# World Series sweeps

First, congratulations to the White Sox on winning the 2005 World Series!

On October 26, David Pinto at Baseball Musings mused that the odds of a team with a 0-3 deficit in a series taking it to a fifth game only three times in 21 tries was 0.0007. Of course, as David noted, this is on the assumption that the two teams are evenly matched.

As David said:

But given the propensity of sweeps after a 3-0 start, I wonder if this start isn’t telling us something about the superiority of the sweeper.

That is, sweeps happen when the teams are not evenly matched. One way to approach this question is to look at run differentials in sweeps over time. There have been 17 four-game sweeps if we disregard the two in which a team won 4-0 but there was also a tie (1907 and 1922).

In those series, the sweeping teams have combined for 369 runs and the swept teams for 160. That is an average score per game of 5.43-2.35 and an average differential of 12.3 runs per series.

In other words, sweeps tend not to happen the way this 2005 sweep did: with four straight close games going the same way.

Of course, the minimum possible differential in a sweep is 4 runs. This time the differential was 6 runs, half the average. Only one other sweep has seen such a low differential (1950, Yankees over Phillies). Only one additional sweep had a differential less than 10 runs.

The Pythagorean method of estimating win percentages based on runs scored and runs allowed says that the 68 games in these 17 series should split 57-11, rather than 68-0. So, David is right that sweeps happen more often than they “should” but maybe not that much more often, because usually one team is, in fact, dominant. But not in 2005.

What we just saw, then, was a very rare type of sweep. And that the White Sox pulled this off was consistent with their uncanny record throughout the regular season in close games. For instance, in 1-run games, the White Sox were 35-19. A basic claim of sabermetrics is that teams that win close games are lucky, and that such luck is unlikely to hold out over the long haul. That is what led many fans, myself included, to predict the White Sox would come back to the pack after their great start to the regular season. And, of course, they did, having their lead drop from 15 games at one point to 1.5; however, that was more a result of the Indians getting hot than the White Sox having their luck catch up with them.

The White Sox not only continued to defy expectations in close games all season long. They continued to do it in the World Seriesâ€”and, for that matter, in their first two playoff series as well. The first two rounds also featured close games, aside from the 14-2 opener against the Red Sox.