Luck in the Journal

There was a nice article on baseball and luck in the Wall Street Journal Weekend Edition on Saturday by Russell Adams that dovetails nicely with some of the things I've written before. One of the passages I found interesting was this about how teams are now factoring randomness into their evaluations:

Most teams are tight-lipped about their use of luck-related statistics, but baseball executives acknowledge that it's becoming part of the decision-making process. Teams including the Red Sox, Indians and Oakland Athletics say they seek to weed out random factors in assessing players.

For the San Diego Padres, currently leading the National League West, measuring randomness will help guide critical decisions as baseball's July 31 trade deadline approaches. Among other things, the Padres use a method to calculate a statistic known as EOPS, which is a combination of a player's expected on-base percentage (the percentage of his at-bats he gets on base safely other than through errors or a fielder's choice) and his slugging percentage (a measure of batting average that gives more weight to extra base hits).

Before the Padres consider spending to beef up their lineup, they want to understand whether their division-leading season is more than just a fluke. "It's important to know: Are we for real or are we lucky?" says Kevin Towers, the team's general manager.

It would be interesting to know how they figure "Excpeted on base percentage" or whether this is simply a combination of OBP and SLUG as has been done before. Adams also talks a bit about the White Sox:

As many fans intuitively know, skill can only take a team so far -- winning a championship requires a certain amount of luck. It's "pretty unusual" for the best team in the league to win the World Series, says Jim Albert, a professor at Bowling Green State University who uses baseball to teach statistics. By several measures, the 2005 Chicago White Sox were one of the luckiest World Series winners in modern baseball history, winning as many as nine extra games due to chance.

Although he has mentions the White Sox being lucky to the tune of nine games in 2005, the analysis I did had them at about 13 games or so but of course some of that is also due to good relief pitching and having a team that can consistently score between four and seven runs per game. Anyway, great stuff.

BTW, Joe Sheehan has a great article on BP where he discusses how the 2006 Sox, much like the 2005 version, are winning via the long ball and how reality and perception are in conflict.