Sunday, March 14, 2010

Jason Bay and The Problem With RBI's

Joe Posnanski published a blog yesterday in which he discussed a recent Bill James essay entitled "The Attribution Problem." The essay can be found in the Bill James Gold Mine 2010. Posnanski relays the concept of the essay:
We attribute the victory won by the team to the individual pitcher — and then conclude, based essentially on that attribution, that the pitcher is the key to victory.

That sums it up pretty well, doesn’t it? The win is all about sleight of hand. Pitchers don’t win games, and pitchers don’t lose games — that should be obvious to everyone. But people decided a long time ago just the opposite: That pitchers do win and lose games.
It is a concept that many of us are already familiar with, both in baseball and in life. We do not always give credit to the things that REALLY help us succeed - sometimes we give credit to things or people that are more obvious, and sometimes we just don't know what is responsible for our success.

Baseball is a truly wonderful place for us to be open-minded and critical of our thought process. Unlike life, we can reconstruct sequences and we have solid records and numbers which we can point to which tell us exactly what happened.

Posnanski's article, however, really focuses on the concept of the "RBI guy."[1] Statistically-minded writers have long been downplaying RBI's as a way to judge the value of players (and correctly so) and this article is more of the same. However, I always like reading Posnanski's take on things and here he adds an interesting tidbit to the conversation.

At the outset of the article, Posnanski asks you to choose one of two options. (1) Your team could have more hits than the opponent or (2) Your team could hit three or more home runs. Which of those two options would give you a better chance of winning? You may be surprised to learn the answer:

Teams that outhit their opponents won 80.3% of the time.
Teams that hit three-or-more homers won 78.4% of the time.[2]

The point is not that RBI on their own are good or bad, but rather, that you should be adding the best players to your team regardless of that single, ephemeral statistic. RBI are good, but hits and on-base percentage are better. RBI is useful, but defense and baserunning is, too.

Punch "Jason Bay" and "Run Producer" into Google, and you'll get thousands of hits. If you limit the search to January 1st to the present, you get 247 hits. He had 119 RBI last season, and has averaged 31 homers per year in the last four seasons. However, Jason Bay was not, and is not, the solution to our problems. As Bill James succinctly stated in his essay:

If you add a low-average power hitter to a bad team, the low average power hitter will lead the team in RBI — and the team will score fewer runs, not more.

That is what I fear with the Bay sitation. Although the Mets will find it almost impossible to be worse than last season offensively, I worry that the addition of Jason Bay will lead to gaudy but unproductive power numbers. With a lineup that is so top-heavy (with Reyes, Beltran and Wright all elite producers) Bay is going to have a ton of opportunities to drive in runs.

But with a low average (only .267 last year), Bay is a feast-or-famine hitter entering the twilight of his career. In watching the games this year, I worry that we'll see a lot of two- or three-run home runs when we do not need them, and a lot of strikeouts in games that are close. At the end of the year, Bay is going to have his 100 or 120 RBI, but I doubt that our offense will be much more productive.

---------------------------

[1]From the article:
Teams don’t score runs because they have uniquely talented RBI men. Teams score runs because more often than their opponents, they put together a string of useful offensive plays — walks, hits, stolen bases, hit-by-pitch, beating out double play grounders, taking extra bases, advancing on throws , on and on and on. That, most of the time, is what lead to runs.* The RBI guy cannot do it himself except with solo home runs. And teams don’t win games by hitting solo home runs. No, really, they don’t.


[2] The usual caveats apply here of course. It's not a scientific study, the numbers only apply to 2009, and the "three home runs" option does not mean that your team necessarily hit more home runs than the other team.

6 comments:

martin said...

the problem with your analysis is that bay is actually a good OBP guy who doesnt make outs. his low average is of little consequence relative to his power and on base percentage. all of what you say would be true if he didnt get on base and if he made outs like jeff francouer.

Brian Mangan said...

You are absolutely right in pointing out Bay's good OBP. However, walks are less valuable than hits and players with low averages tend to have more trouble aging gracefully.

If Bay's bat slows and his average declines, pitchers will just challenge him and he won't be able to draw walks anymore.

Anonymous said...

Yes, but what you fail to take into account is that teams that outscore their opponents win 100% of the time.

In order to outscore your opponents, you need to drive in runs.

martin said...

his career batting average is not particularly low. plus there is no reason to think he would age worse than anyone else. good OBP guys age well. bobby abreu is a good contemporary example. again, your analysis would be spectacular if you change the word "bay" to "francouer".

martin said...

anonymous, you are not understanding the point. big rbi guys who make outs (francouer) will lead their teams in RBI, while causing them to score less overall and lose. rbi are valuable when they come from guys who do not make outs. that is why bay is good. he gets on base and makes the offense work, even if he isnt driving in runs. low OPS guys with RBI are actively hurting their teams, even as they drive in runs.

Brian Mangan said...

Martin, I think he was being sarcastic.

However, your feedback prompted me to write a follow-up post. Hopefully this more accurately states what I was trying to say in my original one.