April 8, 2011

Between The Numbers

Playing the Odds

by Colin Wyers

The dominant story to start this season has been the cold start from the Red Sox, the preseason favorite to crush their enemies, see them driven before them, and to hear the lamentation of their women. Since yesterday’s “no reason to panic” post (I regret nothing), the Red Sox have lost another game–to the Indians, no less.

I’ve mentioned what the cold start doesn’t mean–it’s not a death sentence, as the Sox obviously aren’t an ordinary 0-5 or 0-6 team–but let’s spend some time on what it does mean, with the help of the adjusted standings (given a fresh coat of paint as of today) and the playoff odds report.

The Red Sox performance to date has a direct and an indirect effect on our estimate of end-of-season wins:

• Those are six losses “in the books,” which means that the Red Sox no longer have a chance to win those games. That’s a modest but real drag on their expected wins and losses (after all, it’s still only six games out of 162–nothing to sneeze at, but not definitive).
• It’s additional information on how well the team will perform going forward–obviously less predictive than the Depth Chart-powered expected win percentage, but still meaningful in its own right.

Looking at results going back through 2005, we’ve come up with a weighting scheme to evaluate the predictive value of a team’s observed and predicted results to predict future performance. Today is the first day we’ve incorporated that into the Playoff Odds this year–which brings us to the Red Sox.

The Red Sox started out as heavy favorites to make the playoffs, with an 84 percent chance to win the playoffs. It seems like it would take a pretty significant set of circumstances to shift that probability, but that’s just what happened–an 0-6 start to the season only crops up in about a third of a percent of the simulations ran from the preseason. In sims where the Red Sox started off 0-6, they only made the playoffs about 66 percent of the time.

Now, in the sim, the probability of winning each game is independent–losing the first six games doesn’t change the odds of winning any of the later games. But in real life, we don’t know the true odds of a team winning or losing any one game, we can merely estimate it. And losing the first six games is a signal that perhaps we’ve overrated the Red Sox’ true talent to begin with. So we take the original estimate of team talent and revise it downward.

All of that leaves the Red Sox in an altogether more precarious place than anticipated to start the season-their playoff odds are still above 50 percent, but one could forgive the typical Red Sox fan from feeling as if he's just been sent reeling.

Of course, as quickly as the Red Sox's fortunes have turned, they could turn the other way. They have a weekend series against their principle competitors, the Yankees (and in the course of writing this, they took the first game of the series, thus ending their losing streak for now). That, as they say, is why they play the games.

Colin Wyers is an author of Baseball Prospectus. Follow @cwyers
Click here to see Colin's other articles. You can contact Colin by clicking here