Defying the odds
Can D'backs, Mariners continue to overachieve?
Posted: Tuesday August 28, 2007 11:52AM; Updated: Tuesday August 28, 2007 11:52AM
The Mariners and Diamondbacks, both of whom have a playoff spot that is theirs to lose with five weeks to play, are not just surprise teams because of how they played last year. (Seattle lost 84 games, Arizona 86.) It is also because of the way they have played this year.
The numbers tell you neither team should be this good, with some people claiming those numbers are warnings that collapses are coming. But maybe, just maybe, the numbers are not as trustworthy as we think.
Look at the run differentials for each team, the difference between runs scored and runs allowed that tends to be a good rule of thumb for success. The Diamondbacks have a negative run differential, having been outscored by 34 runs. Seattle is at plus-18.
Just how hard is it to make the postseason with a run differential that low? Of the 88 playoff teams in full seasons under the six-division format, only two teams were worse than plus-18 were the 2005 Padres (minus-42) and the 1997 Giants (minus-9), both of which went three-games-and-out in the postseason. So should Arizona and Seattle be worried? Just how reliable is run differential?
"It can be very useful," said Arizona general manager Josh Byrnes. "But in the second half last year we had a positive run differential (plus-6) and were eight games under .500 (33-41). And I was in Colorado in 2001 when we outscored our opponents by a decent margin (plus-17) and had a losing record (73-89). So it's not perfect."
Byrnes believes in statistical analysis as much as any GM, but his explanations for why his team defies its run differential sound very old school. Byrnes listed five reasons why Arizona has exceeded expectations: 1) in-game decisions by manager Bob Melvin; 2) a reliable bullpen; 3) a lineup with production spread throughout, rather than one that waits for one or two spots to come up; 4) a good bench, and 5) excellent defense. Those qualities, Byrnes said, have served Arizona well in close games. The D-Backs are playing .644 baseball in one-run games (29-16).
In other words, you need to see how Arizona plays rather than what it produces. Perhaps the D-Backs' inspiration should be the 2005 White Sox, who ranked ninth in the AL in scoring and sixth in run differential (tying the '06 Cardinals and '03 Marlins for the worst league rank by a Wild Card Era pennant winner), but played .648 baseball in one-run games (35-19) and won the World Series. Better to be just a little bit lucky than very good.
Arizona's run differential can be passed off largely on just three blowout losses in which it was outscored 38-0. But the D-Backs are a weird first-place team because they do have the worst offense in the league, ranking last in average and OBP and next-to-last in runs.
Seattle, meanwhile, defies convention because its starting pitching has been poor. The Mariners rank 25th in baseball in starters' ERA (5.03), and only the Florida and Texas rotations are easier to hit (.292).
But if there's one common denominator to the surprising success of the D-Backs and Mariners it is that they almost always win games that come down to bullpens. Seattle is tied with St. Louis for the best bullpen record in baseball (24-8). Arizona is next best (25-13). In both cases, the Mariners and Diamondbacks rely on young, relatively cheap relievers. Both teams leave the key late-game outs to pitchers 30 and younger.
This is the year of living dangerously for Arizona and Seattle: keep the game close and feel confident that a key play or a break goes your way. Maybe it catches up to them in September. Then again, I'm beginning to think that baseball's improved competitive balance is making run differential less reliable than it was before so much revenue-sharing, new revenue streams, and the across-the-board increases in front office intellect. I've been saying that the gap between the best team in baseball and say the 15th best team in baseball has been shrinking, and it keeps showing up in postseason series "upsets." So why wouldn't that be true of run differential, too?
Look at like this: In the first six full years of the wild-card format, only three teams made the postseason without finishing among the top five teams in its league for run differential. But in the five years since then, nine teams out of the top five have made the playoffs, and now Arizona (10th) and Seattle (eighth) could make it 11 teams in six years. So maybe, just maybe, the Diamondbacks and Mariners are more legitimate than we think, and our conventional wisdom about run differential less so.