EJ Montgomery's surprising impact, and a mini-Arkansas preview
Everyone who's watched much of UK this season would likely attribute any success the Cats have had to the same players: Hagans, Maxey, Quickley, Richards, maybe some love for Sestina's shooting. But a surprising player grades out as one of the most impactful: EJ Montgomery. I'll try to explain why the data disagrees with the general consensus. But first, I'll offer up some things to watch against Arkansas this Saturday.
What stands out about the Razorbacks?
On offense, Arkansas isn't really anything special. They rank 80th in offensive efficiency per KenPom. The most notable thing is that they rank 256th in 3 point shooting (31.2%), despite having a reputed sniper in Isaiah Joe. Arkansas has 2 guys who shoot often from deep: the aforementioned Joe, and Mason Jones. Neither shoots better than 36%, but there's some additional context. First, Arkansas has a relatively low percentage of threes which are assisted (70%, 13th lowest in country). This means that their shooters are often creating their own shots, and threes in particular tend to be less accurate when that happens. If Kentucky can force Arkansas to create threes off the dribble, they'll be fine; if they allow catch-and-shoot threes, Arkansas will likely outperform their shooting resume.
Arkansas is much more dangerous shooting in transition. 47% of their shots in transition are 3's, and Arkansas hits 38% of them. Joe and Jones are particularly dangerous, as over 70% of their transition shots are 3's and they hit 40%. UK will need to find them in transition and limit those open threes.
On defense, Arkansas has a unique style. First, they allow a lot of defensive rebounds, and these turn into transition shots frequently. However, Arkansas is one of the top teams in eFG% allowed on these shots. It almost seems as though Arkansas lures teams into getting into transition, and is able to close down quickly and challenge shots.
The Razorbacks are very undersized; they rarely play anyone taller than 6'6". Despite that, they are a competent shot blocking team. Adrio Bailey stands 6'6" but is their shotblocking ace, ranking 52nd nationally in block rate. He does most of his shotblocking on shots away from the rim, so UK will need to watch for his help defense challenging on midrange jumpers. The players on UK who shoot a large share of midrange jumpers tend to have size (Richards, Montgomery, Brooks), so it doesn't seem like Bailey should be able to bother their shots too much.
Arkansas hasn't recorded many quality wins this year. Their best win was over Indiana, and 2nd best was an OT win over Georgia Tech. But they also haven't lost at home and will be fired up. If UK is focused they should win this game, but Arkansas will have a lot of fight despite being undersized.
Measuring player impact
One of the great challenges of analytics is measuring the impact a player has on the court. The simplest way involves looking at the per-possession scoring margin when a player is on the court vs off the court. This misses a lot of context however, since it ignores a player's teammates. In an attempt to get a bit more significant measure, I've come up with my own method.
First, I measure the performance for every lineup including a single key player. Then, I look at the other 4 players and pull the data for all of the time they played without the key player. As a last step, I weight this "without" performance according to the time that 4 man combo played "with" the key player. Here's an example:
Hagans, Maxey, Quickley, Richards, and Montgomery have played 145 possessions together; this is 27% of Montogmery's total possessions
Hagans, Maxey, Quickley, and Richards have played 124 possessions without Montgomery; I give their stats without Montgomery 27% weighting when measuring UK's stats without Montgomery
The end result is a picture of how UK performs with and without a player, and it looks like this:
This helps control for a few problems when measuring on/offcourt impact. First, it controls for the other 4 players on the court and looks at the impact of direct replacements for the key player. It also avoids a mismatch in comparing the with and without lineups; if a player plays a lot with a specific 4 player group but those 4 players play very little without him, the results won't be apples to apples. With that being said, here are some insights when running this analysis for UK:
First, both Ashton Hagans and Immanuel Quickley grade out very positively. Hagans' biggest impact is on defense, as UK forces turnovers at a much higher rate and rebounds a higher percentage of opponent misses when he plays. Quickley's impact is felt more on offense, with UK shooting better and turning the ball over less often when he plays. None of this should be very surprising to UK fans when you think about the strengths of these players compared to their most likely replacements.
It's a bit surprising that Tyrese Maxey grades out as a slight negative for UK in this analysis. When I dug into the data, this is entirely due to the Eastern Kentucky game when lineups with Hagans, Quickley, and either Brooks or Whitney at SF blitzed the Colonels. Excluding this game, Maxey has had a positive impact. Again, not surprising considering he plays a lot alongside Hagans and Quickley and therefore gets compared to Brooks or Whitney playing SF.
Another surprise is that Nick Richards grades out with a negative impact. Again, this is due solely to 1 game. Richards plays heavily with 3 guards + Montgomery, and lineups with Brooks in Richards place were absolutely nuclear against UAB. Excluding this game from the analysis turns Richards' impact positive on the season.
The biggest surprise to me was the subject of this article, EJ Montgomery. He grades out very positively, as you can see here:
This isn't due solely to 1 game, either. Several key lineups have seen their results improve when swapping Montgomery in for Richards! My favorite stat is adjusted margin per possession, because it controls for game pace, amount of playing time, and opponent strength. Here are the numbers for some key groups when Montgomery is swapped for other players:
3 guards + Sestina: with Montgomery, +0.25/poss; with Richards, +0.22/poss; with Brooks, +0.06/poss
3 guards + Brooks: with Montgomery, +0.86/poss; with Richards, +0.16/poss; with Sestina, +0.06/poss
3 guards + Brooks + Montgomery has been an awesome lineups. They have played in 7 games, and have outscored the opponents in every one of them. This isn't just against tomato cans, either; they played against UAB, Georgia Tech, Utah, Louisville, Alabama, Georgia, and South Carolina. The one red flag is they only played more than 5 possessions in 2 of these games: 13 poss vs UAB (+16) and 9 poss vs Georgia (+2). They were +3 in 3 poss against each of Utah and South Carolina, however...so they have made an impact quickly even in games where the team struggled.
I don't believe EJ Montgomery is a better player than Nick Richards, despite the data saying that swapping in Montgomery for Richards is a net positive. I do believe, however, that EJ Montgomery has been able to make more of a positive impact than he maybe gets credit for. It appears like Montgomery is most impactful when he is the key low-post big. Montgomery is a much more active and effective player when playing with Richards:
Montgomery without Richards: 58% eFG%; takes 20% of UK's shots; collects 22% of opponent misses and 9% of UK's misses
Montgomery with Richards: 47% eFG%; takes 20% of UK's shots; collects 16% of opponent misses and 7% of UK's misses
When Montgomery plays alongside Richards, he doesn't gets shots in the paint. He takes less than 10% of his shots in the paint with Richards, but 50% when Richards sits. Given that Montgomery hits 75% of his shots in the paint and 37% in midrange, it's much more effective to get him shots near the basket if he's going to play.
Interestingly, Richards doesn't change his offensive profile depending on whether Montgomery is in the game or not. He takes just over 50% of his shots in the paint either way. Montgomery takes a backseat when playing alongside Richards, and there's an argument to be made that the two should play together less than they do if Montgomery's value is to be maximized.
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.