Basketball Analysts' Mistakes of Certainty

I am no fan of the mainstream media and I find most college and NBA analysts barely tolerable (except the comedy routine of Jeff Van Gundy, Mark Jackson and Mike Breen - those guys are classic). I tire of newspapermen disregarding bloggers because they did not go to a proper journalism school or because they are not full-time writers or because they lack a press credential.

I believe the best writing and the best analysis occurs through blogs. Not every blog, of course, as there are as many bad blogs as there are bad newspaper writers. However, the best bloggers - guys like Kevin Pelton, Henry Abbott, Ryan McNeil, Ken Pomeroy, Free Darko, Kevin Arnovitz and others - exceed the content and analysis of most "real" journalists and analysts.

The problem, I believe, is the curse of knowledge, or more specifically, partisan thinking. When a former NBA player or coach analyzes a game on television, he believes that he is an expert because he played the game. As an expert, he starts to believe that his opinions and beliefs are facts. When he analyzes games, he colors his commentary to reinforce his beliefs.

The "amateurs," however, start blogs because of their passion for the game. They approach the game not from a belief in their own expertise, but as a desire to learn more. Rather than preach about their own beliefs, they ask questions. They offer theories and then do the research to see if they are right. They do not offer opinions as an expert; they conduct research. While an NBA analyst attempts to prove himself right with his commentary, the "amateur" attempts to find answers.

The easiest example over the last few years has been the Phoenix Suns' defense with Mike D'Antoni (and, before that, the Sacramento Kings' defense with Rick Adelman). Another recent example has been the plight of the offensively-challenged UCLA Bruins.

Over and over, NBA analysts called the Suns' defense terrible; however, in 2007-08, the Suns ranked 13th in defensive field goal percentage. 13th is not great, but it is above-average, and above-average certainly is not terrible. However, to listen to NBA analysts, one would think the Suns were the worst defensive team in history. As I wrote early this season:

"Watching the Suns last night was annoying because I had to listen to Jackson talk about how much the Suns have changed on defense with Terry Porter even though they gave up 98 points to a team that Tim Legler said "would struggle to score points." The Spurs played without its best offensive player and still had 98 points in what was a slower tempo game than a Suns game of old. So, has their defense improved, as Jackson suggested over and over again, even ridiculing Mike Breen when he pointed out that the Suns were 13th in FG% defense last season?"

Jackson wanted to see improved defense because he firmly believes, despite the objective statistics, that Phoenix played no defense last season. He allowed his personal bias to affect his analysis. In How We Decide, Jonah Lehrer cites a study of political pundits who suffer from the sin of certainty:

"Although they were paid for their keen insights into world affairs, they tended to perform worse than random chance...Tetlock also found that the most famous pundits in his study tended to be the least accurate, consistently churning out overblown and overconfident forecasts. Eminence was a handicap...Tetlock notes that the best pundits are willing to state their opinions in 'testable form' so that they can 'continually monitor their forecasting performance.'"

Bloggers write in "testable form" and then use objective measures to "monitor their performance." Recently, UCLA basketball fans and the popular media have complained about UCLA's offense. However, having watched many UCLA games, I did not see a significant difference betwen their offense from this season and past seasons. UCLA has been a low scoring team since Ben Howland was hired, but they play at a very slow pace. Their offense is typically efficient.

After arguing with some other alumni that UCLA stumbled against Arizona State not because of its offense, but because of its defense and rebounding, I decided to find some objective data and used Ken Pomeroy's statistics. As it turned out, UCLA, at the time, was 5th in the nation in offensive efficiency, while its defense was outside the top 30. For the past three seasons, UCLA offensive efficiency had progressed each year, moving into the top 10 last season, but its defense had been near the top of the top 10 for three straight seasons. The change from last season to this season was not offense, as everyone suggested, but defense. The L.A. Times, however, ran a series of articles and numerous "Letters to the Editor" complaining about Howland's offense and talking about the offensive ineptitude. However, the offense, in the games that so frustrated the viewers at home, actually became more efficient, as UCLA moved to #1 in the nation in offensive efficiency in the week when the Times blasted its offensive execution and questioned Howland's coaching.

The problem is that the Times and the UCLA fans could not separate the tempo of the game. They saw low scoring and equated it with bad offense. Similarly, Mark Jackson saw the Suns surrender a lot of points in a very fast tempo game and equated it with terrible defense. However, objective statistics proved otherwise.

In Lehrer's book, he writes that "like partisan voters, they selectively interpret the data so that it proves them right. They'll distort their thought process until it leads to the desired conclusion." He looks at Republicans and Democrats and the way they rationalize comments. Basically, people hear what they want to hear. If a Republican hears a Republican president contradict himself, they overlook it. However, if a Democrat contradicts himself, they find the statements incoherent (Same if you flip Republicans and Democrats). People with strong feelings ignore facts if they do not support their strong feelings.

When passionate UCLA fans believe that UCLA's offense is terrible because they see a low scoring game, they cannot reconcile the objective data that UCLA ranks as one of the best offensive teams in the country, just as NBA analysts who believe that good defense means preventing the other team from scoring cannot reconcile a better than average defensive team that gives up over 100ppg. They allow their prejudices to distort their viewpoint.

Bloggers, on the other hand, are passionate about the game or their favorite team, but they constantly seek new answers. They look beyond the obvious, while the television analysts stay on the surface. While television analysts focus on points per game, bloggers invent new statistical measures to create different ways to look at and analyze the game. They are not satisfied with the traditional approach, so they pose hypotheses and find answers. In this way, bloggers provide more useful information than the mainstream media and the former NBA players and coaches analyzing the game on television. Despite their "amaterur" status or lack of pedigree, they do not allow their biases to shape their commentary, but instead search for new answers to their personal questions, and they share these answers with their audience, creating, in the process, new ways to look at the game.