dkSez : : : : : : Don Kahle's blog

What Statistics Show (and Tell)

The PAC-12 collegiate basketball season is roughly half over, so now is a good time to debunk the debunking of “hot hand phenomenon.”  If Erin Boley or Chris Duarte have hit their last three shots, should the point guard feed them the ball to attempt a fourth?

(For those of you who don’t like reading sports analysis on the editorial page, don’t worry. This column isn’t really about basketball. It only starts there.)

Three scientists in 1985 analyzed every field goal attempt by the Philadelphia 76ers over a complete season. Every shot attempted and made was sequenced. The data set appeared complete.

They found a strong tendency for each player’s performance to revert to his individual mean. Simply put, if a player shoots 50 percent from the field, three made shots in a row tend to be followed (eventually) by three missed shots.

So, no. The point guard should not dish the ball to the player who has had a streak of success. The “hot hand phenomenon” became the “hot hand fallacy.” (Later studies have refined the original findings, without overturning the original conclusion.)

And yet. Coaches, fans, and point guards have continued the “tradition” of giving the ball to the person who has had recent success. Why? It’s not habit. Boley and Duarte weren’t born in 1985.

Fans love the drama of “hot hand” improbabilities. My son and I once saw Luke Jackson score 26 straight points for the Ducks. Even from Mac Court’s third balcony, it was a thrill to behold. But coaches are not paid for memories. Coaches are paid to win. 

In the hyper-competitive context of a tournament game, why not take every calculable advantage? Because not all advantages are calculable. Statisticians don’t measure the roar of the crowd, unless you count the Oregon mascot repurposing a rolling trash container to contain the din. That wins games too.

We can’t quite measure how far or how fast a player lunges for a loose ball. Or how loudly they warn a teammate of a pick or a trap. Or when an opposing coach uses a precious time out to slow the pace. Or what generally dispirits the opposing team.

These may all be calculable, but why bother? We know what works. Feed the hot hand. The benefit of success is greater than the cost of failure.

Here’s the point. The experts and their statistics are not incorrect. But they are wrong nonetheless.

Statistics are great. Odds are true. They are also seldom true enough to capture anything that looks like life. We should stop pretending they represent reality. None of us live our lives as if they do.

Nor would we want to. A world where every variable is measured and accounted for? No thanks.

UO Football Head Coach Mike Bellotti once was asked if he thought his team had any chance against a star-studded USC juggernaut that Saturday. He pursed his lips and shrugged, “That’s why we play the game.”

Life is filled with immeasurables — uncertainties. We should be glad for that. It’s why we play the game.

==

Don Kahle (fridays@dksez.com) writes a column each Friday for The Register-Guard and archives past columns at www.dksez.com.

Tags: No Comments