. These possessions tend to come when the defense is set, and teas have to execute their offense. Mitchell's individual defense on opposing guards likely limits them from creating quality shots in the flow of their offense. Opponents shoot just 39% on shots taken 11 or more seconds into a possession with Mitchell in, compared to 45% when he leaves the game. These possessions account for almost 80% of opponent shots, so the difference in shooting has a significant impact, and accounts for the majority of Mitchell's defensive impact.
In summary, Donovan Mitchell's subpar shooting relative to other UofL options mitigates some of his positive offensive qualities, and his impact on that end is a wash; however, his ability to defend deep into the shot clock without fouling limits opponents ability to get good shots, and has a significant positive impact on UofL's defense. All told, Donovan Mitchell is a significant part of why UofL's defense is so strong.
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