Interleague play opens on Friday night, with the notable difference that for the first time, teams won't merely be crossing over and playing their own division in the other league.

Actually, it's more complicated than that. Because of the different sizes of the divisions, and the desire to continue certain matchups, teams within the same division are playing fairly disparate schedules.

Now, we already have an example of how that can affect a pennant race. In 2001, the Cardinals finished two games ahead of the Giants and claimed the NL wild-card slot (they technically tied for first place in the Central). However, whereas the Giants played the AL West, a division that had the best record in baseball and included the two best teams in the game, the Cardinals played the AL Central, with two of the four worst teams in the league and no one better than 91-71. That doesn't factor in the unbalanced schedule's impact, with the Giants playing in a much tougher division than the Cardinals.

Might that happen this year? Yes, and perhaps even worse. The unbalanced schedule already screws up the wild-card race, so that's a lost cause. However, the gerrymandered interleague schedule of 2002 may play a role in determining the division winners. Some teams are playing much easier interleague schedules than the teams they're battling with within their division, and in cases where races may come down to a game or two, that could be the deciding factor.

Let's look at the data. For contenders, we've calculated the current winning percentage of non-common interleague games. In other words, because the Yankees and Red Sox both play the Diamondbacks, Rockies, and Padres, those games cancel, leaving the games you see in the chart below. If a team plays two non-common series against an opponent, their record counts twice.

The difference between playing the Braves six times and the Mets six times may be the difference in this division. The designation of Atlanta as Boston's "natural rival" is a specious one, based on the Braves having originated in Beantown, and may cost the Sox the AL East crown.

The Indians' opponents appear to have been pulled at random. The White Sox and Twins get a big benefit from the "natural rival" clause, with six games against the only two sub-.400 teams in the National League.

Not having a "natural rival" helps the Mariners, who get games against the Padres and Cubs as those teams sink slowly out of sight.

The "natural rival" thing takes its greatest toll here, as the division with four teams bunched together sees two of them playing six games with teams .633 or better, and two getting six with teams at .404 or worse. The Mets are also the only team playing all three good AL Central squads. We'll see if the Expos and Marlins can take advantage of this to maintain their position as the trade deadline approaches.

The Cards and Astros get some benefit from the "natural rival" rule. NL Central teams are only playing 12 interleague games.

The Rockies have to play the three best teams in the American League, so if they're still hanging around in three weeks, I'll be surprised. The gap between the Giants and Dodgers is entirely due to the surprisingly divergent fortunes of the Angels and A's.

MLB has a half-baked interleague schedule. They're so intent on keeping the only matchups -- maybe a half-dozen -- that mean anything that they've completed messed up the schedule. If this doesn't make the point that interleague play exists for 20 games a season, I don't know what else can.

Keeping "natural rivals" might be good for the gate for a few games, but is it worth it if it costs a team a playoff spot? The Mets may be paying a very big price for three sellouts in June: sellouts in October.

You can check out more work from the team of writers of the Baseball Prospectus (tm) at their web site at baseballprospectus.com. Joe Sheehan can be reached at jsheehan@baseballprospectus.com.