Yesterday, I took the delight in releasing the hard work that had consumed my life for the past week. Of course, I learned afterwards that Baseball Reference was already keeping track of one of the things I thought I had discovered, which I admit was a little heartbreaking. It’s ok though, I’ll live. Creating three things from scratch was still pretty fun to do, even if I complain every now and then about how soul-crushing it occasionally was for me. Today’s post is looking at a player’s ability to drive in runners in situations with runners in scoring position and (the existing statistic) the ability to drive in any runner.

First, there was some rules that I created to judge if a player should be rewarded (successfully advanced or scored a runner) or penalized (unsuccessful attempt) based on the outcomes during a plate appearance.

- In order to be rewarded or penalized, the hitter had to complete his plate appearance. For example, if a runner was picked off or caught stealing, the hitter was not penalized for failing to advance the runner. Continue with the new situation.
- Likewise, if the runner successfully advances a base without help from the hitter (stolen base, wild pitch, etc.), the hitter is not rewarded. Continue with the new situation.
- If a batted ball leads to an error, the hitter is not rewarded for advancing or scoring runners if they would not have advanced or scored without the error. Say there is a runner on 2nd base and the batter hits a grounder to the second baseman, but he is charged with an error, causing the runner to score. The hitter is credited with advancing the runner one base (because he most likely would have made it to 3rd without the error) but not for scoring the runner or advancing him two bases.
- If the runner successfully advances at least one base but is thrown out later in the play, the hitter is rewarded for all runners/bases that were succesfully advanced. For example, there is a runner on 1st and a ball is hit to the outfield. The runner advances to 2nd but is thrown out at 3rd base. The hitter is rewarded with advancing the runner one base.
- If a play ends the inning in a manner not described by the previous point, the batter is not rewarded for advancing runners that would have remained on base/scored if the inning had not ended. For example, say there are runners on 1st and 2nd with 1 out. The hitter grounds into a 4-6-3 double play, ending the inning. The hitter is not rewarded for moving the lead runner from 2nd to 3rd.
- Unless the play ended the inning, a hitter was rewarded for scoring/advancing runners, even if he made an out (sacrifices, GIDP with a runner scoring, etc.). Note that this means that I did not look at RBI, but rather whether or not a runner scored as a result from the hitter putting the ball in play.

Due to my rules, there were a few problems that were created.

- The rules above were subjectively chosen. They could have been modified, causing my results to change.
- When there were runners in scoring position, I counted all runners that scored rather than just those that scored that were already in scoring position. (If there were runners on 1st and 3rd and the batter hit a home run, he was rewarded with driving in 2 runners even though only one of them was in scoring position). I had intended to look at only runners in scoring position, then promptly forgot about them once I started looking through the 2009 season.
- Hitters were credited with doing something “good” even if they left the game in a worse situation than when they started their plate appearance. What I mean by this is that each situation in the game has a run expectancy, as shown by the matrix below. In some cases, even if you advance a runner, the out that you sacrificed reduces your chance of scoring a run in the new situation.

In other words, in 2005, if there was a runner on 2nd with 1 out and you sacrificed the runner to 3rd, your probability of scoring one run dropped from .41 to .254. Note that this matrix does not account for the probability of scoring more than one run.

I don’t know the significance of the following graph, other than it’s a little interesting to note that if a hitter scored well in one statistic, he generally scored well in the other three. The correlations are fairly strong (around .85 or so, roughly) when pairing any two of the four statistics, which confirms the assumptions made from the graph. By the way, check out how awesome Joe Mauer was last year.

I apologize, but I feel that this post is getting rather wordy and messy, so I’m going to continue everything on a second post. If there are any additional questions, comments, or concerns, feel free to leave them in the comments section, whether it’s for this post or for the next one.

January 21, 2010 at 11:53 am |

Here’s a table that shows how many runs are expected to score, rather than the probability of scoring:

http://www.tangotiger.net/RE9902.html

As a side note, I disagree with awarding a hitter who makes an out but moves a runner ahead on the bases. Advancing a runner by making an out is a negative play in every situation. I appreciate the work you’re doing, but I can’t go along with anything that rewards a hitter for making an out.

January 22, 2010 at 6:20 am |

A couple interesting things about your graph:

1) Span > Gomez (Not too surprising, but still interesting)

2) Punto does his job. He advances runners well, and scores them at around the team average. I’d be interested in a graph that correlates advancing runners with improving or hurting the situation.

3) Delmon Young didn’t suck (I know, it’s a personal prejudice). I’d be interested in a graph that shows how his numbers changed over the year. I have a sense that, like the rest of his stats, they peaked near the end of the year.

4) Cuddyer is below average when he’s not hitting the long ball. He lead the team in HR, but he also led in SO and AB (excluding Cabrera). His RBI numbers appear to be more about persistence that consistency.

January 22, 2010 at 6:51 am |

Also, I have to disagree with Steven on sacrificing always being a bad play. The point is not that getting a hit is always better than sacrificing, of course that’s true. The point is that the sacrifice is always better then just getting an out.

Two examples (using Steven’s graph) using simplest outcomes:

1) Runner on first with 1 out (.573)

– Hit that advances the runner to 2nd (.971)

– Sacrifice (.344)

– Out (.251)

2) Runners on first and third with one out (1.243)

– Hit that advances both runners one base (.971, run scores)

– Sacrifice, both runners advance one base (.344, run scores)

– Sacrifice, man on third can’t advance (.634, no run scores)

– Out (.538, no run scores)

In fact, I could argue that the in the second outcome of the second example you’ve actually improved your situation because you scored a run (1 + .344 > 1.243).