from Amazon
Monday, August 07, 2006
Runners Driven In (RBI - HR)
John Walsh takes a look at runners driven in, by opportunity:
http://www.hardballtimes.com/main/article/confessions-of-an-rbi-fanatic/
I also highly recommend Tom Ruane’s infamous look at Joe Carter, using the 24 base/out matrix:
http://www.retrosheet.org/Research/RuaneT/rbipro_art.htm
Now, here’s what you can also do…
First off, Walsh has it right, by excluding HR. R+RBI-HR is the right way to do it. If you don’t believe me, read this:
http://www.tangotiger.net/rp.htm
If still not convinced, that’s fine. This isn’t the place for that talk. I’ll probably update that research piece, and you can comment at that time. Right now, the focus is on “runners driven in”.
Next thing I would do is use a version of the Ruane chart, but excluding HR.
Finally, and this is the kicker, you need to have a “tablesetter” chart as well. A walk still keeps the opportunities available for the next guy, and in some cases, moves the runners on base an extra base. As well, an out really reduces the opportunities to the next guy.
So as not to impact the batter’s on-base performance as well, remove him from the base/out equation.
As an illustration: say that Barry Bonds comes to bat, with a runner on 1B, 2B and 1 outs. He walks. Bases are now loaded, with 1 out. This is a good thing! A very good thing. What he does is changes the base/out situation from 1b,2b, 1 out, to 2b,3b 1 out. That basically improves the expected RBI for the next batter by .25 runners. (Remember, I’m not including the additional RBI opps he’s giving by getting on base.)
On the other hand, if he had struck out instead, he reduces the RDI opp of the next batter by .01 runners.
This process is a step in the right direction. Some of you may now be asking: “ok, you took care of the tablesetting for the next batter, but how about for all subsequent batters?”. You are now welcome to Linear Weights.
This table here shows you how to break down the run values of each event into its three components:
http://www.tangotiger.net/rc2.html
All you now have to do is do it for the 24 base/out matrix. So, in the case of Bonds walking with guys on 1b,2b, 1 out: his “getting on” run value is .28 runs. His “moving runners over” value is .40 runs. His total run value is +.68 runs, which we know from here:
http://www.tangotiger.net/RE9902.html
Where we can see that the RE with man on 1b,2b, 1 out is .971 and with bases loaded 1 out is 1.650, for a difference of .679 runs.
Had he made an out, his “getting on” and “moving runners over” is zero, and his “inning killer” is .505.