I love Mgoblog. I love UMHoops. I love Brian, Ace, and Dylan. Obviously trust Dylan’s analysis more for basketball. Now back to Big Ten basketball.
Let’s keep it rolling: Nebraska -2, Michigan o137.
Still thinking about PSU/IU.
Thought this article was very interesting, particularly given Beilein’s propensity to auto-bench when a player picks up two fouls: http://pickandpop.net/2018/01/brad-stevens-on-foul-trouble/
1 Like
This came up in my Twitter mentions recently with the Matthews thing vs. Illinois. I pointed out that Beilein has relaxed the auto bench with guys that he trusts not to foul. Zak Irvin played heavy minutes vs Oregon in the first half with 2 fouls, for example.
The list of players who he trusts not to foul on this roster… I think it might only have one name (Abdur-Rahkman).
1 Like
I’d hope Zavier could join that list, as well. He is only averaging about 2.2 fouls/40 mins.
True, but every time Beilein talks about his defense right now it is in the context of “He used to foul all the time and now he’s learning not to.”
He is doing a significantly better job there, just one of those things where the past sticks with you.
I’m tempted to tweet this to JB
https://mindyourdecisions.com/blog/2014/05/06/game-theory-tuesdays-why-basketball-coaches-shouldnt-bench-players-in-foul-trouble/
Doesn’t seem to discuss (I might have missed it just skimmed it) that minutes at the end of the game tend to be more important though, does it? Beilein’s thing has always been auto-bench with the caveat that the game is not out of hand.
There’s definitely a middle ground (I see guys like Jaren Jackson foul out with 10 minutes left in a game)… and Beilein probably falls too far on the cautious end, others too far on the aggressive end. I do think that he’s been significantly less cautious than he used to be fwiw.
Are they? Points count the same, no matter when you score them.
Not sure why they even call it a game theory model since it just a simple statistical model. What it leaves out are things like opposing coaches attacking your star player, psychology, and perhaps other real world stuff. I have no idea if anybody is collecting data that could be used to test the model, but it’s doable.
I think most coaches would tell you yes. Obviously from a pure statistics standpoint, no all points count the same.
Personally, I’d rather have my PG on the floor in the last 5 minutes of the game than the 10-5 minute stretch of the 1st half. I think most coaches would agree. Just my 2 cents.
1 Like
Are minutes at the end of the game more important? Or are they just overdetermined? I go back and forth.
Both? Insofar as the players on the court feel their end of game actions are actually directly determining the outcome, it is probably better to have starters in at the end of the game because they will probably be better at functioning in the face of those feelings, which are often based upon the perception that those end of game actions are actually causing the outcome.
The perception is a part of reality even if it is a misperception.
2 Likes
Autobench isn’t so bad this year because of depth.
2 Likes
I am not saying one player is better than another but I must confess I actually like it when MO, Matthews, or Robinson pick up fouls because I really enjoy watching our younger backups play. (Part of the excitement is due to the unknown aspect probably.)
1 Like
One thing to consider (besides the whole is a FT with ten seconds left the same as a FT with ten minutes left in the first half):
In close games, the final minutes might actually be different than the minutes halfway through the first half – more possessions (because of fouling), more foul shots, etc. Or, conversely, longer possessions and isolation plays where one team is trying to milk clock. Also, are the strategic decisions (e,g, whether to pull the ball out and run clock or not) the same (not just of the same importance, but literally different calculations) at all junctures of the game. I don’t have the answers, just wondering whether the standard response to auto bench that a point is a point necessarily accounts for the full picture (even leaving aside clutchness type arguments).
4 Likes
The points technically count the same, but the later you are in a game the more each point affects your win probability, so I feel like there is at least some statistical backing to go along with the other stuff mentioned that can’t be simplified down to numbers. Of course if you’re out of the game at halftime that doesn’t matter, but with our depth I don’t see that happening.
2 Likes
All else being equal, points at the end of the game are more important. Take a look at a Win Expectancy graph - taking a two point lead with 5 seconds left in the game increases your chances of winning much more than taking a two point lead with 5 minutes left in the first half.
The challenge, of course, is for the coach to foresee whether all else will be equal. If benching a player with foul trouble prevents the game from being close at the end, then it’s not worth it.
1 Like
@umhoops - Dylan I hope you got in early on tonight’s Michigan over 137. All the way up to 141 now. That’s a BIG move.
Definitely a big move. The KenPom total is only 134 too.
IU/Penn State is all the way down to a pick-em. I would stay far away from that game. I think Penn State is the better team, but at Indiana is a tough spot. PSU needs a lot of wins quickly.