Joe Maddon…victim.

Confirmation bias is the most pernicious of all biases, the most natural, and the hardest one to deal with, since it is hard-wired into everyone’s brain. It is nearly indistinguishable from wisdom and experience, you see, but it is a bias nonetheless, and like all biases, makes us stupid. Confirmation bias prevents us from accepting and processing new information objectively, and leads us to see it in the light most favorable to what we already believe, sometimes when that light is decidedly dim or even non-existent.

Baseball is full of vivid ethics lessons, and the post-season, with such high-profile games and thick media coverage is annually an ethics smorgasbord, if you look hard enough. Saturday, Tampa Bay manager Joe Maddon, widely regarded as the smartest manager in the game, showed us how confirmation bias works, and the damage it can do.

The Rays are playing the Boston Red Sox for the privilege of contesting either the Oakland A’s or the Detroit Tigers for the American League Championship. Down 1-0 in a best of five series, Maddon sent his ace, David Price, out to even the series. Price got clobbered. He gave up two runs, including a home run to David Ortiz, a left-handed slugger whom an lefty ace like Price should not allow to hit “dingers,” in the very first inning.When the Red Sox swung, they usually made contact with the ball—another ominous sign, for Price, when he’s right, will get swings and misses. By the eighth inning, Price had given up six runs, and was trailing 6-4, yet he was still in the game. This is extremely unusual in baseball—a starting pitcher who gives up six runs in a single outing is rare, and a pitcher who is allowed to remain pitching long enough to be so generous in a post-season game is more than rare—it is remarkable. Yet there was Price, starting the eighth, well after his opposing hurler, Sox righty John Lackey, had been replaced after giving up his fourth run. Predictably (I thought), Price gave up yet another run, in the form of another majestic, homer by Ortiz, and finally Maddon went out and pulled him.

I then remarked to my wife that I could not recall ever seeing a starting pitcher in a playoff game stay on the mound long enough to give up seven runs so late, and that this was especially strange for Maddon, who is known for his “quick hook.” Maddon, in his post-game comments to assembled sportswriters, explained his thinking and resulting actions: Price was pitching great, he said. He was just unlucky, that’s all. Some bloops fell in; some fielders messed up. Price was as good as ever.

This was clearly not true. The manager  was not watching the same game, or the same pitcher, that the rest of us were. Yes, some weakly hit balls fell in, but there were plenty of sharply-hit balls that went right at Rays fielders too. Price was hit hard. He wasn’t locating his pitches with usual precision, he wasn’t fooling the batters, he wasn’t getting the lefties in the Sox line-up out, and he would have, if he was sharp. Not only did the commentators in the broadcast booth see this, but I could see it too. Why couldn’t, or wouldn’t, Joe Maddon, the managerial guru?

Pure, unvarnished confirmation bias. David Price is his best pitcher, so Maddon interpreted what he was watching to confirm that status. Would he have ever left a lesser pitcher in the game to give up so many runs in a crucial game? Literally never.

That’s confirmation bias.

Beware.