October 10, 2003
What Happened to the A's?
In the wake of the A's' loss to the Red Sox in the Division Series, the fourth straight year in which they've bowed out in the first round, there's been a maelstrom of psychoanalysis, criticism, and...oh, what's a good Chris Kahrl word?...foofaral! Yes, there's been foofaral a-plenty as talking heads, and some thinking ones, try to explain four straight series losses.
Many of the rationalizations are flat-out wrong, even counterfactual. There's still a popular notion that the A's are a "sabermetric" team, following the walks-and-power, damn-the-defense approach that defined them back in the late 1990s. Actually, the A's are a pitching-and-defense team, have been for two years now, and were especially so this year with the addition of Chris Singleton and the commitment to Mark Ellis at second base.
Accusations that the A's lose in the postseason because they can't play defense are patently absurd. The A's prevent runs far, far better than they score them. What they don't do is score enough runs; in fact, the Red Sox triumph over the A's should be see as a validation of troglodyte baseball. The Sox are much better offensively and don't have a real good defensive team outside of a few players. They won, so where is all the praise for that approach? (I'll leave it to the reader to discern where these facts intersect with the media's preconceived notions.)
An extension of this is the idea that the A's are a particularly bad baserunning team, something that no one was really talking about before the series, or even 26 innings into it. It wasn't until the A's actually made a mistake on the bases that everyone started calling them poor fundamentally. Do you know what a fundamental is? It's the last thing you did wrong.
I submit that there's little evidence that the A's are a poor fundamental team; it's just another example of the media taking specific high-profile events and extrapolating them into character flaws, in the same way that a big hit in the right time brands someone forever as a "clutch hitter." Just to proffer one piece of data, the A's outscored their projected runs (according to Clay Davenport) by 27 this year. If they're a bad baserunning team, it seems to me that it would show up in that figure; at the least, it would be a huge barrier to that kind of overperformance.
I'm probably stumbling into some Gilovich here, but the way in which people so desperately want to take the disconnected events of four playoff series, four Game Fives, nine chances to clinch, and find a pattern in them says more about those people than about the A's. The A's have failed in many different ways in those nine games: their starters have gotten bombed; they haven't hit; they've made bonehead baserunning errors; their best reliever has failed. None of those things have happened in every game, and the circumstances of all these losses differ wildly. The A's have performed poorly, and they've been unlucky.
Why does this keep happening? The standard stathead explanation is that the playoffs are a crapshoot. Beane famously, and reasonably accurately, called them "luck." Beane isn't wrong, but we need a better way of articulating this idea. When you use the word "luck," it sounds like making an excuse. Moreover, it's not really accurate; while luck plays a factor in short series, it's not really the core problem with evaluating a team--or should I say, judging it--based on its playoff performance.
The phrase "small sample size" is also bandied about in these discussions. That's closer to what we're trying to say, but it doesn't play well outside the room. Baseball fans and the media that feed them don't care to understand statistical significance all that well, and tend to shun anything that even sounds like senior-year math.
So what we need is a way to explain how playoff series, despite the importance that gets placed upon them, are just five- or seven-game series between two teams that are usually reasonably close in talent, performance to date, and expectations. Some team is going to win three games of five or four games of seven; the order in which those wins occur and which teams gets them aren't very meaningful in and of themselves. It's the superstructure built up around these games, and the disproportionate significance placed upon them, that makes them seem like more than what they are. Baseball doesn't change, though. In a short series between teams good enough to make the playoffs, any outcome is possible, because very little separates the teams. The outcome will be determined primarily by how well those teams play in the series, with luck a factor in the result. Same as it would be in May, same as it would be in August.
It is completely fair to say that the team that wins a playoff series played better in that series. The logical leaps from there are where everything breaks down: winning a playoff series does not mean a team is better than its opponent. It means that it was better over the course of that series. The team that loses doesn't necessarily have rampant flaws that need to be picked over repeatedly; they just didn't play as well as their opponent did in those games. The team that wins isn't comprised of better, tougher human beings.
To watch a team win nearly 400 games in four seasons amidst the toughest competition in the major leagues, then decide that because it went 8-12 in four series against the pick of the league, it is fatally flawed, is to give best-of-five series far too much credit for their ability to separate good and bad. We know this because every year, some last-place team wins five in a row, and some first-place team can't win for a week. The Expos sweep the Giants; the Padres ruin the Dodgers' season. You can't come to an accurate conclusion about baseball teams in five games. All you can do is point at the one winning three games and say, "Go forth."
It's an unsatisfactory response, but it's the truest one. The A's didn't lose to the Red Sox, or for that matter to any of the other the Division Series opponents, because there's some gaping hole in how Billy Beane assembles a baseball team. The A's have basically been a different team each time around; while the core talent has been the same, the supporting cast has changed dramatically, changing the team's shape with it. There's nothing endemic to the A's that connects Terrence Long's dropped fly ball to Jeremy Giambi's speed to Mark Mulder's ill-timed injuries to Keith Foulke's inability to close out Game Four.
The A's lost in 2003 for the same reason they lost in 2002, 2001 and 2000. They didn't play as well as their opponent did. Trying to find reasons that go deeper than that leads to amateur psychology and the kind of twisted logic that ends up raising baserunning and relay plays to the same importance of hitting for power and throwing strikes. They need to improve--two corner outfielders who can hit and a platoon partner for Eric Chavez, just to start--and doing so would go a long way to correcting the real problem: not scoring enough runs.
"Luck" and "small sample size" haven't taken hold as explanations, but it is imperative that we find a way to communicate the idea that short playoff series, while a necessity in determining a champion, aren't good barometers of relative quality or useful tools in discerning a team's strengths and flaws.
Oh, yes...the LCSs. There weren't many surprises last night in New York. The Red Sox didn't score much off of Andy Pettitte, and Derek Lowe did about what he does in road games. There seems to be some agitation over Grady Little's decision to not save Lowe for Boston, but I don't see that he had good options. The only way to get Lowe and Pedro Martinez two starts apiece in this series without using one on short rest was to start one of them in Game Two.
In Florida tonight, Kerry Wood and Mark Redman are the starters for Game Three of the NLCS. I think this is where we start to see the Cubs' problems; I expect Wood, like Carlos Zambrano, to be a little off, and the Marlins to make the Cubs pay for their 100-year-old outfield. Redman is a bit like Mike Hampton, and the Cubs have yet to show any patience at the plate, so Flailapalooza II is likely. I'll say, Marlins, 5-3, and note that it could also be something like 8-5.
I know the weekend columns have been very popular, and I appreciate all the positive feedback. I will try and do at least one this weekend, but I may have to just come back Monday with a monster. Check the site for updates and your inboxes for a newsletter. And enjoy the weekend!