How to Win in the Playoffs: According to BP

As long time readers of my blog, and fellow obsessive compulsives who has read through all most old posts here and at my old blog, might recall, I've been saying that the best way to win in the playoffs is to go in with dual aces like Koufax and Drysdale, Randy Johnson and Curt Schilling, and I think we have them with Cain and Lincecum, at least they appear to be thus far. If Sanchez can join them, all the better, and now I have to add Misch to the equation, I love them strikeout pitchers. However, it is one thing to have a theory, another thing to try to prove it.

Then I ran across a study by Baseball Prospectus in their latest book called "Baseball Between the Numbers" and Chapter 9, Section 3, is titled from a paraphrase of a Billy Beane quote from "Moneyball": Why Doesn't Billy Beane's Sh*t Work in the Playoffs. In it, the authors of that chapter section investigates what are the factors that contribute to success in the playoffs. (BTW, I wholeheartedly recommend the book, only $10 at Amazon plus shipping, no shipping if you order only $25, what a deal!).

Why Scoring Doesn't Correlate and Pitching and Defense Does

In the book, BP studied Playoff Success Points (PSP), which they created for this investigation. In it, points are assigned based on various characteristics of each playoff team. For example, a team gets 4 points for winning the World Series. And they studied the 180 playoff teams between 1972 and 1995. What they looked at first was the correlation coefficient between PSP and each team's runs scored per game and runs allowed per game (that's earned runs plus unearned runs over a season).

They found that runs allowed accounts for 0.22 of the change in PSP for a team, which, while it seems small, still signifies that pitching and defense matters to a certain extent in the success of a team in the playoffs. Which is better than what the offense contributes: essentially a zero correlation between PSP and runs scored per game. Thus, while preventing runs correlates with post season success, scoring them does not. There literally is no relationship between regular season offense and post season success, based on the results of this study. I found that very amazing.

They also noted, "Does this mean that defense really does win championships after all? The short answer is yes, probably." What is "clear is the diminished importance of offense in the playoffs." They then observe, "It isn't that good pitchers have a structural advantage against good hitters, but that good pitchers have a structural advantage against good-hitting teams." That advantage comes to the forefront in the playoffs, when all the teams can hit pretty well, they added.

Digging Deeper

This is confirmed by their analysis of individual offensive, defensive/pitching, and other possible factors. Among offensive stats, as was shown on an overall basis with runs scored is shown for the individual offensive metrics, except for stolen-base attempts which had a correlation of 0.13, which is slightly more than half the correlation for runs allowed.

Pitching and defense has a number of correlations. There is a 20% correlation between all starting pitcher VORP and PSP (VORP is BP's metric for valuing a player over a replacement player), and most of that is delivered by the Top 3 starting pitcher VORP which has an 18% correlation with PSP.

There is an even stronger effect in the bullpen, specifically the closer: "The relationship between the closer's Win Expectancy, adjusted for replacement level and the opposing lineup (WXRL, also known as the increase in the probability of a pitcher's team's going on to win the game given the game situation...) and PSP is quite strong - stronger in fact than when we look at the bullpen as a whole." Which means that the performance of the non-closers is not of significant importance to playoff success. Which is not to say you should not have a good bullpen, but that it is not significantly correlated to playoff success as defined by PSP.

The one with the highest correlation to post season success is the Opponent's Batting Average, which is pretty obvious. But interestingly, avoiding walks doesn't seem to have much relationship with playoff success. And that ties into the point above about facing good teams solely in the playoffs: you don't let the good hitters beat you, it's better to walk them.


This also ties into this correlation: Fielding Runs Above Average has one of the higher correlations with playoff success at 0.16. The better your team's defense, the lower the Opponent's Batting Average. In addition, there would be less errors that could lead to unearned runs that could cost you a key game in the playoffs. A team's playoff fortunes can swing from one end to the other of the success spectrum based on one game where the defense fails you. I think Giants fans can relate to that (koff, Jose Cruz, Jr., koff).


Another interesting relationship found was that between a team's playoff experience and playoff success. Though only a slight relationship - about 0.12 - plus it was skewed by the Yankees last dynasty in the late 1990's, they don't think that this hypothesis can be dismissed out of hand. However, the relationship isn't statistically significant, especially when one considers that it is the better players who tend to accumulate a lot of postseason experience.

So obviously, you need a good enough offense to win given your pitching, but once you get into the playoffs, having the best offense doesn't do anything to help you achieve success in the playoffs, as it is apparently not only negated by facing the other team's best pitchers more often, but rendered to be a non-factor in playoff success.


Deeper Analysis: Regression


Next they did a regression analysis using all the variables in order to identify the factors that are the most fundamental and direct relationship with PSP. The following are the factors that significantly ties into playoff success:

• Closer WXRL
• Pitcher Strikeout Rate
• Fielding Runs Above Average

The importance of an above average closer is clear: post season games are usually close contests between evenly matched teams. And that results in many more opportunities for the closer to pitch in high-leveraged situations. I would also add here that in the playoffs, one game swinging from a loss to a win can be pivotal: instead of being 2-1 and one game from winning the first round, you could be 1-2 and one game from being eliminated; or instead of being tied 2-2 in the second round or World Series, you are down 3-1 and one game away from elimination.

The importance of the strikeout is also clear: the good hitters can't harm you in any way if you strike him out. This is particularly important when facing the good offenses you normally run into in the playoffs. Also, good hitters tend to tee off against finesse pitchers while they lose some of their advantage against power pitchers.

Lastly, again, it is clear why good fielding is important: the ball is put into play more often than not, so having good defensive players is key to not giving away runs. Also, the pitcher might feel more comfortable challenging hitters if they know they have a good defense behind them when the ball is put into play.

However, when you can combine a strikeout heaving pitching staff with a great defense, it can become nearly impossible for its opponents to get hits and generate rallies. Of the 33 teams to win the World Series from 1972 to 2004, only 5 had a below-average defense. Defense is important against good offenses since good hitters put hard-hit balls into play and test the defense far more frequently and strikeouts are important because it reduces the number of balls put into play.

These three do not strongly determine success but if the team can be strong in each one, you improve your chances of winning greatly. When the 180 teams are ranked by these three measures (averaging the three ranks), 8 of the top 10 won the World Series and one of the teams that did not win ended up losing to the team with the better average composite score. The bottom 10 didn't do anything in terms of World Series success, all of them did not reach the World Series.

Sabean Ahead of the Pack, the Times

What has been the hallmarks of a Brian Sabean team? He has always seeked a strong closer and a great defense, plus, of course, there was Barry Bonds. What has Brian Sabean been advocating lately plus drafting and trading for? Creating speedy teams that can steal bases and play good defense. What type of rotation should we have for the foreseeable future? One led by Cain and Lincecum, two top strikeout pitchers plus there's Jonathan Sanchez waiting in the wings plus now Misch had a very nice first start, after striking out a lot as a reliever. And as I have shown above, from the BP study, are all shown to be key determinants of playoff success and particularly World Series success.

What has Sabean naysayers been complaining about Sabean? About his lack of position prospects. About his lack of good hitters. As another chapter in the book makes clear (very interesting read and I've only read maybe 10% so far), you need to balance offense and defense/pitching on your roster, you can't focus just on pitching to the detriment of offense, nor vice-versa. So it is not like Sabean can shirk this GM duties on offense, so that is not a point I'm trying to make here.

However, to get the players that contribute in a significant way to improving your chances of winning it all in the playoffs, as outlined above, the best way to get these types of players is to focus your draft on pitchers and to draft a lot of them. You need to kiss a lot of frogs along the way - Grilli, Vogelsong, Ainsworth, Williams, Jensen - to get your aces (plus a dash of luck - Lincecum - but as the old saying goes, you sometimes have to make your luck), particularly when you are drafting back in the first round, where the odds of finding a good player - period - is not that good at all, even worse than the odds of rolling a 7 on your first roll in Craps.

And how hard is it to find good defensive players? It appears that the key is to field a strong defensive team, with a couple or three key offensive players, based on what I've seen in studies that analyze batting orders: leadoff, cleanup, and maybe 5th or 2nd spots. Fans complain constantly about one player or another who is deficient offensively but no team fields a lineup full of good hitters, most have average or poor bottom of the lineup hitters plus someone with no power hitting second.

Thus half your starting lineup is probably OK to average hitters but good defensively and be relatively cheaper, leaving the bulk of your payroll for your strongest hitters because the pitching staff is relatively cheap since you are focused on developing pitchers. For example, as maligned as Pedro Feliz is, despite his poor offense at a classic power position, he is very good defensively (near the top of most of the advanced stats I've seen for 3B) and his bat is good enough to hit in the 7th or 8th spots. They key is to stock good players elsewhere in the starting lineup.

This reminds me of economics, David Ricardo's comparative advantage principle that if each country (or organization) focuses on what they do best, the economy will be better off overall, with more production. That is essentially what Sabean has been doing with his focus on drafting pitchers and acquiring other teams' pitching castoffs, like Scott Eyre, Alan Embree, and Felix Rodriguez.

By doing this, eventually he will be a major supplier and controller of two of the main areas of proven significant connection with playoff success, closers and high strikeouts. Once he gets the pitching factory to produce pitching surplus, he will be able to trade off his suprlus to other teams for parts he does need. With a pitching staff that is pretty much complete and full of pitching prospects - except for a dominating closer - the Giants appear to be capable of doing this going forward as young prospects like Misch, Wilson, Sadler, Anderson, Sosa, Cowart, and Joaquin, for examples, continue to develop and advance in our system, and new draftees like Tim Alderson join the system.

Meanwhile, he has a team full of complementary offensive pieces that will hold the fort until we can obtain the better offensive players. With an extra $13M added to the payroll for 2008 from the trade of Matt Morris, the Giants should have around $40-50M cleared from the payroll for 2008, money that can be spent on obtaining better players for the starting lineup. Up for grabs right now is an outfield spot (unless Bonds somehow re-signs; doubtful given the Giants pronouncement that 2008 is not necessarily a competitive year), 1B, 3B, and SS.

With Roberts and Winn capable of playing all three positions, the Giants might look into players like Andruw Jones and Torii Hunter. I'm not sure who else are premium players coming up in the next two off-seasons, but one or two good signings, plus a trade, probably of Lowry, should yield the players we need.