Which is UofL's best lineup?
What lineups has Louisville played the most?
Louisville has 5 lineups who have have played 75 or more possessions together this season:
Snider/Mitchell/Adel/Johnson/Mathiang (325 poss)
Snider/Mitchell/Adel/Spalding/Mathiang (163 poss)
Snider/Mitchell/Adel/Spalding/Mahmoud (153 poss)
Snider/Mitchell/Adel/Johnson/Mahmoud (146 poss)
Snider/Mitchell/King/Spalding/Mahmoud (75 poss)
Let's assume that UofL is going to rely on these 5 lineups in the most important games this postseason. How can I evaluate these 5 to determine which is UofL's best lineup?
Looking at the stats, and quality of opponents
I'm going to rate these lineups across 9 different stats:
Adjusted +/- per possession
Effective FG%
Opponent effective FG%
Turnover rate
Opponent turnover rate
Offensive Rebound %
Defensive Rebound %
Foul rate
Opponent foul rate
I'll rank the lineups 1-5, with 1 being the best and 5 being the worst, for each of the 9 stats. I'll also rank them 1-5 on how many possessions they've played, to give credit to lineups who have sustained their performance across a larger sample.
I'll add up the rank for each lineups across all 10 measures to get a total; a lower total is better.
I'll do this 3 separate times:
Full season stats
Stats vs KenPom top 50
Stats vs 4 or 5 opponent starters
This will give me a view of which lineups have done well against tougher competition, or against opponents' stronger lineups. I'll weight their performance as follows:
25% for full season ranking
50% for ranking vs KenPom top 50
25% for ranking vs 4/5 starters
So, the final output will be a ranking of these lineups 1-5, based on their performance across the most relevant stats and weighted for performance against stronger competition.
So, what do the results say?
The results say UofL's coaches are smart...and fans may be too!
When I tallied up the stats, here's how they ranked:
1) Snider/Mitchell/Adel/Johnson/Mathiang - 26 pts
2) Snider/Mitchell/King/Spalding/Mahmoud - 26.5 pts
3) Snider/Mitchell/Adel/Spalding/Mahmoud - 29.75 pts
4) Snider/Mitchell/Adel/Johnson/Mahmoud - 33.75 pts
5) Snider/Mitchell/Adel/Spalding/Mathiang - 34 pts
So, 2 lineups are clearly above the others: UofL's usual starters,
and the lineup who started recently vs Syracuse. The two have very different strengths and weaknesses however.
Snider/Mitchell/Adel/Johnson/Mathiang, UofL's most common lineup this year, stands out for their ability to avoid turnovers and rebound well (especially defensively). They've been UofL's most played lineup against top 50 teams and against 4/5 starters, so they've been leaned on heavily. They aren't particularly weak in any area, except for the fact that top 50 opponents have shot very well against them. They are generally solid in every stat category, and compare favorably against all of the alternatives. Given that we have a pretty sizeable sample of their play, it's fair to say that UofL can rely on this lineup.
Snider/Mitchell/King/Spalding/Mahmoud have played much less (less than 1/4 as much as the starters) but have had notable success against some tough teams. In particular, this lineup was excellent against Indiana, Duke, and at home against Syracuse. Their results are skewed by the Syracuse game, where they were +18 in 13 possessions with a 100% eFG%. Interestingly enough, this lineup did not play together in either of UofL's final 2 games. This lineup has been a monster at forcing turnovers and offensive rebounding, but pretty bad at defensive rebounding and fouling. It's tough to have a lot of confidence, given that they have a small sample size, but they have played 3 outstanding games against top-50 caliber teams. It's definitely worth giving them some time together in the ACC tournament to see if they can sustain their performance.
While we're on the topic of lineups with small sample size, one more stands out: Hicks/Mitchell/Adel/Johnson/Mahmoud. This lineup was +27 in 33 possessions in the rout of Pittsburgh, and was +6 in 6 possessions in the loss to Virginia on Dec 28th. They've barely played in the rest of the season, but put up 2 strong performances in the 2 games where they played. It's worth a shot at seeing if they can recapture some of that magic this postseason.
Given that I just wrote last week about the fact that Deng Adel is better than VJ King, and that it seemed premature for fans to anoint the Snider/Mitchell/King/Spalding/Mahmoud lineup, I do feel a bit humbled to see the results of this analysis. The reality is, VJ King has had a pretty uneven season, but he's performed very well in this lineup...well enough that they deserve to get some consistent time together this season. Based on my analysis, I'd still say UofL's starters grade out as their best lineup, but I think they have a very nice 2nd option to turn to if they need a boost.
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