This story appears in the Sept. 19, 2011 issue of ESPN The Magazine.
IT'S BEEN FUN watching so many zombie baseball franchises rise to pennant contenders this season: the Brewers, D-backs, Indians and, for a while there, the Pirates. Who could have predicted the success of all these sorry teams?
Well, we did, using a method I call the Fruit Bowl, because it mixes statistical apples and oranges. The Fruit Bowl, in essence, is a formula for projecting which teams can get better through inertia. Fans are biased toward action: We all want our favorite clubs to sign gazillion-dollar free agents or import the next superstar from East Timor. But the Fruit Bowl argues that sometimes the best thing a team can do is nothing much at all. And of the 10 teams the Fruit Bowl said would improve in 2011, seven actually have, by an average of 14.0 wins per 162 games through Aug. 29. Arizona was No. 1 on the list. Milwaukee, Pittsburgh, Cleveland and Seattle rounded out the top five.
So let's look at how the Fruit Bowl works and what it has to say about potential 2012 surprise teams. First, the apples: Teams improve when they simply cut or trade below-replacement-level players. Every year, many managers give valuable plate appearances and innings to players who perform worse than what could be expected from a readily available replacement. In 2010, the Indians wasted 717 at-bats on Trevor Crowe (.634 OPS) and Luis Valbuena (.531). By simply removing those two from their lineup in 2011, they gained two or three wins. Of course, it can be tough to bench a slumping star like White Sox slugger Adam Dunn, whose 2.5 wins below replacement is worst in the majors. But there's really no excuse for the Mets to squander a big league roster spot on a no-upside scrub like Willie Harris (.655 OPS, -0.6 wins above replacement).
Now for the oranges in the Fruit Bowl: Sometimes a team's record is likely to improve by luck alone. In any season, some snakebitten squad might have a high number of hits but not many runs to show for it, or it may have a high number of runs but surprisingly few wins. Over time, however, a team's wins and losses will tend to reflect its underlying talent. And thanks to data from Baseball Prospectus, we can find out how well every MLB team should be performing, based on its statistical components. (Prospectus calls this "second-order winning percentage.")
Now to bring the Fruit Bowl together. The recipe calls for simply adding a team's apples (the wins it didn't get because of players who are below replacement value) and oranges (the difference between its actual and expected wins). If a team's sum is negative, it should improve next year if it merely discards albatross players and its bad luck evens out. Last season, for example, the Mariners had -9.7 Fruit Bowl wins, meaning they were due for a nice pop of nearly 10 wins this season. It's easy to see why: Seattle blew more than 1,200 ABs on negative-value players who had little future with the club, like the now-discarded Milton Bradley, surely the best example of addition by subtraction. Not coincidentally, through Aug. 29 the M's were on pace to win 12 more games in 2011 than they did last year.
So who's due for a near-effortless bounce-back in 2012? Well, the Astros clock in at -12.4 Fruit Bowl wins. Sure, they're on track for 108 losses, but based on their K's, walks and fly ball rates, their starters should have an ERA around 4.08 rather than their current 4.68. (Opponents have knocked an unusually large proportion of fly balls over walls against the Astros this year, but HR/FB rates typically regress toward the league average over time.) Houston does have the only bullpen in MLB with more blown saves (22) than saves (20), but replacement-level players are probably more plentiful at reliever than at any other position. Although the Astros have been outscored by 154 runs, a sabermetric formula known as the Pythagorean Theorem says that should be good enough for around 54 wins, not the 47 they've earned through August. So Houston has been both bad and unlucky.
But if they do next to nothing next year, that could very well change.
Peter Keating is a senior writer for ESPN The Magazine.