Print and Go Back Page 2 [Print without images]

Thursday, March 30, 2006
Updated: April 4, 7:54 PM ET
Why don't the A's win in October?

Special to Page 2

The following chapters are adapted from "Baseball Between the Numbers: Why Everything You Know About the Game is Wrong." Copyright (c) 2006 by Baseball Prospectus. Excerpted by permission of Basic Books, a member of the Perseus Books Group. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.

From chapter 9-3: Why Doesn't Billy Beane's S*** Work in the Playoffs?
by Nate Silver and Dayn Perry

With Billy Beane as their general manager, the Oakland A's have earned the lavish praise of the statistically inclined for succeeding through the use of quantitative analysis and despite tightly constrained budgets. From 2000 through 2003, the A's averaged 98 wins per season and made four consecutive postseason appearances despite annual payrolls that ranked twenty-fifth, twenty-ninth, twenty-eighth, and twenty-sixth in the major leagues, respectively.

Yet not only have Beane's teams never advanced past the first round of the playoffs, but they've lost all of their series in galling fashion. In 2000, defensively challenged center fielder Terrence Long lost a flyball in the sun, allowing the Yankees to score a critical run; the A's lost the series 3 games to 2. In 2001, Jeremy Giambi infamously failed to slide and was tagged out at the plate after a Derek Jeter no-look relay flip. In 2002, Billy Koch gave up a home run to the Twins' A. J. Pierzynski in the ninth inning that provided the margin of defeat in the deciding game of the American League Division Series. The following year, Oakland allowed the Red Sox to come back from down 2 games to none to win in 5, taking the final 3 games of the series by a total of 4 runs. To hear some tell it, the A's have been serial victims of bad luck.

Others argue that recent A's teams have been improperly constructed for success in the postseason. Beane himself seems flummoxed over their failures. As he said in the best-selling book Moneyball, "My sh*t doesn't work in the playoffs." Baseball analysts have had a similar reaction. While we have exhausted a lot of energy debating what doesn't work in the playoffs, we have thrown up our hands when it comes to explaining what does work.

Click here to buy "Baseball Between the Numbers" by the Baseball Prospectus team of experts.

Baseball is very hard to predict. Every supposed rule -- "defense wins championships," "you have to play Smallball to win," "an experienced team performs better in the playoffs" -- has far too many exceptions. (For counterexamples, we need look no further than the last three World Champions. The 2003 Marlins rated as a below-average defensive club, the 2004 Red Sox disdained Smallball, and the 2005 White Sox had almost no postseason experience.)

That there is a great deal of luck involved in the playoffs is an incontrovertible mathematical fact. While it takes 162 games to sort out who is the better team in the regular season, a playoff series can be decided in as few as 3 games. Moreover, while there are both good and bad teams in the regular season, those teams that reach the playoffs are closely bunched together in ability level.

Under the current playoff format -- which requires three series wins to claim the World Series title -- it is rare for any team to have more than a 25 or 30 percent chance of winning the whole enchilada. This is something that first-place-or-bust owners like George Steinbrenner would do well to keep in mind. Billy Beane's A's didn't merely fail to win it all -- they lost four straight times in the ALDS. While their opponents weren't pushovers, the A's were favored based on regular-season winning percentage on all four occasions.

The probability of the A's losing all four of these series consecutively is 3.5 percent, or odds of about 27-to-1 against. Since we're only talking about four playoff series, we've got a small sample size that limits our ability to extrapolate meaning from these losses. But a 27-to-1 shot is unlikely enough that it's still worth exploring whether there was something in particular about the A's that made them less equipped for postseason play than their regular-season record would suggest. The metric that we'll use is Playoff Success Points (PSP), a handy invention that assigns credit to teams as follows:

• 3 points for making the playoffs
• 3 points for winning the LDS
• 4 points for winning the LCS
• 4 points for winning the World Series
• 1 point for each postseason win
• -1 point for each postseason loss

More from "Baseball Between the Numbers"
Is David Ortiz really a clutch hitter?

Does baseball need a salary cap?

The highest possible PSP is 25, for a team that sweeps through all 11 postseason games. The lowest is 0, for a team that gets its 3-point spot for making the playoffs, but fritters it away by being swept out of the LDS in 3 games. The system is intuitively consistent. The "worst" possible World Series team, a team that plays the maximum number of games in each of its postseason series, receives 17 points. Meanwhile, a team that sweeps through the LDS and LCS but loses the World Series in 7 games receives a PSP of 16.

Since 1972, there have been twenty-seven teams that made the postseason in spite of having below-average offenses. Of these, seven won the World Series: the 1985 Royals, 1987 Twins, 1990 Reds, 1995 Braves, 1996 Yankees, 2000 Yankees, and 2005 White Sox. All of these teams, except the 1987 Twins, had excellent pitching staffs; it's hard to make the playoffs with a below average offense unless you have an excellent pitching staff.

Conversely, twenty teams have made the postseason with below average run prevention. None of them won the World Series, and only two (the 1982 Brewers and 1993 Phillies) even played for the championship. Sixteen of the twenty lost the first playoff series in which they played.

Does this mean that defense really does win championships after all? The short answer is yes, probably.

Run production as a whole hasn't had much relationship with playoff success. Neither have any of the individual offensive metrics. The A's postseason struggles have sometimes been attributed to their tendency to rely on walks and home runs, but there is no evidence that teams that play Smallball instead fare better in the postseason. Although stolen-base attempts have a slight (but statistically insignificant) positive relationship with PSP, sacrifice-hit attempts have a slight negative one. Speed Score, a composite of five different offensive statistics that provides evidence about a player's wheels, has no relationship with PSP at all. Nor do teams that hit well in the clutch in the regular season see that advantage carry forward into the playoffs.

There is a lot more to look at when it comes to pitching and defense, though. The relationship between the performance of the closer and PSP is quite strong -- stronger in fact than when we look at the bullpen as a whole. Put differently, the performance of non-closer relievers is of very little importance in the postseason. Between the absence of fourth and fifth starters who often require bullpen help and the willingness of managers to stretch their closers into multiple-inning stints, it is the secondary relief pitchers who get squeezed in the playoffs.

Of all the statistics in our study, the one with the highest correlation to postseason success is opponents' batting average. Certainly, preventing hits is very important in the playoffs -- when you're matched up against good offensive clubs, it's vital to stop them from stringing together hits and starting rallies. But we also need to think about how a team can go about preventing hits. The best ways to prevent hits are: (1) to strike the batter out, so that he doesn't put the ball into play in the first place and (2) to catch the ball when it is put into play.

The other interesting finding is that avoiding walks doesn't seem to have much relationship with playoff success. A good rule of thumb for pitchers is: Let bad hitters beat themselves, but don't let good hitters beat you. While walking a hitter who can't hit the ball out of the infield is tragic, pitching around a guy who can slam the ball 450 feet can be advisable. Pitchers encounter a lot more of the latter kind of hitter than the former in the playoffs. Finally, we see that the two defensive metrics examined here -- Baseball Prospectus' Fielding Runs Above Average (FRAA) and unearned runs -- show a promising correlation with PSP.

After any number of permutations of the twenty-six variables in our database, we identified three factors that have the most fundamental and direct relationship with Playoff Success Points. These variables are as follows:

• Closer's performance
• Pitcher strikeout rate
• Defense

Striking batters out, catching the ball, and having a good closer wins championships. It makes a good deal of sense why each of these variables is so important in the postseason. The following table shows the rankings of the top ten and bottom ten playoff teams in our sample of 180 teams. Note how the worst performers fare poorly in all three of the key categories.

Highest Composite Ranking
Team Closer FRAA K Rate Average Result
1. 1979 Orioles 31 9 48 29.3 Lost in World Series
2. 1990 Reds 38 15 49 34.0 Won World Series
3. 2001 D'Backs 37 61 10 36.0 Won World Series
4. 1998 Yankees 23 72 10 37.3 Won World Series
5. 1984 Tigers 1 63 51 38.3 Won World Series
5. 1984 Tigers 1 63 51 38.3 Won World Series
6. 1979 Pirates 20 29 70 39.7 Won World Series
7. 1979 Mets 36 32 63 43.7 Lost in NLCS
8. 2002 Angels 24 5 106 45.0 Won World Series
9. 1995 Indians 12 62 67 47.0 Lost in World Series
10. 1978 Yankees 53 47 47 49.0 Won World Series

Lowest Composite Ranking
Team Closer FRAA K Rate Average Result
180. 1997 Giants 167 152 167 162.0 Lost in NLDS
179. 2005 Red Sox 164 141 131 145.3 Lost in ALDS
178. 1989 Blue Jays 120 166 144 143.3 Lost in ALCS
177. 2000 A's 155 107 150 137.3 Lost in ALDS
176. 1974 Orioles 66 168 173 135.7 Lost in ALCS
175. 1995 Rockies 161 109 136 135.3 Lost in NLDS
174. 1981 Expos 118 173 111 134.0 Lost in NLCS
173. 1996 Orioles 160 138 102 133.3 Lost in ALCS
172. 1981 Brewers 143 83 172 132.7 Lost in ALCS
171. 1979 Reds 91 150 151 130.7 Lost in NLCS

What does this mean for Billy Beane's A's? We see that one of his clubs, the 2000 version, ranked fourth from the bottom in composite ranking. This team featured a young offense with budding stars like Eric Chavez, Miguel Tejada, and Jason Giambi, coupled with a cheap veteran pitching staff. The starter in the decisive ALDS Game 5 was Gil Heredia, a finesse pitcher who treaded water by avoiding mistakes and doing just enough to give his offense a chance. However, he was no match for the Yankees, who crushed him for 6 runs in one-third of an inning. Meanwhile, A's closer Jason Isringhausen had yet to really establish himself. And while the infield defense was reasonable, the outfield of Ben Grieve, Terrence Long, and Matt Stairs was one of the worst defensive groups in recent memory. Things got a little bit better in the seasons thereafter. The Big Three of Barry Zito, Tim Hudson, and Mark Mulder established themselves, and Billy Beane, learning from his experience, began to pay more attention to relief pitching and outfield defense.

By 2003, the A's had become a legitimately strong postseason club, but they still succumbed in 5 heartbreaking games to the Red Sox. It was neither bad luck nor a design ill suited for postseason play that did the Athletics in but a combination of the two. Miguel Tejada and Keith Foulke left after 2003, and Tim Hudson and Mark Mulder the year after. Young shortstop Bobby Crosby and 2005 Rookie of the Year closer Huston Street have since emerged. But in 2004 and 2005 at least, Billy Beane's window of opportunity had closed.

Next: Does baseball need a salary cap?