October 14, 2010
A Brief Meditation on the Power of Sabermetrics During the Postseason
My apologies for giving this piece a title more suited to a Sufjan Stevens song than a baseball article, but I’m a little bit hyper today. Like many of you, I’m struggling through this two-day break before the playoffs begin again on Friday night, featuring an ALCS showdown between the Texas Rangers, coming off their first playoff series victory in franchise history, and the New York Yankees, who’ve won 28 playoff series and nine world championships since the Senators/Rangers franchise came into existence in 1961. Over in the senior circuit, the Giants and Phillies tee it up on Saturday night with one of the more compelling pitching matchups of recent vintage: Tim Lincecum vs. Roy Halladay, who in their two playoff starts have combined for two complete games, one no-hitter, four baserunners allowed, 23 punchouts, and 48 instances of a broadcaster saying “that pitch just wasn’t fair.” I can’t wait, because this is going to be good—in fact, given how exciting (if sloppy) the playoffs have been so far, I’m more excited about them than I’ve been in years.
To me, this two-day gap feels like a deep breath before the plunge, as the arrival of a “real” (i.e., best-of-seven) series means the playoffs have become serious—I’ve never quite understood why a divisional series is best-of-five, since the loser is just as fully required to go home as if it were a full seven, but that’s a discussion for another day. To fill this short breathing space, the interwebs are teeming with series previews, player profiles, predictions, and occasional pablum. Sometimes this analysis will give heavy weight to things like clubhouse chemistry, clutch hitting, veteran steadiness, and other factors that are difficult to quantify. Since you’re reading this at BP, however, you’re probably interested in a more sabermetric take on each series, and I’m here to help. Here’s the best, most informed, most statistically advanced analysis of each series I can give you:
(cue crickets chirping)
OK, that’s not completely true. Colin Wyers, Eric Seidman, and Matt Swartz, our resident Navajo Stat Speakers, have done yeoman’s work putting together game-day PECOTA projections and pitcher previews, and their output has been interesting, entertaining, and informative. But if there’s one overriding principle everyone interested in baseball analysis should always keep in mind, it’s that nearly anything can happen given a small sample. In a series that lasts four-to-seven games, most any batter can produce like Mickey Mantle or Tony Gwynn for a short period of time, just as he can during any given week in June. Even the worst teams in baseball can be world-beaters for a short stretch—as an example, on five different occasions this season, the Pirates won four out of five games.
This holds true for in-game strategies as well. Over the course of a season, a manager prone to calling for a sacrifice bunt in situations where it almost certainly lowers his team’s run expectancy is going to cost that team wins, as will a manager who structures his lineup inefficiently or doesn’t maximize his bullpen usage. Baseball writers are right to point these errors out. Over the course of a week, however, these small inefficiencies have much less opportunity to accumulate and therefore play less of a role in determining who wins a series, as compared to their effect on winning enough games to make the playoffs.
You already know this to be true, of course, as do I, as does Billy Beane, who famously said “my sh*t doesn’t work in the playoffs.” There are those who read that statement and take it to mean that the methods used by more saber-friendly organizations are “tricks” that have value only during the regular season, but are doomed to fail against the best teams in the league once the playoffs start. Rarely do those folks remember, or understand, that the rest of the quote is this: “My job is to get us to the playoffs. What happens after that is … luck.” While “luck” was perhaps an unfortunate choice of words, since it makes it sound like there’s no advantage to being the better team (something that’s patently untrue), Beane’s point was that the market inefficiencies he was trying to exploit could help him leverage enough small advantages over the course of a season to get him into the dance, but once the playoffs start the teams are too evenly matched and the time horizon too short for those things to make a difference. The best teams almost always make the playoffs, but the better team frequently doesn’t win a short series. Dayn Perry and Nate Silver wrote this in Baseball Between The Numbers, after cribbing Beane’s quote for a chapter heading: “That there is a great deal of luck involved in the playoffs is an incontrovertible mathematical fact.” Their attempts to determine the factors that lead to playoff success formed the basis of our Secret Sauce calculation, but even that noble effort hasn’t show any great predictive power.
As I said, everyone reading these words probably know this, since we’ve all been using advanced metrics to help us stay competitive in our fantasy leagues for years now—leagues that usually depend on a full season of baseball, where there’s enough time to apply this knowledge. It’s an easy thing for us to reconcile, despite what may be said by those who are not only disinterested in the statistical side of baseball but can be actively hostile towards it: sabermetrics (for want of a better term) can go a long way towards helping your team’s GM put together a consistent winner over the long haul, and the fact that we may have difficulty predicting the outcome of a single game or a single short series in no way invalidates that fact, regardless of what may be said by those who not only aren’t interested in sabermetrics, but are downright hostile to it.
Once the playoffs start, it’s easy for me to just sit back, relax, and let the games wash over me, to revel in their beauty and drama, their heroes and goats, their nail-biting personal confrontations and otherworldly athletic displays, and everything else that makes baseball such a wonderful game. I can take pleasure in watching the Buddy Biancalanas of the world enjoy their unexpected moments in the sun just as much as (or even more than) I can enjoy watching superstars like Halladay and Lincecum shred opposing lineups, regardless of their respective PECOTA projections. This makes me no different than anyone else reading this, and no different than any other baseball fan. As for my in-depth ALCS and NLCS analysis, I’m picking the Yankees and the Phillies, the teams I thought were the most talented in each league back in March, and nothing since then has changed that opinion. When all is said and done, though talent doesn’t make you a lock, it’s what matters most.