Bad Luck and the Yankees

Baseball fans have very short memories. A player comes out of nowhere and has a fantastic season, despite bad career numbers? He’s a Hall of Famer. Alex Rodriguez has an awful postseason, just a year after he practically carried the team to the World Series? He’s a choker. The New York Yankees have a couple bad offensive weeks? Forget the season stats and the player pedigrees, there’s something wrong.

This phenomenon is not at all surprising, and is certainly understandable. Not just in baseball, but in all situations, we don’t like the idea of events being due to bad luck, or just random distribution. We like to believe that there are reasons for things that happen, whether they be the results of baseball games, horrific natural disasters, or just bad traffic on the way to work. Often, these events do have reasons – they weren’t just bad luck – but other times, bad things that happen aren’t anyone or anything’s fault; they were just natural consequences of this horribly unpleasant thing called randomness.

Don’t believe me? Consider a hypothetical baseball team playing a full 162-game season. Now let’s say the team goes 87-75 for a .537 winning percentage. However, this record would have been better if not for a horrible stretch in the middle of the season in which they lost 10 of 12 games. Take out that stretch, and they’re a .567 team, good for about 92 wins on the season, which would probably put them in the playoffs.

Now most of us would likely conclude from this information that this is a pretty good team, and that we can expect them to be good going forward (assuming the team doesn’t change). Even including that bad stretch, the team is well above average. But guess what? This hypothetical team was really just a fair coin flipped 162 times* (and in case you’re wondering, this is the first and only simulation that I ran).

Yep, a team that should have won half of its games actually won well over half, the same number of wins as the World Champion 2000 Yankees, completely due to randomness and luck. There was no skill here, no players or managers to assign credit to. No, this team can only thank good luck for its success.

What can we learn from this exercise? First of all, any team, even if they play like a .500 team, can get lucky and put together solid seasons. So when a team like the Yankees starts out slow, or a team like the Baltimore Orioles starts out hot, keep in mind that there is a chance that those teams are simply products of random distributions of events.

The other lesson comes from that bad stretch in the middle of the hypothetical team’s season. During that stretch, fans would surely come up with loads of reasons for the poor performance. Maybe the team wasn’t hitting with runners in scoring position, or maybe the starting pitching was falling apart. But this stretch clearly didn’t affect the team’s performance moving forward, as they were very good apart from it. What would have been perceived as the team falling apart was really just an unlucky clumping of losses in an otherwise fine season.

I’m sure you can see where I’m going with this when we relate this lesson back to the Yankees. Remember that awful stretch in May in which the Yankees at one point lost 6 of 7 games and couldn’t hit with RISP? Well look at them now. They’ve won 6 in a row entering tonight’s game, they’re having no problems in the clutch, and they’re the best team in the American League, in the toughest division in baseball.

So next time you are tempted to attribute a bad stretch to bad performance, “choking” or whatever other reason you can think of, consider the possibility that they’re just victim of random distribution. It’s unsatisfying, unintuitive, and incredibly nerdy, but it’s always a possibility. I believe that the Yankees are a great team, and I’ll try to continue believing that no matter what their record shows.

*via http://www2.whidbey.net/ohmsmath/webwork/javascript/cointoss.htm