Kenpom analysis finds Steve Fisher may be top in-game coach
Kenpom has done some analysis of end-game and in-game coaching ability and identifies Steve Fisher as possibly the best. Beilein also rates highly.
http://kenpom.com/blog/index.php/weblog/entry/whos_the_best_in_game_coa…
you're burying the lede: yet another article to anoint Beilein as a top tactician.
Wow, he's 17-1 in overtime games over the past seven years. That's crazy.
If you watch the Fab Five doc, it seems every other game the narrator mentions went to OT, and obviously we kept winning them.
"San Diego State has blown just six of 105 halftime leads against D-I teams over the past five seasons..."
That's pretty remarkable to me as well, that Fisher is winning about 94% of the games where he leads at halftime over the last five years. Very impressive. I would have to wonder how man coaches could claim a better record in that particular scenario - I suspect it is fairly select company.
I think there was something mentioned last weeked about SDSU not losing a single game under Fisher when leading with like 5 minutes left or something like that (I'm probably exagerrating that number and I don't have the patience to look it up). Anyways, it was pretty impressive sounding. Talk about a shut down closer...
That is covered in the Kenpom article.
SDS has only lost 35 games total in the last five years. So, about one in six of their losses was a game they led at halftime. Is that unusual? How am I supposed to know? I suspect it is, on some level, but how significant is that, when you checked every single coach in CBB, shouldn't some be this good just by chance?
And besides, isn't the first half part of the game too? If he was already leading at halftime in these games, how is holding that lead a sign of "halftime adjustments"? Don't you just need to keep doing what you were already doing? I just don't think this analysis gets anywhere meaningful. Yes, it looks like Fisher (recently) has some unusual numbers, eh he also has had good teams in that stretch in a terrible conference, eh.. what does it mean? Probably not much.
I generally like statistical analysis, but Kenpom is starting wear thin with me. Analyzing every aspect of the game takes some of the fun out of it.
Because it's rational to equate a game with the universe?
March 28th, 2014 at 10:14 AM ^
Or what kenpom does to science? It's not science. There are no controlled experiments, there are no testable hypotheses that ever actually get tested, there are no predictions made by Kenpom theories that we can then go test.
Essentially, what he does is mine data for clickbait. Yes, the overall ranking of teams is interesting, but the blog posts like this one, to me, are not. They are just mining data for outliers, which of course there are many, and then putting those outliers into lists and making unsubstantiated claims about them.
At worst, he hunts for correlations without any measure of how likely he is to find such correlations just by chance. Correlations don't show anything causal anyway, but these correlations are more likely just chance observations from a population of unlimited size (hundreds of coaches, and he can run his analysis as far back in time as he wants, giving him nearly unlimited options to find *something* to say). Again, this ain't how we learn things about the universe at all! And it's not how we'll learn things about basketball, either.
There is no report of the uncertainties or the variability inherent in these measures, which we would need to make any assessment of the significance of these results. I'm not sure if Kenpom ever actually does statistical tests. Obviously, if you look at the performance of hundreds of coaches at any given metric, you will find a very large variability, and some large outliers, purely by chance. The whole field of statistics is supposed to assess whether these results are likely to be real or likely to be chance observations. I don't really see Kenpom doing anything like that (but I admit, I haven't looked super closely).
In this case, these issues are on top of the already dodgy bit of assuming that something like "wins over expectation starting at 5:00 to go in the game" is a metric that measures coaching ability. Seems like a lot of things will contribute to something like that, espeically free-throw shooting, for example. It also doesn't seem that Ken has made much effort here to justify the assumptions he's making, or to do more than a hand-wave at claiming these measures have face validity.
So with unclear validity, and no actual statistics, these are just numbers, which may say what Kenpom thinks they say, or certainly may not, and I don't see any much reason to believe one or the other. I fear that this new wave of "statistics" and "advanced metrics" is being generally poorly applied, and then parroted by those who understand very little about what these numbers might actually mean, and this includes the "experts" at the top such as Ken Pomeroy and Nate Silver.
...said no one ever(?)
He was a very good coach here. I honestly feel there is no chance in hell they win in '89 without him talking over. The Fab 5 get all the credit for those runs in '92 & 93, but despite all of their talent, it was Fisher's defensive schemes that made them so consistently good come tournament time IMO. He's always been more of a defensive coach than an offensive one, and I think that leads to people undervaluing him. Defense is not flashy or exciting, but in one and done situations, it tends to always show up.
The Fab Five stain will never go away, but Fisher also produced one of the great moments in MIchigan sports history.
It is sad that he is so little thought of now; perhaps other are like me, latently appreciating him without thinking about it much. And perhaps there can be some situation where he coaches the visiting team at Chrisler, or receives some small honor, and receives a loud ovation, and we can remember him for the worthwhile things he added as coach here.
March 28th, 2014 at 12:03 AM ^
I think the stain that will never go away is people who still think its OK to use "Fab Five" as a shorthand for the Ed Martin scandal.
If I remember right, the only game Fisher's Fab 5 lost when up with 5 mins to play in a game was the National Title Game vs. UNC. Even though, Webber rebounds the ball, down 2 and with seconds left (with an open Rob Pelinka in the corner for 3) before he calls the famed timeout.
never understand why Rose didn't come grab the ball from Webber in that situation. I guess everyone was panicking because there was so little time left. We needed the ball in his hands.
SDSU plays tomato cans. On their best years, they would barely be a .500 team in the Big Ten. Beilen delivered the goods in the erstwhile Big East and the Big Ten.
I will be eternally grateful to Fisher for delivering Michigan's one NCAA Championship. He did a great job of coaching. However, Glen Rice scoring 184 points, a record that has stood the test of time since 1989, played a much bigger part than any strategy on Fisher's part.
Fisher's best "strategic move" was to take the pressure off of the team, which traditionally choked under Bill Frieder's neurotic "leadership" in the NCAA Tournament. His other "best strategic move" was to have Perry Watson "deliver" Chris Webber and Jalen Rose to Michigan. We all know how that worked out.
Fisher is a solid coach, but he hasn't proven that he is a master against the best competition on a long-term basis.
There was more to it than that in 1989. Freider was an excellent recruiter but a poor in-game coach and his teams chronically underperformed in big spots, particularly in the NCAA tournament. As soon as Fisher took over there was a dramatic change in the flow of the game, in particular the substitution patterns were better, players were fresher, shorter dry spells, less foul trouble. I remember that being a striking difference and it obviously worked out well.
It also showed with the Fab-5.
Not the least bit surprising he's being recognized now. He may not have top talent but he'll get a lot out of it.
I think beyond Ed Martin, the Fisher legacy is impacted in the minds of some Michigan fans by the disappointing performance of his Traylor, Taylor, Baston, Bullock etc. teams featuring 2 first round NCAA exits and a NIT championship. I never liked those teams, and wonder if Fisher was struggling to bring together a group as talented as the Fab Five as opposed to finding talented players who play together and fill needs.
Anyway, I think his success at SDSU suggests that his accomplishments at Michigan weren't solely based on Ed Martin. I also think the fact that the team he would have coached in his last season won the B10 tournament and a game in the NCAA under Ellerbe (who no one thinks is a better coach than Fisher) is an indication that he would have done okay that year as well had he been allowed to stick around.
March 27th, 2014 at 10:21 PM ^
First Fisher had the fab 5, next he had the 5 flops. As good as his first few teams were, it's hard to understand how he under performed so much was such great talent 95 through 97.We had two consecutive top 3 or 5 classes in 1995 and 96. I can't think of when I've seen classes like that under perform so much. The general concensus I remember back then was Fisher just let his horses play when he had the fab 5. The North Carolina players were criticizing the Michigan coaching staff after the National Championship game for not making sure the players knew how many time outs they had.
March 27th, 2014 at 10:57 PM ^
...that it's a huge stretch to assume that best final five minutes of game performance = best in-game coach. What about the other 35 minutes, a.k.a, 87.5% of the game?
March 28th, 2014 at 12:26 AM ^
March 28th, 2014 at 12:31 AM ^
It's real! /s
SDSU being outscored 9-6 since 5 min. mark.
Way too many games with lower-level opponents outplaying us. His teams generally played good defense but they didn't make good decisions on offense. They were exciting but aggravating to watch. I'm glad he's done well at SDSU.