PDA

View Full Version : The Postseason Secret Sauce



D-Man
08-27-2010, 11:26 PM
A few years ago, Nate Silver introduced the idea of the special ingredients that go into a postseason run. Of course, he wasn’t the first analyst to examine what drives postseason success; however, his analysis was exhaustive, data-driven, and interesting. I won’t belabor Silver’s study on the “Secret Sauce” here; instead, I’d like to focus on the implications of his work and what that means for the Reds’ potential postseason chances. [For the jinx police: please note that I was careful to qualify postseason chances as potential.]

For those interested in the Secret Sauce, see any of these articles:
http://www.baseballprospectus.com/article.php?articleid=5541
http://www.baseballprospectus.com/unfiltered/?p=548
http://nymag.com/guides/fallpreview/2009/sports/58501/

Silver’s most significant finding was that regular season offense performance is irrelevant to postseason success. Here’s what Silver said about why strong offenses really don’t matter in the postseason:


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.

This conclusion may be unsettling to some people. . . Offense is half of the game, and to suggest that it doesn’t correlate with postseason winning is preposterous. However, the postseason is littered with great offensive teams that have had limited postseason success. Recent history provides several examples: the 2006/2007 Yankees (the modern-day Murderer’s Row), the 2005 Red Sox, early 2000s A’s, and mid-90’s Indians. Even the 2001 Mariners (927 runs and 116 wins!) failed to reach their postseason expectations. So be prepared to toss aside the Reds’ league leading offense. If the Reds were to traverse the regular season gauntlet (back off, jinx police), the offense would a big reason why they may make the postseason. For purposes of postseason chances, the offense is largely irrelevant. Sad to say.

Instead, Silver’s research indicates that three run-prevention attributes—the key elements of the postseason “Secret Sauce”—reflect how well a team performs in postseason:

• A good defense, as measured by Fielding Runs Above Average (FRAA).
• A power pitching staff, as measured by normalized strikeout rate.
• A good closer, as measured by WXRL.

Baseball Prospectus tracks these three measures on its stats page. The current results for the Reds paint an overall picture that isn’t pretty: the Reds are 6th, 22st, and 20th respectively in those three categories. Overall, the Reds are 18th in the majors and rank behind the acclaimed Cubs, A’s, Angels, Mets, and Marlins in possessing the ingredients of postseason secret sauce. Most notably, the Reds are *last* among NL challengers and 9th among the 11 MLB contenders. That’s hardly encouraging.

How much can we trust these measures, and how likely are they to change before the end of the year? Most importantly, what can the Reds do to optimize their chance of winning in the postseason?

Defense
Clearly, the defense is outstanding. If Rolen and Phillips are healthy, then the Reds can go to the bank knowing that the defense would be a difference maker.

Likewise, Reds fans should be prepared to throw their shoes at the TV (or in the direction of the dugout, if you can get tickets) if Gomes were to start any postseason game in LF. Giving away outs--even in left field--often results in early postseason exits.

Strikeouts
The starting pitcher element is trickier. The Reds are lacking a strikeout ace among the rotation candidates. Cincinnati’s starters have survived via rotation depth, great defensive plays, a potent offense, and grinding out innings. Cueto is probably the Reds nominal ace, but his K rates are below his career numbers, and it’s likely he will hit the wall in September. Arroyo is the staff innings-eater, but his K rates are painfully low and his peripherals are scary for someone with a sub-4.00 ERA. Leake reached his wall in June. Harang hasn’t pitched in months and may have reached the end of his rope. Wood has been the team’s best pitcher for the past month or so (San Fran start notwithstanding), but he’s a rookie. And he has only pitched one game in Cincinnati, which warrants mentioning with the significant home/road splits for nearly everyone in the rotation. Bailey and Volquez have both shown some promising signs and provided mixed results; both pitchers are coming off injuries too. In total, this rotation doesn’t paint a pretty picture for October success. Not good at all.

The wild card in this scenario is that we've learned how difficult it is for the staff to pitch in the GABP during the dog days of July-August. But don't know how the stadium will play in October games at night. Consequently, the threat of the easy taters may be lessened in the postseason, and strikeouts may be easier to come by. Who knows?

My position is that the Reds should roll the dice with those pitchers most likely to pitch an outstanding October baseball game, even at the risk of embarrassing the club with a bad performance. Go big or go home. I think the front three should include Cueto, Bailey, and Volquez. All three have thrown in a few stinkers in the past week. . . Nevertheless, they have the best “stuff” and would be the most likely candidates to dominate on any given night. Arroyo should start a game 4, if the Reds need a fourth starter. It will be hard to pass up the veterans (Arroyo and Harang) for a postseason rotation slot, but I think it’s the best shot the Reds would have of advancing in the postseason.

Closer
You probably don’t need me to remind everyone that Cordero hasn’t had a good year. To his credit, he has looked much better in his past 6-7 games. Let’s hope he’s made the appropriate adjustments and is turning the corner.

If not, then the Reds need to get creative with the bullpen. It’s interesting that Silver’s research suggests the rest of the bullpen doesn’t matter as much as the closer, but that’s probably a function of how closers are used in the postseason (Mariano Rivera as the archetype) rather than a reflection of the importance of middle relievers. It's doubtful that Dusty use the Reds best relievers (i.e., Massett and Rhodes) in the same way that the Yankees have used Rivera. A reasonably good approach might be to employ a short leash with the starters and bring in Masset and Rhodes to snuff out the high-leverage situations. Perhaps Chapman can also fill a similar role when he is called up, but relying on a rookie with a month of big league experience might be too much.

redsfandan
08-27-2010, 11:32 PM
Thousand Island dressing?

Seriously, I'll disagree on this part:

My position is that the Reds should roll the dice with those pitchers most likely to pitch an outstanding October baseball game, even at the risk of embarrassing the club with a bad performance. Go big or go home. I think the front three should include Cueto, Bailey, and Volquez. All three have thrown in a few stinkers in the past week. . . Nevertheless, they have the best “stuff” and would be the most likely candidates to dominate on any given night. Arroyo should start a game 4, if the Reds need a fourth starter. It will be hard to pass up the veterans (Arroyo and Harang) for a postseason rotation slot, but I think it’s the best shot the Reds would have of advancing in the postseason.

Gotta have Arroyo as one of the projected postseason starting pitchers instead of Volquez. But that's just me.

D-Man
08-27-2010, 11:43 PM
Gotta have Arroyo as one of the projected postseason starting pitchers instead of Volquez. But that's just me.

Sadly, with Arroyo's 4.7 K/9, he's likely to end up with a postseason performance reminiscent of Chien-Ming Wang (ERA north of mid-7).

That doesn't mean that I would choose Volquez as an ideal postseason starter, not by any stretch. I just don't see the Ks coming from anyone else.

Rotation depth is nice in August, but it's irrelevant in October.

hebroncougar
08-27-2010, 11:45 PM
How'd the Cards do with their studs last year? Those 2 "should" define what makes a staff successful in the post season. As should the big 3 of Atlanta for years. It's all who's hot at the right time. There is no "secret sauce".

paulrichjr
08-27-2010, 11:53 PM
Secret sauce - Someone coming out of nowhere and dominating

Reds 2010 secret sauce = Chapman

IslandRed
08-27-2010, 11:58 PM
Gotta have Arroyo as one of the projected postseason starting pitchers instead of Volquez. But that's just me.

I agree with that. I get the Secret Sauce and agree with much of it, but it's one thing to compare teams and isolate the K/9 as a success factor, but quite another leap to suggest pulling a pitcher that's more consistently effective in favor of another pitcher solely because of K/9.

D-Man
08-28-2010, 12:03 AM
How'd the Cards do with their studs last year? Those 2 "should" define what makes a staff successful in the post season. As should the big 3 of Atlanta for years. It's all who's hot at the right time. There is no "secret sauce".

I think you're missing a few key points:

*If you read the third article cited above (nymag one), Silver had the 2009 Dodgers as the 2nd best team in the postseason, better than the Cards. By inference, the model probably predicted the Dodgers to beat the Cards, and they did. Remember, it's not postseason aces--it strikeouts per inning, closers, and defense.

*The importance of what Silver did is that he didn't focus on a specific example (e.g. the 2009 Cards lost but should've won); instead, he gathered 30 years of data, added it up, tested some theories, and looked at the big picture. What were the prevailing trends? That's what he's sifted through for our benefit.

traderumor
08-28-2010, 12:07 AM
All well and good, but it really isn't like there is anything teams can do about it, even if its absolutely true. What it takes to survive the 162 game grind may not play well in postseason, but you have to do that to get there in the first place. The Cards may very well be a perfect example of a very dangerous postseason team with Wainwright and Carpenter that is very possibly not going to make the postseason.

Brutus
08-28-2010, 12:09 AM
I think you're missing a few key points:

*If you read the third article cited above (nymag one), Silver had the 2009 Dodgers as the 2nd best team in the postseason, better than the Cards. By inference, the model probably predicted the Dodgers to beaten the Cards, and they did. Remember, it's not just postseason aces--it strikeouts per inning, closers, and defense.

*The importance of what Silver did is that he didn't focus on a specific example (e.g. the 2009 Cards lost but should've won); instead, he gathered 30 years of data, added it up, tested some theories, and looked at the big picture. What were the prevailing trends? That's what he's sifted through for our benefit.

Those things are definitely important. And I think there's an element of truth to what Silver said (his natural slant toward Baseball Prospectus aside)... but at the same time, the playoffs are also a giant crapshoot. In a short series, especially a 5-game series, anything can happen between two good teams.

D-Man
08-28-2010, 12:11 AM
I agree with that. I get the Secret Sauce and agree with much of it, but it's one thing to compare teams and isolate the K/9 as a success factor, but quite another leap to suggest pulling a pitcher that's more consistently effective in favor of another pitcher solely because of K/9.

Is it a leap? Really?

Let me put it a different way. Arroyo has started 6 games against likely playoff teams (Padres, Braves, and Cards). In 26 innings (less than 5 innings per start), he's surrendered 24 runs.

http://espn.go.com/mlb/players/splits?playerId=4416

Tornon
08-28-2010, 12:12 AM
Jeff Suppan absolutely dominated in the 2006 postseason. If Suppan can have that much success, I'd hope Arroyo could at least be a viable starter. Arroyo has struggled in the postseason in his career, but that was in the AL and he's matured as a pitcher since then

redsfandan
08-28-2010, 12:16 AM
Those things are definitely important. And I think there's an element of truth to what Silver said (his natural slant toward Baseball Prospectus aside)... but at the same time, the playoffs are also a giant crapshoot. In a short series, especially a 5-game series, anything can happen between two good teams.
Yep, that's so true.

Is it a leap? Really?

Let me put it a different way. Arroyo has started 6 games against likely playoff teams (Padres, Braves, and Cards). In 26 innings (less than 5 innings per start), he's surrendered 24 runs.

http://espn.go.com/mlb/players/splits?playerId=4416
That's a valid point about what Arroyo has done vs possible playoff teams. But, I'm one that thinks that k/9 isn't quite as important as k/bb and I just don't think we should expect too much from Volquez this soon.

D-Man
08-28-2010, 12:27 AM
Here's an element that I find particularly telling, re: the worst ranking teams. . .


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.

redsfandan
08-28-2010, 12:33 AM
It's "pungent"? Does it need some deodorant?? :confused: What the ...

IslandRed
08-28-2010, 12:42 AM
In 26 innings (less than 5 innings per start), he's surrendered 24 runs.

He got bombed in his one start against San Diego, true enough, but San Diego isn't going to the playoffs because of their offense. He's pitched better against better. That rocking aside, he's done as well in his five starts against Atlanta and St. Louis as anyone else on our staff is likely to do. I certainly don't think today's Volquez gives us a better chance to win a playoff game. Next year's, probably, but not today's. And here's why. While strikeout rate is a component of success in the Secret Sauce, here's a stat I bet would show up in a reverse study looking to isolate reasons teams fail in the postseason: walking too many batters.

blumj
08-28-2010, 12:43 AM
Jeff Suppan absolutely dominated in the 2006 postseason. If Suppan can have that much success, I'd hope Arroyo could at least be a viable starter. Arroyo has struggled in the postseason in his career, but that was in the AL and he's matured as a pitcher since then

Arroyo only made 2 postseason starts that I remember, one was excellent, vs. LAA, the other extremely bad, vs. NYY. He did a very good job in relief against mostly RHHs in '03-'04, just had one really awful relief appearance in '05.

Brutus
08-28-2010, 12:45 AM
Here's an element that I find particularly telling, re: the worst ranking teams. . .

Baseball Prospectus is known for trying to form the questions to best fit the answers. Sounds like Silver took some liberties to find a study that supported what he already hoping was the case. The BP people love to do anything that makes their own 'stats' look good.

Nothing against Silver. I respect his work... but the postseason is too volatile to really affirm anything too strongly.

Are closers, defensive players and terrific starting pitching important? Absolutely. But we already knew that.

D-Man
08-28-2010, 01:04 AM
He got bombed in his one start against San Diego, true enough, but San Diego isn't going to the playoffs because of their offense. He's pitched better against better. That rocking aside, he's done as well in his five starts against Atlanta and St. Louis as anyone else on our staff is likely to do. I certainly don't think today's Volquez gives us a better chance to win a playoff game. Next year's, probably, but not today's. And here's why. While strikeout rate is a component of success in the Secret Sauce, here's a stat I bet would show up in a reverse study looking to isolate reasons teams fail in the postseason: walking too many batters.

In addition to the SDG massacre, Arroyo also had two miserable starts against the Cards (mixed in with two excellent spots against STL). In total, that's three great starts and three bombs against playoff-caliber clubs. When I look at it in the aggregate (>8 runs per 9, 50% bomb rate), Arroyo isn't going to provide a very good chance to win in the postseason. Moreover, pitchers with his profile (low Ks, bad peripherals) are usually the ones who get smacked around in the postseason.

Fair enough re: walks.

P.S. I should make it clear that I don't view Volquez as the answer to the postseason rotation. I see him as the "least bad" option on the table, warts and all.

D-Man
08-28-2010, 01:08 AM
Arroyo only made 2 postseason starts that I remember, one was excellent, vs. LAA, the other extremely bad, vs. NYY. He did a very good job in relief against mostly RHHs in '03-'04, just had one really awful relief appearance in '05.

Overall postseason ERA of 7.41. Here are the game logs:

http://www.baseball-reference.com/players/gl.cgi?n1=arroybr01&t=p&post=1

redsfandan
08-28-2010, 01:27 AM
Overall postseason ERA of 7.41. Here are the game logs:

http://www.baseball-reference.com/players/gl.cgi?n1=arroybr01&t=p&post=1
That ONE bad start, out of only two postseason starts, kinda skews things a bit doesn't it? Looks like a small sample size to me. Plus it was 6 years ago. I get that you don't like Arroyo that much. Could that be influencing your preference of Volquez?

D-Man
09-29-2010, 01:07 PM
Bump, for obvious reasons. (Yay!!!)

For the skeptics of the secret sauce, I found a rather recent blog post suggesting that the three elements of this theory have not been statistically significant since ~2002. See here:

http://www.rationalpastime.com/2010/09/weak-sauce-secret-sauces-predictive.html

Nevertheless, my take is that the Reds should apply a few general principles of the secret sauce to the postseason:

1.) The Reds need someone to perform like an strikeout ace. Volquez appears to have earned a rotation turn, and he is the most likely to get strikeouts. He is probably the key to the Reds postseason chances. My preferred rotation would be Cueto, Volquez, Bailey, and Arroyo if a 4th starter is needed (presumably, Wood has reached his inning ceiling.)
2.) Baker needs to have a quick hook with all starters. He's been great in managing the personalities of the team (e.g., giving questionable players a long leash throughout the regular season), but he needs to take a no-nonsense approach throughout the playoffs.
*Defense is key. The Reds need to rest Rolen and Philips this week, and need to be judicious in deploying Gomes (middle-inning defensive replacements).
*The Reds need their best relievers pitching in the high-leverage innings. Rhodes, Chapman, and Masset need to be pitching in nearly every game where a reliever is needed. Again, Baker needs to take a no-nonsense approach with Coco.