The RBI Game: More informative than actual RBI

While calling this a game might be a stretch, I think it’s an interesting way to see what really affects a player’s RBI totals.

Here’s how the “game” works. If you want to play along, it will require a four-function calculator and the internet, both of which you already have because you are using some form of computer technology. I know this because I am an astute individual.

  1. Pick any player from 2012 that got at least 200 plate appearances.
  2. Find his total number of homeruns.
  3. Find his total number of baserunners (the BR column).
  4. Find his slugging percentage with runners on base.

I have linked to pages that will help you find the information quickly. Once you have all that information, follow these steps.

  1. Take the homeruns as they are, multiply baserunners by 0.15, and multiply the slugging percentage by 59.71.
  2. Add those three things up, and then subtract 26.97 from the total.

The number that you get at the end is an estimate of how many RBI that player racked up in 2012. Notice, I didn’t cheat and use actual RBI figures, but instead I used other things that might have helped a player accumulate RBI. Homerun totals and SLG% with runners on base have everything to do with the player himself, but total baserunners  in front of our hitter have just about nothing to do with the hitter himself. Let’s see how the formula—er, I mean, game!—works. I’ll take some volunteers you might recognize: Ichiro Suzuki, Jesus Montero, Casper Wells and Mike Morse.

Player Ichiro Montero Morse Wells
HR 9 15 18 10
BR 374 351 259 176
BR x 0.15 56.1 52.65 38.85 26.4
SLG 0.360 0.386 0.491 0.473
SLG x 59.71 21.50 23.05 29.32 28.24
Sum – 26.97 59.63 63.73 59.20 37.67
Actual RBI 55 62 62 36
Difference 4.6 1.7 2.8 1.7

Being able to estimate a player’s RBI totals so accurately using just three pieces of information is pretty neat.* Considering Ichiro played for two different teams, that estimation was pretty solid…and it was the worst one of the bunch! But the main thing I wanted to point out the importance of having runners on base for a player’s RBI total. The baserunning component was a key part of the formula for each player—as you can see above—yet a player has very little (if any) control over that.

I may be preaching to the choir, but I think that any statistic which is partly a function of a player’s teammates, like RBI, should not be used in place of statistics that are almost wholly functions of the individual himself, like maybe homerun totals and slugging percentage. I think this is a major reason why the Sabermetric community has disposed of RBI as a relevant statistic, replacing it with statistics that represent an individual’s ability independent of his teammates.

For the same reason, you won’t find much mention of RBI in my articles. I prefer more daring abbreviations like wOBA and UZR.

*Nearly 93 percent of players with at least 200 plate appearances in 2012 could be esimated to within 10 RBI of their actual totals using this formula. 67 percent of these players recorded RBI totals within five RBI of their estimates. 

  • http://twitter.com/TheMockingJ sgreen516

    This isn’t really a “predictive” tool. I can’t use it to determine Ichiro, Montero, Morse or Wells 2013 RBI totals with it. As I see it, I’d still need to know how many Home Runs, SLG % and BR they will have in 2013 to predict their RBI totals in 2013. Unless you have an independent way to predict those other variables?

    • Matthias_Kullowatz

      You’re absolutely right. I was playing fast and loose with the term “prediction.” Really it was more “estimation.” But the fact that I could estimate RBI so closely, and that I had to use baserunners to do it, says a lot about RBI as a statistic I think.

      As for prediction, I could check every player’s SLG, HR and BR in 2011 to see if they were predictive of 2012′s RBI, and that would be more “predictive” as opposed to explanatory. I think we would find that BR is still a necessary component.