Baseball Prospectus "Secret Sauce" - Prediction on Postseaso (1 Viewer)

[wilked]

Some of you might remember the discussion last year, check it here

http://forums.footballguys.com/forum/index...=348032&hl=

Basically, last year BP said "The Red Sox, in fact, might be one of the best Secret Sauce teams of all time, ranking in the Top 3 in all three categories."

We all know what happened in the playoffs

For those not in the know, the secret sauce attempts to break down the elements that make a difference in the postseason. It has found the following to have a correlation to postseason performance:

Adjusted Strikeout Rate
Fielding
Closer performanceBeing strong in these categories does not guarantee success in the playoffs, but the numbers suggest that it gives an edge.

That said, here is the list of this year's Secret Sauce rankings (surprisingly, with the same #1 and #2 as last year)

http://www.baseballprospectus.com/statisti....php?cid=280104

1. Red Sox

2. Cubs

3. Toronto

4. Angels

Anyway (and I am looking at you Capella), this is not a Ra-Ra Red Sox post, but moreso to generate discussion. As most know I like these statistic-based analyses, and in particular find this interesting as it goes against the Moneyball / Beane line of the postseason being a complete crapshoot

 

[wilked]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

 

[Sonny Lubick Blowup Doll]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

So where do the rest of the playoff teams rank?

 

[Sonny Lubick Blowup Doll]

Poor Toronto...

 

[wilked]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

So where do the rest of the playoff teams rank?

Not quite sure what you are asking here...

 

[gump]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

Where do the Sox and Cubs rank on this since-1972 list?

 

[Capella]

Manufactured stat.

 

[gump]

Looks like the WXRL is the Rays weakness. I'm guessing that has something to do with the closer listed who we will hopefully not see close again.

 

[Sonny Lubick Blowup Doll]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

So where do the rest of the playoff teams rank?

Not quite sure what you are asking here...

Nevermind, I hadn't gone to the link. KC is fifth.

 

[wilked]

Manufactured stat.

it's not a stat, but rather a correlationWhat do you mean by manufactured? They took a number of stats, tested them against postseason success, and found 3 that correlate. Not sure why you think that is manufactured?

 

[wilked]

Pulled this from last year's discussion, but I think it is good to remind people that this stuff isn't hogwash

Since 1972, out of the Top 10 teams with the highest "Secret Sauce" ranking, 7 won the WS. 2 lost in the WS, and one lost in the NLCS to another Top 10 teamOf the Bottom 10 teams, none of them have reached the WS.

So where do the rest of the playoff teams rank?

Not quite sure what you are asking here...

Nevermind, I hadn't gone to the link. KC is fifth.

Ya, it is not predicting who will get to the postseason, but who will have an edge in the postseason. KC has a killer closer, pitchers who can strike a guy out, and a good defense. That does not account for the face that they can't hit the broad side of a barn though...

 

[Your Mother]

I buy into this, although I do think there are a couple flaws. One, I don't think adjusted K rate is modified to account for shortened rotations. Two, I can believe that a closer is important, but I have a hard time believing that bullpen rank is more important than closer rank.

It isn't a "manufactured" statistic, but it's not a silver bullet either.

The whole rundown is in the indispensable "Baseball Between The Numbers", in a chapter entitled "Why Doesn't Billy Beane's #### Work in the Playoffs", or something to that effect.

 

[blackjack23]

Does this "Secret Sauce" take injuries into account?

 

[The Man from Laramie]

As most know I like these statistic-based analyses, and in particular find this interesting as it goes against the Moneyball / Beane line of the postseason being a complete crapshoot

Not really. If you read the initial analysis from the Baseball Prospectus book, they still said luck and random chance (or put more accurately, those factors not otherwise considered by the team) was the biggest factor in explaining postseason success. IIRC, it was a bigger factor than the above three metrics. I think those metrics might have combined for 27% of the explanation.EDIT: I just looked it up. It said something along the lines of "it would be misleading to describe these three factors as a secret sauce. In fact, they only account for explaining 11% of the success." They looked at 26 separate factors, and these three had the largest effect, accounting for 11%. I would imagine luck probably accounts for ~75% of playoff success.

 

[Your Mother]

As most know I like these statistic-based analyses, and in particular find this interesting as it goes against the Moneyball / Beane line of the postseason being a complete crapshoot

Not really. If you read the initial analysis from the Baseball Prospectus book, they still said luck and random chance (or put more accurately, those factors not otherwise considered by the team) was the biggest factor in explaining postseason success. IIRC, it was a bigger factor than the above three metrics. I think those metrics might have combined for 27% of the explanation.EDIT: I just looked it up. It said something along the lines of "it would be misleading to describe these three factors as a secret sauce. In fact, they only account for explaining 11% of the success." They looked at 26 separate factors, and these three had the largest effect, accounting for 11%. I would imagine luck probably accounts for ~75% of playoff success.

Right. It's not a complete crapshoot.

 

[kaa]

I still say the MLB playoffs is too much of a crapshoot to warrant playing 162 games. My continued recommendation is to weight each playoff series according to won-loss differential. Something like:

If team A won 5-9 more games in the regular season than team B

In a 7 game series, team A needs 3 wins and team B needs 4.

In a 5 game series, team A needs 2 wins and team B needs 3.

If team A won 10+ games in the regular season than team B

In a 7 game series, team A needs 3 wins and team B needs 5.

In a 5 game series, team A needs 2 wins and team B needs 4.

I think this is a fair compromise between those who want a 162-game season to have a lot of meaning and those who want a lot of playoff teams. You can let more teams in the playoffs, but they are going to have the decked stacked against them when they play top teams.

 

[[icon]]

You're insane BGP.

 

[Sammy3469]

I buy into this, although I do think there are a couple flaws. One, I don't think adjusted K rate is modified to account for shortened rotations. Two, I can believe that a closer is important, but I have a hard time believing that bullpen rank is more important than closer rank.

It isn't a "manufactured" statistic, but it's not a silver bullet either.

The whole rundown is in the indispensable "Baseball Between The Numbers", in a chapter entitled "Why Doesn't Billy Beane's #### Work in the Playoffs", or something to that effect.

Why? Basically its saying that pitching 0s the 7th and 8th are more important than doing so in the 9th...which makes sense since most starters don't get out of the 6th. Basically the rest of the bullpen pitches twice as much as the closer unless you have starters who go deep into the game.

 

[blackjack23]

Does this "Secret Sauce" take injuries into account?

 

[RedRaiders]

Does this "Secret Sauce" take injuries into account?

Is that the excuse for the pitching staff getting their ### kicked in three straight games?

 

[Liquid Tension]

Some of you might remember the discussion last year, check it here

http://forums.footballguys.com/forum/index...=348032&hl=

Basically, last year BP said "The Red Sox, in fact, might be one of the best Secret Sauce teams of all time, ranking in the Top 3 in all three categories."

We all know what happened in the playoffs

For those not in the know, the secret sauce attempts to break down the elements that make a difference in the postseason. It has found the following to have a correlation to postseason performance:

Adjusted Strikeout Rate

Fielding
Closer performanceBeing strong in these categories does not guarantee success in the playoffs, but the numbers suggest that it gives an edge.

That said, here is the list of this year's Secret Sauce rankings (surprisingly, with the same #1 and #2 as last year)

http://www.baseballprospectus.com/statisti....php?cid=280104

1. Red Sox

2. Cubs

3. Toronto

4. Angels

Anyway (and I am looking at you Capella), this is not a Ra-Ra Red Sox post, but moreso to generate discussion. As most know I like these statistic-based analyses, and in particular find this interesting as it goes against the Moneyball / Beane line of the postseason being a complete crapshoot

Maybe MoneyBall was right as the above was WAY off. My opinion is that the items you mentioned (and some offensive categories) are good indicators, but the reason why it is a crapshoot in the playoffs is that even the best teams win only 60% of the time against all the teams. When you have nothing but the better teams the odds of winning are reduced from the 60% and in a 5 or 7 game series one event can flip the series. On top of this, the playoff schedule is VERY different from the regular season with all the days off and the absurd amount of time off before the WS (like this year).

 

[Doctor Detroit]

 

[wilked]

Update for this year

http://www.baseballprospectus.com/statisti....php?cid=280104

Yanks, Dodgers, Giants a clear top 3, Sox start of next tier

From a Red Sox perspective, looking quickly the Sox get KILLED on their D (they rank 2nd and 3rd on the other two categories)

The new acquisitions should help that tremendously

Yankees very well-balanced

San Fran hurt by no dominant relief pitcher

This suggests Philly and the Angels won't be going very far

 

[Greco]

Fernando Rodney may be the worst closer ever to go 31 for 32. The Tigers in the 5 spot here eases some of my anxiety. Some.

 

[JaxBill]

Update for this year

http://www.baseballprospectus.com/statisti....php?cid=280104

Yanks, Dodgers, Giants a clear top 3, Sox start of next tier

From a Red Sox perspective, looking quickly the Sox get KILLED on their D (they rank 2nd and 3rd on the other two categories)

The new acquisitions should help that tremendously

Yankees very well-balanced

San Fran hurt by no dominant relief pitcher

This suggests Philly and the Angels won't be going very far

Hope that it works as well as last year's ranking where they didn't rank the Phils in the Top 4.

 

[wilked]

It is not a perfect system, just shows where a team has an edge. Historically it has been shown to work, but naturally there will be year-year variation