UofL vs Southern analysis: So much garbage time!
So much garbage time....what does it do to the stats?
UofL's win over Southern was thorough (final margin of 50 points!) and expected (UofL was a 26 point favorite entering the game). When a team with a new coach and players in new roles gets such a dominant win, it's a great sign that a lot of good things came together, and I like to try to find interesting takeaways. The final margin presents a problem, as both teams likely recognized at some point that the game wasn't going to be competitive and no amount of effort was going to change that. I need to filter out everything that happened in garbage time, and focus on what happened when both teams were putting forth a reasonable effort. Luckily, we can do that by looking at some smart analysis others have done.
Ken Pomeroy posted an article on his website kenpom.com this September trying to identify garbage time. He looked for the point margin at various times in the game when the results after that showed no correlation to the results before. His analysis is here and indicates that there are a few points where this happens, and the team that's ahead basically stops trying to extend their lead. His analysis highlights that with a 25 points lead with 10 minutes left, or an 18 point lead with 5 minutes left, or a 15 point lead with 2 minutes left, the rest of the game is essentially meaningless. So, I've added a feature to my analysis this season that allows me to filter out any possessions where the margin and time remaining match one of these conditions. Below you'll find stats profiles for the Southern game, both including and excluding garbage time. First, with garbage time included:
Then, without garbage time:
Of the 77 possessions in the game, 20 can be excluded as garbage time. Comparing the two using Table A, you can see that UofL's adjusted margin per possessions is a little better if you include garbage time (+0.46 vs +0.41), the Cards shot better with garbage time (66% eFG% vs 59%), and turned the ball over less (17% TO rate vs 19%). Southern shot a bit better when including garbage time (42% eFG% vs 37%). The effects aren't huge, but they are present.
I'll let you comb through the data on each stats profile, but I did want to call out a couple interesting points. First, per Table B, UofL's eFG% on possessions lasting more than 20 seconds was 67% for the full game, but only 33% if excluding garbage time. In on-garbage time, UofL scored 4 points on 6 shots in these situations, but got 8 points on 3 shots in these situations just in garbage time. Looking at the full game you'd make the case that UofL did fine late in the shot clock, but looking at just the part of the game that mattered you'd come to the opposite conclusion.
Second, individual player impacts can be affected by garbage time. In Table E, I show the Same 4 Impact for VJ King and Ryan McMahon. If you look at each of the stats profiles, you'll notice that VJ King's "With" possessions stays the same at 41 in each, but the "Same Four Without" possessions drop from 25 (for the full game) to 14 (excluding garbage time). Ryan McMahon has the opposite situation, where his "With" possessions jump but "Same Four Without" stay the same. This means that VJ King did not play in garbage time, but the 4-man lineups he played with did play together in garbage time, and the opposite for McMahon. King's impact will likely suffer when compared to stats put up by his replacements in garbage time, while McMahon may have his impact inflated by garbage time. To get a truer picture of how each player impacted the game, we should exclude garbage time. So that's what we'll do! We will focus on the stats profile excluding garbage time for the rest of the analysis below.
VJ King was a bit unfortunate in this game in that UofL played pretty well when he was in, but really took it to another level when he was replaced. UofL shot really well and limited Southern's shooting when King played, but were unbelievable when he sat. UofL rebounded at a very good level when he played, but were otherworldly when he sat. About the only stat where UofL was below average with King in the game was in committing turnovers; King committed 4 on his own, so he definitely shares some of that blame. The other really interesting stats was UofL's offensive rebounding; the Cards grabbed 83% of their misses when the only lineup change was swapping out VJ King. UofL's offensive rebounding was amazing in this game, and a real step up from a lackluster rebounding game in the opener. UofL was at their best when Jordan Nwora and Dwayne Sutton shared the frontcourt. Both players are solid rebounders despite being somewhat undersized, and they played together during all 14 of the possessions listed as "Same 4 Without" for VJ King. If UofL is going to keep playing just 1 traditional big man, they will need Nwora and Sutton to keep up their effort in getting rebounds.
Ryan McMahon's impact on the game takes a hit when excluding garbage time. He is the least dynamic of UofL's guard options, and UofL clearly takes a hit in both rebounding and drawing free throws with him in the game. This isn't a new phenomenon; last season he had the worst rebounding percentages and lowest rate of drawing fouls of any UofL guard. It's largely an effect of his style of play, as he is frequently outside the 3 point arc and doesn't look to drive the ball. He brings more shooting than any other UofL player, which is clearly valuable; UofL shoots much better with him in the game and scored slightly more points per possession. UofL is more undersized this season, which shows up in a couple of key ways. First, rebounding is more of a team effort. Second, more driving lanes are open on offense without a second post player. McMahon's weaknesses become a bigger issue with this style of player as compared to last year, where his shooting was crucial to spacing the floor for 2 big men to operate. It remains to be seen over the course of the season if his particular strengths and weaknesses lend themselves to UofL's new style of play.
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