September 20, 2006

Lies, Damned Lies

Secret Sauce

by Nate Silver

With the pennant races settling down more quickly than we might like, it's not too early to start thinking about which teams might have the pole position in the post-season derby. As Dayn Perry and I found in Baseball Between the Numbers, regular season success is no guarantee of playoff performance. Rather, there are three particular characteristics of teams that win more than their share of post-season games. These characteristics are as follows:

• A power pitching staff, as measured by normalized strikeout rate.
• A good closer, as measured by WXRL.
• A good defense, as measured by FRAA.

Of the dozens of team characteristics that we tested for statistical significance, in terms of their relationship with winning post-season games and series, these were the only three that mattered. Ending the year hot doesn't make a whit of difference, for example, nor does having a veteran club, or a smallball offense.

More remarkably, all three of these characteristics relate to run prevention, rather than run scoring. That does not mean that offense is of no importance in the playoffs. But there is a lot of noise in the postseason record, and offense did not produce enough signal to emerge through it. The reasons are too complicated to get into here, but have to do with what happens when good offenses face good pitching. Pitching does have some tendency to dominate these match-ups, whether they occur in the regular season or in the playoffs. Because "plus pitching" versus "plus hitting" duels occur more frequently in the post-season, we tend to notice the effects more then.

In any event, this "secret sauce" is fairly pungent. The two teams that rated most favorably in these categories in the 2005 playoffs were the White Sox and the Astros, who met in the World Series. The formula also predicts the success of some surprise World Series winners like the 1990 Reds and 1979 Pirates. Conversely, of the ten post-season teams since 1972 that rated worst in the "secret sauce" rankings, none advanced beyond their LCS.

It's less controversial to imagine why the "secret sauce" characteristics that we have identified make such a difference. Closers pitch a much larger percentage of a team's innings in the playoffs than they do in the regular season; since 1996, for example, Mariano Rivera has thrown 5.1% of the Yankees' regular season innings, but 10.4% of their post-season innings. And those closers tend to pitch in higher leverage situations, with the preponderance of close contests when good teams get together. (It is the closer specifically who seems to matter; middle relievers are of less importance in the postseason, as both closers and starting pitchers are used more aggressively by their managers.)

Strikeout rate is important for a couple of reasons. Firstly, it is the single most reliable indicator of pitching performance. Pitching is subject to enough luck that it is not uncommon for an entire team, over the course of an entire season, to have an ERA which differs markedly from the quality of its peripheral statistics. Strikeout rate, on the other hand, is quite stable from season to season. Secondly, as we documented in Mind Game, power pitchers tend to have a leg up against power hitters, and there are a lot of power hitters in the robust offenses of playoff-bound clubs.

Defense, finally, is critical to prevent good offenses from putting rallies together. Offense is relatively non-linear; it requires stringing hits and walks next to one another to generate runs. Avoiding crooked numbers early in the game allows for the superior bullpen to win the game later.

In the table below, I've ranked the 30 MLB clubs based on their performance in the three key departments. Strikeout rate is measured by the Davenport-translated strikeout rate, as available on the team DT pages. Although strikeout rate is not highly dependent on park effects, it is rather dependent on league; National League pitchers record strikeouts about 4% more often than their American League counterparts. The DT translations correct for this. WXRL refers to the performance of the closer only, defined as the pitcher who has recorded the most save opportunities for his club over the course of the season. (Entertainingly--or not--Bob Wickman qualifies as the closer of both the Indians and the Braves).

             EqK9    RANK    FRAA   RANK   WXRL    RANK     Composite
---------------------------------------------------------------------
Blue Jays     6.6     5.5     +20     6    5.21      6        17.5
Twins         6.7     2.5      +4     13   5.95      3        18.5
Angels        6.7     2.5      -1    17.5  6.78      1        21
Mets          6.3     10      +17     8    5.37      4        22
Red Sox       6.6      5.5    -13    22.5  6.53      2        30
Rangers       6.0     14      +23     5    2.88      16       35
Astros        6.4      9      +27     2    0.51      24       35
Yankees       5.9     18.5     +6    11    5.18      7        36.5
Mariners      6.0     14       -3    19    5.23      5        38
White Sox     5.9     18.5    +14     9    3.78      11       38.5
Phillies      6.5      8       +0    16    3.27      15       39
Tigers        5.9     18.5    +26     3    2.21      18       39.5
A's           5.8     23      +18     7    3.49      13       43
D'Backs       6.6      5.5     +2    15    0.70      23       43.5
Padres        5.7     26       +5    12    4.97      8        46
Dodgers       5.8     23       +3    14    4.52      9        46
Rockies       5.7     26      +31     1    1.89      19       46
Cubs          7.1      1       -1    17.5 -0.97      28       46.5
Orioles       6.0     14      -18    25.5  3.87      10       49.5
Pirates       6.2     11.5    -30    28    3.46      14       53.5
Reds          6.2     11.5    -10    21    1.74      21       53.5
Giants        5.7     26      +24     4    0.02      25       55
Brewers       6.6      5.5    -14    24   -1.15      29       58.5
Braves        5.9     18.5    -13    22.5  1.77      20       61
Cardinals     5.3     29.5    +12    10    1.04      22       61.5
Marlins       5.8     23      -18    25.5  2.39      17       65.5
Nationals     5.3     29.5    -29    27    3.66      12       68.5
D-Rays        5.9     18.5    -40    30    0.02      26       74.5
Indians       5.9     18.5    -33    29   -0.86      27       74.5
Royals        5.6     28       -7    20   -1.21      30       78

I've already spoiled much of the suspense, but let's look at the secret sauce performance of the eight teams that project to qualify for the playoffs as of this morning:

• The Twins ranking does not--repeat, does not--include Francisco Liriano. But even without Liriano, the Twins have a very good pitching staff. It all begins with Johan Santana, of course, who is a couple of degrees of order better than any other post-season bound starting pitcher. But it extends to Boof Bonser, who has quietly put together one of the better second halves in the league, as well as the bullpen. The Twins' defense, which was so shaky early in the season, also rates as an asset, as is consistent with the defensive reputation of most of the starters (it has also been a blessing, of course, to get Tony Batista and Shannon Stewart out of the lineup). In short, the Twins are poised to make some noise.
• The Queens Gothams would seem to bear little resemblance to the Piranhas from the Prairie, but in some ways the Mets and Twins are similar clubs. Both teams have excellent closers, and defensive assets all over the field. Each will live and die based on the health and performance of its #1 starter. And both may be bound for the World Series.
• I've written before that this Yankees club is the strongest iteration of the dynasty since 1998. Much of that assessment is tied up in the powerful offense, which this method will neglect. In the run prevention department, the Yanks are a mixed bag. They do have an adequate defense for the first time in years, as well as Mariano Rivera. But pitchers like Chien-Ming Wang aren't usually the type to find post-season success. The catch is that the three AL teams the Yankees are most likely to face in the regular season aren't power hitting clubs; if the White Sox can pull out a miracle, on the other hand, they might give the Yankees real trouble.
• The good news for the Tigers is that performance down the stretch run bears little relationship with post-season success. The bad news is that they still don't make for all that great a playoff team. While the defense is fantastic, the starting pitchers don't miss enough bats to have an advantage against teams like the Yankees, and they have the wrong man installed as closer--replacing Todd Jones with Joel Zumaya would move them ahead of the Yankees.
• The Athletics have gotten only 35 innings of work out of Rich Harden this season, and deserve to rank somewhat better than this if he is healthy (and effective) in October. On the other hand, the absence of Bobby Crosby really weakens the defense; if there's one team that really could have used Neifi Perez, it's the A's. Oakland has pulled wins out of nowhere all season long--if I had told you before the season everything that would go wrong with the A's, you'd probably have guessed 70-92, rather than 92-70. They'll need to continue to do so in the playoffs.
• Has there ever been a team more average than the Padres? All eight prospective post-season position regulars have an OPS between .750 and .844, while all four projected post-season starting pitchers have an ERA between 3.63 and 4.25. Yes, the Padres have Trevor Hoffman, but this is not the sort of top-heavy club that the playoff schedule tends to favor.
• I've been touting the Dodgers as a sleeper team, but Chavez Ravine masks some pretty deep problems on the starting pitching staff, and their up-the-middle defense can be clunky at times. Word is that Grady Little is seeking an injury exemption to sneak Kirk Gibson onto the playoff roster.
• The Cardinals, finally, are a truly poor post-season club. Chris Carpenter and the defense are assets, but this is much the same group of pitchers that was destroyed by the 2004 Red Sox, and we're likely to see a lot of results like this one. Losing Jason Isringhausen, at least, figures to be a blessing in disguise.

Here, ladies and gentlemen, are your "secret sauce" playoff results, based on projected matchups as of this morning:

Twins 3, Yankees 1
Tigers 3, A's 2

Mets 3, Dodgers 0
Padres 3, Cardinals 1

Twins 4, Tigers 2
Mets 4, Padres 1

Twins 4, Mets 3

Twins-Mets would be a lot of fun, particularly given the gamesmanship over how best to deploy Santana and Pedro Martinez. That said, if my formula is disproven by a team like, say, the Tigers, I wouldn't be too upset.

Nate Silver is an author of Baseball Prospectus. 
Click here to see Nate's other articles. You can contact Nate by clicking here