David R. Henderson  

Moneyball and Randomness

PRINT
From National Affairs... The Best and the Brightest...

Earlier this year, I posted about sports and randomness. I noted that many people fail to account for the role of randomness in sports. I also noted that Billy Beane, the general manager of the Oakland A's, gets randomness and I quoted a story from Michael Lewis's book.

I went from memory but here is the exact statement from the book, edited with asterisks for obvious reasons:

Billy Beane had been surprisingly calm throughout the team's playoff debacle. Before the second game against the Twins, when I'd asked him why he seemed so detached--why he wasn't walking around the parking lot with his white box--he said, "My s**t doesn't work in the play-offs. My job is to get us to the play-offs. What happens after that is f***ing luck"

By the way, I had the luck (there's randomness again) of seeing the A's win their 20th game in a row, a game that Lewis writes about in Moneyball: My friend I went with and I were talking about it all the way home and I didn't notice that, while I was driving 75, we had moved into a stretch in which the speed limit was 55. I got popped for a speeding ticket. Randomness.


Comments and Sharing


CATEGORIES: sports economics



COMMENTS (8 to date)
ThomasL writes:

I think people have the hardest time accepting randomness where their is traceable human agency involved.

In baseball you've got the pitcher and a batter and the fielders, and all of them are actually doing things that they could have done differently. It is a realm where you have controlled actions mixed with uncontrolled luck, and that mixture seems to make people particularly uncomfortable.

People have historically been accepting of the idea that lightning, tornadoes, or earthquakes are random. Not nearly so accepting that the football kicker is.

I think that acceptance of non-human randomness is fading as well though, as people begin to believe that they comprehend the world. Quite a number of people seem to think something went wrong that Katrina even hit New Orleans... not that the response was poor, but the fact that there was a hurricane that struck at all.

I have talked about this a lot with my father. He works in heavy industry, and his company has a "zero accidents" safety policy. I understand the sentiment of wanting their people to be safe, but from what he has told me of their policies, it seems to me they have created a fictional, "riskless" world for themselves. Anything short of that is blameworthy.

Because most accidents are the result of some human mistake, they imagine a perfectly safe world is the natural state, if only these pesky humans would stop ruining it. Perfect safety is the starting point, and it will stay that way if only they eliminate the possibility of human error. They may be more or less correct, but I think that "if only" is quite a leap to believe in.

I suppose I'd make a horrible safety officer if I went around saying, "Do your best, but bad things happen," but I think it is the truth, while their version is fiction.

The reason this is so important, is that once the impression has shifted, anything that goes wrong anywhere, for any reason, has to have be done by some person. In that world, and I think it is the world the general population is coming to believe in, there is no such thing as an accident, risks, bad luck, or vis major--but there is plenty of fault.

Jesse writes:

There is a brilliant xkcd comic on sports randomness:
http://xkcd.com/904/

joshua writes:

The illusion of randomness in sports or lightning strikes is a result of incomplete information. If we had the capacity to observe the entire Earth right down to the movement of atoms, no earthquake or lightning strike would be surprising.

In addition to understanding the universe's laws and movements with perfect precision, if we had the ability to control our interactions with the same precision, no swing of the bat or roll of the dice would be random, either.

But we cannot determine the perfect angle and speed at which to toss a dice from a 4 and make it land on a 6, and even if we could calculate it we do not have the ability to execute it. Thus, the illusion of randomness.

Jesse writes:

Joshua,

If Heisenberg's theory is correct (very strong evidence poitns to yes), then the universe is fundamentally random. Even with theoretically perfect knowledge, our prediction envelope will be limited by how long quantum effects take to propagate to macroscopic results.
In the case of lightning, I would guess this would be on the scale of seconds, since it is a near quantum trigger event (the first electron that has sufficient energy to tunnel through the air gap).

joshua writes:

Jesse,

Good point. My previous comment almost assumes a Newtonian world. Quantum stuff sure throws a wrench into things!

Arthur B. writes:

This is getting off-topic but:

Heisenberg doesn't say anything about randomness, it describes deterministic properties of the wave function. Bell's inequality does go a little further but it only rules out local hidden variables to explain quantum phenomena.

And since we're already off-topic:
The reasons for editing out profanity do not seem obvious at all. They're an integral part of the English language. The use of asterisks is about as futile as prohibition and likely originates from the same misguided puritan ethic.

Mike Rulle writes:

I have always loved sports and statistics (!), even as a child. I first heard of Bill James in the early 70s (or before) when a friend of mine went to Kansas State and brought back tales of a local guy who said "RBIs don't count" or something like that. I love the Sabremetrics stats and have read some of the research debunking the concept of "hot streaks" as a predictor of future behavior. As a Yankee fan, I also witnessed Jeter's HR in '96 which was clearly a fan interference. What would have happened if that HR was called back?

Still, Beane's statements about playoff time is the ultimate in radical reductionism. Athletes are not robots programmed by a randomizer. Small decisions made by individual players and coaches can have enormous impacts. Anyone who has played sports is very familiar with the very real concept of "choking", which I define, counterintuitively, simply as the willed desire to fail (approximately). These competitive results, not unlike efficent market theory, result in ex-post appearance of randomness----which most of the time is correct----but not all of the time.

Beane takes the obvious (randomness is sports), which is merely the baseline, and then washes his hands of what matters (winning).

GIVCO writes:

A road that drops a 70mph limit to 55mph and then back to 70mph may be a speed trap under your vehicle code, which means the violation is null.

Never, ever admit what speed you were traveling at. That confession wipes out all your challenges to their evidence (i.e., speed determined from an airplane is inadmissible in California last time I looked at the VC).

And never, ever let a cop search your car. Its good practice for you and especially for the cops.

Comments for this entry have been closed
Return to top