Howland and Decision-Making Bias

From reading the study, I am not very convinced by the results and interpretation. I actually find it pretty disingenuous statistically.

From a statistical standpoint, my main gripe is that they acknowledge that high draft picks might go to worse teams, and then say that they “control” for this by including a variable for whether or not the are going to a team with a winning record.

An oversimplistic explanation of the problem with this is:
dividing teams into “winning record” teams and “losing record” teams does not accurately describe the difference in quality, and does not remove the potential statistical bias: the team with a losing record and the #1 draft pick is likely to be significantly worse than the team with a losing record and the #15 draft pick; the team with a winning record and the #16 draft pick is likely to be significantly worse than the team with the #30 draft pick.

A top-5 pick is likely to log major minutes immediately for a team at the bottom of the heap (unless you are drafted by the Clippers… and then you are expected to get injured). A #15 pick going to a solid team is likely to be a bench player. A #16 pick going to a solid team is still likely to get some time off the bench; the #30 pick going to the Lakers may be farmed out the the NBADL for a year.

The authors could have controlled much more precisely for how good a team is by controlling for the number of wins. This would make it hard to find an effect of draft order on results, because it is so correlated with the number of wins, but there is some ‘random’ variation introduced by the lottery (where an average team can end up with a top-3 pick) and by trades involving draft picks (where a bad team might end up with an additional low 1st round draft pick) that could be used to identify the effect. The authors either tried this method, found no significant results and didn’t tell the reader (a dishonest but common practice) or didn’t try it (lazy) or didn’t think of it (unimaginative).

I’m not sure that this follows from the academic argument about sunk costs / escalation effect

He drafted guys like Anderson and Ragovic, and I believe has a natural psychological bias to not see what is very clear for the rest of us.

I think that the argument from the literature would not be that he cannot see the underperformance of the players, but that he wants to keep pouring time into them in the hope that they end up justifying the previous sunk cost investments by eventually turning out to be good. If you give up on the players now, the previous investment is ‘lost’, but if the players eventually pull it together, the previous investment may appear to have been justified. Unfortunately, this also means that the longer you stick with something that isn’t working, the more ‘sunk cost’ you feel the need to justify and the harder it is to cut your losses and move in a different direction.

The analogy to the NBA isn’t great, given that we don’t have draft picks and money to invest in a player, and all players on the college roster are ‘paid’ the same. However, a parallel argument could be made to suggest that a coach is more likely to give playing time to a senior in whom 4 years of coaching has already been invested than to a replaceable freshman.

A change in personal bias would require a shift in the way CBH thinks and the way his assistants can challenge his thinking. If the current assistants are a bunch of “yes men” who will tell CBH anything he wants to hear, then the culture will pretty much remain the same. CBH needs to surround himself with assitants that both he and the college basketball world respect. People who can honestly tell him no from time to time, and whom he will listen.

The problem, however, doesn’t solely lie with the assistants. CBH himself must be flexible enough to change. And that has been the rub for some time now. He hasn’t demonstrated a willingness to do that easily. I personally think that is what makes a coach great. That is, the ability to objectively look at your team situation, admit your plan hasn’t been working, and make changes that suit the strengths of your current players.

I live in the Bay Area and follow the De La Salle football team pretty closely. Early this season, DLS had some huge problems running their offense. Coach Lad would say it was the worst offensive performances he had ever seen at the school. After a few games of painfully watching his team struggle, he did something many thought was unthinkable. He benched his veteran senior QB in favor of an untested sophomore. That sophomore went undefeated in every game he started, including the state bowl game. Benching a senior QB sent a message to his team. If you aren’t playing well, you won’t play. Period.

I waited for CBH to take that kind of leadership and risk. It just didn’t happen this year, and we all suffered by watching Rago jack up brick after brick and play matador D. What we need is not just a personnel change, but a personal change as well.

Back in a BN article here , there was a little tidbit quoted from a Daily News Q&A, John Gold asking the tough questions:

DN: Was there any internal conflict about playing zone?

BH: No, but we should have done it sooner.

Assuming CBH is telling the truth, that tells you a lot about the advice he is getting from his assistants. The question is often asked “how can CBH not see this when x,y,z posting on BN have been calling for this for weeks?”. According to the above, the question should be extended to assistants – and the apparent lack of differing opinions is a worry.

Howland’s email address at his Morgan Center address ?

Their base salaries, and by the same token, University of California systems’ primary funding source come from our tax $$$$ collectively. Of course they have their outside income from various endorsements, speaking or apearance fees. etc.

So when we are taxpayers, theoretically we all have a say in it. But then it’s one thing to say all we want, and if we are not big time alumni contributors in the category of David Geffen & others, for example, Dan Guererro, Swainson the vice chancellor won’t bat an eyelash. Strill, we do have a right to be heard.

Towards the end of the Lavin “malaise”, someone posted Carnesale, the former chancellor’s email for us to write to.
I did. His secretary responded courteously, saying the athletics department would look into our concerns. Of course we all knew by then the new AD Dan Guerrero had something up his sleeve already, pending the conclusion of the season.

Now I don’t mean we would all be flooding the chancellor’s office or elsewhere with calls for Howland’s dismissal. It’s just that he needs to hear our input, so precisely analysed, objective without any personal bias nor animosities here at BN. Some zingers here and there may be, but we are humans and we are frustrated.

these are income sports. You’ re right. The revenues generated directly subsidized, or paid for the programs’ expenditures. Sometimes even the entire athletic department. Ohio States’ is a good example. It’s the largest, most profitable one in the country, and operates like a self supporting private corporation even within the university itself.

Coaches’ university salaries cannot even compare with their other fringe benefits within the whole salary package. That I know too. But still, our opinions cannot go unheeded even though they can easily write us because tax payers, in this current economic climate, do not count much anyway unless you are Warren Buffet or heirs to the Wal Mart empire.

In the pros, money is at stake, as well as validation of someone being a top draft choice.

All you have to do is look at the QBs who have been selected over the last 12 years. Top draft choices get more time to prove that they suck. In 1999, for instance, Couch, Akili Smith and McNown were taken in the first round. All sucked. Couch was selected first, and he got 5 seasons to prove he was terrible, getting 59 starts. never improving. Akili Smith was selected 3rd and lasted 4 seasons, 17 starts (going 3-14) with a QB rating of 52. McNown was selected 12th and lasted into a 2nd season (14 starts), never to play again, despite having better numbers than Smith. This is the way it works in all sports. Get drafted high, you get more chances. They’re paying you more money, they want to see a return on the investment. They also don’t want to be the team that gave up on a player only to see him blossom elsewhere.

In college, money is not an issue. UCLA is not making a bigger investment in Dragovic than it is making in any of its other scholarship players. Nor is Dragovic a “high draft pick” in this analogy. Dragovic was only lightly recruited (as many foreigners are). Lane was a 4 * recruit. In fact, the only 3* recruit of the 5 freshmen is Nelson, who is 2nd in minutes among freshmen in Pac-10 play. One of the two 5* recruits in last year’s class is the person who has received the least amount of playing time among those 5 who comprise the most disappointing recruiting class in UCLA history. In fact, he’s received far less time in the aggregate than two players who are no longer with the team (Holiday and Gordon), including one who was a lower rated recruit and plays (or played) the same position (Gordon).

Looking at the evidence, Howland does not favor more highly touted recruits. Nor does he favor less touted recruits. I don’t think he favors Anderson. After all, he yanked him from the starting lineup when there was no clearly superior alternative (and Abdul Hamid has as many or more flaws than Anderson, with less potential). Putting aside Dragovic, I see no evidence that he does anything other than playing the guys who he thinks help the team win now. That’s it. If not for that rule, he would have kept Keefe in the starting lineup last year even after it was painfully obvious that he was no threat to score. He would not have moved Lee to the 1 this year with Anderson being a true point guard.

As to Dragovic, isn’t it also possible that he is a better alternative right now at the 4 than Brendan Lane (who is the only other 4 on our roster)?

While I think that Brendan is more talented than Dragovic and will be better than Dragovic when their careers are over, I do not think that fans can honestly say that playing Lane 30 minutes and Dragovic 10 will give us a better chance of winning Thursday’s game than playing Dragovic 30 minutes and Lane 10 minutes does.

Right now, Dragovic is the better player. He has more experience. He makes fewer mistakes with the ball. On a team with few shooters, he knows how to get his shot off. He knows where to go on defense. The problem with Dragovic is that his effort, to all of us, appears lacking. My personal opinion is that Howland thinks he can motivate Drago to play hard, and when he does, he is our best option at the 4. Moreover, when not playing at 100%, Dragovic’s lack of effort hurts the team probably on a similar level to how Lane’s lesser current ability and greater number of mistakes hurt the team.

Now, this does not make me a Howland apologist for the disaster we see on the court every week. It is Coach Howland’s fault, and he must be held accountable. This also doesn’t mean I like Dragovic. I do not. I never have. I must have counted 20 baskets scored by Stanford that were in his area (and about 5 could have been avoided with 100% effort).

No, I think that Howland plays Dragovic because he has to try the best he can to win. He thinks Dragovic gives us the best chance to win. And Dragovic, playing at 100% of his capability does give us the best chance to win. He can’t say we’re not going to try the rest of this year but instead wait til next year. That wouldn’t be fair to the seniors. So, while I would like to see Lane play more, I do not agree that his not playing has anything to do with Howland having decision-making bias. Instead, it’s an accurate belief that Dragovic is better right now, an inaccurate belief that he can motivate Drago to play to his ability, and a possibly accurate belief that even when Dragovic doesn’t play to the best of his ability, he still is no worse than Lane.

Now, can he and should he bench Drago when he loafs on defense? Absolutely. Should he hold him accountable for failing to get rebounds and for jacking up 3s early in the shot clock? Yes again.

But I am not buying that Howland is playing Dragovic because he’s a top recruit with too much invested in him.