On redmen message boards someone broke down +/- for Malik last night.....For all the Ellison Love/Hate. Post from message board below:
Ellison was -18 in the first half. Rest of team was +5 without Malik on the floor, a 23 point swing.
Second half was a little better. Ellison was -5, rest of the team was +4 without Malik on the floor, a 9 point swing.
For the game Ellison was -23. The rest of the team was +9 without Malik on the floor, a 32 point swing.
If you remove a -3 for rest of team and a +3 for Ellison when the game got out of hand in the last few minutes, it's -26 and +12, a 38 point swing in ~36 minutes of basketball.
And it wasn't like it was only driven by one good/bad run or something. If you remove the end of the game, Ellison had 6 sessions on the court. Only 1 was positive, a +1 when he scored the 8 quick points. Over same stretch of game the rest of the team had 5 sessions without Ellison on the court. Only 1 was negative, a -2 in the last 30 seconds of the 1st half when Creighton had a tip in near the buzzer.
Individual +/- is a flawed statistic in my view, especially in small samples. I don't think one player being +4 and another being -4 in a game is statistically meaningful, or even necessarily means the player in the positive had a better game. Bigger fan of +/- for entire 5-player combinations.
But when the numbers are extreme it's difficult to ignore. And these numbers are absolutely astounding. Especially for a player that played the third most minutes on the team last night after playing the second most against Georgetown, while not playing particularly well in either game.
Plenty of other guys didn't play well last night, but this seems to be a growing trend for Ellison with significant minutes for not a lot of production. What is the staff trying to accomplish here?
+/- is a weird stat because I don't think it accurately reflects individual performance, but rather the unit around him. There were lineups he was in that could not do anything, and everyone was negatively impacted. I am sure he had a definite negative role, but there needs to be a deeper analysis. For example, Javale McGee has the highest +/- Per 36 because he plays around Steph, Klay, Dray, and KD. If Malik was starting on Duke with Kennard, Allen, Tatum, and Jefferson then he also might have a good +/-. I think we need to get an exact +/- of how he does with the starters and the bench unit to see who he plays better with.
On KenPom., last night, no starter had above a 100 Offensive Rating which is awful. Looking at the crazy formula that Dean Oliver created, 100 seems to be a good night. Ellison was the second highest starter last night, with only Sham being in front.
Which stat line would you rather have?
Min/Pts/2PM-A/3PM-A/FTM-A/OR/DR/A/TO/BLK/STL/PF
Player A: 24/6/3-8/0-3/0-2/1/6/1/4/2/1/2
Player B: 25/9/1-2/2-6/1-2/1/5/1/3/0/0/4
A is Ahmed and B is Ellison. Ellison is the scapegoat of the season. He will not be used this much next year. His minutes will be cut in half next year with Simon taking his spot. In my opinion, Ahmed has had some awful games. He charges in and has no idea what to do with the ball or takes a quick three that he somehow makes and does the same thing the next possession but even faster. I have never seen him correctly run a fast break because he refuses to pass. Ahmed has a higher usage rate than Ellison but his ORtg is 94.9 compared to Ellison's 99.0. Ahmed is at the bottom of the Big East for ORtg for players that use at least 24% of possessions in a game. In total, the only players he is above regardless of possession usage is Yakwe, Brandon Cyrus, Sanogo, and Quentin Gooden. Ellison's offensive game closely compares to McCallum from Depaul or Diallo from Providence. I will love that as a bench piece next year.
He is not as bad as people make it him out to be.