Monday, March 9, 2020

A Coach of the Year Polemic

What is the Coach of the Year award for?

If you answer this question by looking at results, the answer is pretty clear: the Coach of the Year award typically goes to the coach of the team that most exceeded expectations. Another way of putting it:

I'm here to say that's the way it should be.

Just like the preseason AP poll tends to give a fairly pure view of how good knowledgeable people think a team will be—before dirty game results sully the analysis with overreactions and the like—preseason expectations for a team give us a pretty pure view of how talented people think the roster is.

There are many reason a team could outperform its presumed talent. Obviously the presumptions of talent could be quite wrong. That happens all that time. It could be just luck. That no doubt happens all the time. But generally speaking when a team outperforms its presumed talent level, I think it's fair to attribute at least some of that variance to coaching—superior training, development, and game strategy.

That's a sound theoretical justification for giving the Coach of the Year award to the coach of the team that most outperforms its preseason expectations. It's not at all dumb. Let's just accept it.

There are two main objections to this regime:

1) Why shouldn't preseason favorites be eligible for coach of the year?

and, relatedly,

2) Recruiting is coaching, too, and this doesn't account for that.

Both of these objections are wrong.

First, under this regime, coaches of preseason favorites are eligible to win Coach of the Year, and they do. Even favorites can wow us with their overachievement. For example, John Calipari was national Coach of the Year in 2015 despite having probably the most talented roster in the one-and-done era. Bo Ryan won B1G COY in 2015 leading a preseason top-5 juggernaut. There are many examples.

Second, and I cannot emphasize this enough, recruiting is not coaching. Coaching has a general meaning that applies to all sports: training players, developing players, and directing game strategy. Recruiting, program-building, fundraising, glad-handing, press conferences, etc.—these are all things college basketball coaches have to do, but none of those things are coaching. The Coach of the Year award need not (and should not) consider them.

Most sports have a Coach of the Year award. In professional sports, coaches don't get credit for having great players (even if they happen to also be the GM that drafted them) because it's understood that coaching and roster construction are different things, and the Coach of the Year award is for the coaching part. Just because the person called "Coach" happens to both coach and recruit doesn't mean recruiting is coaching. It isn't. It just isn't.

The existing regime for deciding Coach of the Year focuses, appropriately, on coaching: instruction, development, game strategy. That's reasonable and appropriate. There's no need to muddy the waters trying to evaluate the non-coaching duties.


  1. Fair points Bart. But I will counter both of your counters to objections #1 and #2.

    1) Of course preseason favorites are "eligible" for the award. The question is: why are they graded on a different curve for the award? If a coach has to outperform already high preseason expectations to even receive consideration, then the deck is unfairly stacked against their chances. It also enters into the equation a coach actually being given implicit extra credit for underperformance in recent seasons. Is it even a question that Pat Chambers' once-strong Coach of the Year candidacy was buoyed by the fact that his team lost 10 in a row last season? Wasn't he the coach during those games?

    2) Recruiting and roster management are two of the most important roles that a coach fills throughout the both the season and offseason. So if the award is going to be "Coach of the Year" then everything should be taken into account. If we are awarding "Coaching Performance This Season" then I would agree with your criteria. But if it's "Coach of the Year" then we should a) look at everything it means to be a coach, and b) take into account everything that happened during the year. If we want it to mean something different, then we should use different words or have a clearly defined criteria that makes it clear we're interpreting the words differently. Is this a bit pedantic? Maybe. I get how this is done in practice, and that Coach of the Year is shorthand for "coach who outperformed expectations." I just the think the actual words we use for this stuff should mean something, or we should lay out specific criteria so we're all on the same page. Otherwise, the award simply doesn't have as much meaning.

    In my ideal Coach of the Year world, you could argue that the deck would be undoubtedly stacked against a Pat Chambers or Steve Pikiell from ever winning the award. The best coaches would undoubtedly win it most years. To which I say: fine. The best coaches should usually be the top contenders for "Coach of the Year," rather than being graded on a different curve because of their past success. (It's the same reason LeBron and MJ didn't win more MVP awards.) This particular Big Ten season was unique in that a lot of coaches who might normally have been in the running delivered really good coaching jobs worthy of consideration. But you don't hear many people mentioning Tom Izzo, and that just seems like an indictment of the entire process we're using. He's not the slam dunk winner; but he's certainly one of the most deserving choices if Coach of the Year is more than just performance against preseason expectation.

  2. on the 2021 team projections, do you only focus on top 10 players?

    also, why are some players missing, for example Oscar Tshiebwe is not showing on WVU for 2021, even though he has not declared, as well as some others I've noticed. Just curious if you are projecting them to leave and therefore removing player.

    found your site after hearing about it, great job!

    1. Yes, projected contributors only lists top 10 for a variety of reasons.

      If players are missing that should be there that either means I don't know about them or I have them coded incorrectly. It's hard to track all 3500+ players! Tshiebwe did originally indicate he was exploring NBA and I had him excluded for that reason but I see that he has stating he is returning so I'll take him off - that list. Thanks for pointing out, and please always feel free to let me know of this stuff because as I said it's a difficult challenge for me to do it by myself.