No, not this 24.

Baseball is very unique when it comes to team sports. Unlike basketball, hockey, and soccer, action isn’t continuous. There’s a definitive start and end to each event in a baseball game. Football, like baseball, also has this trait. However, in football, there are almost endless possibilities at the end of each play. There could be any number of seconds remaining, 100 different ending yard markers, and numerous different scoring possibilities dependent on where the ending yard marker is, even different spots between the hash marks for the ball to be spotted.

Baseball holds little of this uncertainty. There are a set number of scoring opportunities (the 9 innings) and there are a finite amount of possibilities inside each inning after each play. There could be a runner on first and one out, there could be bases loaded and two outs, and there could be bases empty and nobody out. Let’s take a look at all the possibilities inside an inning.

So we see we have 8 different possible states on the bases, and each one can happen at each out number. That makes 24 possible states inside of each inning. This is key behind how much of linear weights and other sabermetrics works. Everything that happens in a play can eventually be summed up by transitions from one state to another. Let’s take a look at a sample inning, thanks to www.retrosheet.org .

BOTTOM 8TH INNING:

Baserunners: ___ Outs: 0 Runs Scored: 0

Sabathia was called out on strikes;

Baserunners: ___ Outs: 1 Runs Scored: 0

Cameron singled to left;

Baserunners: 1__ Outs: 1 Runs Scored: 0

Durham flied to right;

Baserunners: 1__ Outs: 2 Runs Scored: 0

Braun homered [Cameron scored];

Baserunners: ___ Outs: 2 Runs Scored: 2

Fielder struck out; 2 R,

Outs: 3 Runs Scored: 2
Inning Summary: 2 H, 0 E, 0 LOB. Cubs 1, Brewers 3.

So here we have 5 transitions. Let’s take a numbers-free approach to analyzing the effects these transitions had on the inning. First, the strikeout. Clearly, this was a detriment to the team. The single following clearly contributed to the team scoring. The fly out was, again, harmful (it’s funny how outs tend to be bad…). The home run marks a transition to a worse state (it’s better to have runners on), but the 2 runs scoring on the play far outweighs the loss in the transition. Finally, the strikeout is, once again, bad.

The more interesting thing to look at here is how we allocate the credit for producing the two runs. The gut reaction would be to just give it all to Braun for the HR (this is what RBIs do). Another possibility would be to give one to Cameron for scoring and the other to Braun for scoring (this is runs), or to give one to Cameron and one to Braun each for getting a hit (batting average) or getting on base (OBP). Or, you could give Braun four times as much credit for amassing 4 total bases to Cameron’s 1 (SLG), or maybe some combination of the last two (OPS).

The reality is that none of these truly capture the dynamic of run production. Tomorrow, I will post a second part to this article explaining the quantitative use of these base/out states.

Tags: Balking Traditionalism, Introductory Sabermetrics, Jack Bauer, Ryan Braun

This entry was posted on Friday, January 16th, 2009 at 2:45 am and is filed under RFB. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.