Change of leagues can throw curves

"Not only are they changing leagues but home ballparks. They'll be pitching in Fenway Park as opposed to Wrigley Field or Minute Maid Park or Petco [Park]. They're also changing defenses. I think it's our job as we project performance from a scouting standpoint and a statistical standpoint that we take all those variables into account."

A study of the last 10 seasons shows that the ERA of a pitcher who ended one season in the NL and began the following season in the AL climbed by an average of 0.46. Clay Dreslough, the Lead Designer and Lead Programmer for the Baseball Mogul line of simulation games, did the following calculations, beginning with anyone who ended 1995 in the NL and began 1996 in the AL.

Those pitchers allowed 8,964 earned runs in 17,899.1 innings their last season in the NL, an ERA of 4.51. Those pitchers, in the AL the following season, allowed 10,175 earned runs in 18,415 innings, an ERA of 4.97.

One point that can skew the data should be considered.

"Pitchers tend to switch leagues after a poor season," said Bill James, senior baseball operations advisor for the Sox and author of the annual Bill James Handbook. "Derek Lowe going to Los Angeles now, rather than after his 2002 season. This creates a `weighted' sample, which tends to weight the data in a way you might not anticipate."

Lowe, for instance, had ERAs of 2.58, 4.47, and 5.42 the last three seasons.

"One would expect that he would have some tendency to recover, absent other changes," James said. "A pitcher has ERAs over a three-year period of 2.50, 4.00, 5.50, it is fairly likely that he will recover some in the fourth year, simply because he has not performed as well as he is capable of."

Bear in mind that the AL is the more offensive league. In the last 10 seasons the average AL ERA was 4.69, while the NL average was 4.32.

The most obvious reason why pitching in one league is different than the other is the bottom-of-the-lineup dynamic.