Posted by Andy on November 5, 2007
OK, here is part two of yesterday's post.
Yesterday, we saw that representing runs scored by inning using percentages was a bit misleading. Here, instead, is average run scoring by inning, AL vs NL. The results are averaged by team, and also averaged over the period 2003-2007:
So what this tells you is that an average team scored about 96 runs in the first inning (totaled over the course of the season.)
As you can see, the AL and NL averaged virtually identical numbers in the first inning (so that paragraph I wrote yesterday wondering why the NL seemed to score more in the first inning was a waste of everybody's time...sorry 🙂 )
But then the NL scoring drops way off in the 2nd inning and 3rd inning, trailing the AL all the way until about the 7th or 8th inning, when the gap closes.
Let me re-display the above graph in terms of difference seen in the NL, using the AL as the baseline:
So you can see that there's virtually no difference in the 1st inning. The second and third innings are the biggest disparity, in terms of fewer runs scored in the NL vs AL, at about 12% each inning.
I submit to you that essentially the entire difference in run scoring between leagues is due to the pitcher batting in the NL. Looking at the above graph, consider the following:
- There is no difference in the 1st inning because the pitcher never bats unless the batting team has already scored at least 3 runs.
- The 2nd and 3rd innings are those where the pitchers bat most often and where the NL falls the shortest of the AL.
- In these days of very few complete games, pitchers rarely bat in innings 7 through 9 (either pinch-hitters are used, or double switches are made to avoid the pitcher's spot from coming up) and the NL just about catches up with the AL in those innings.
Don't believe me? Here is one more piece of evidence.
Using the PI Team Batting Event Finder, here are all of the plate appearances this season by pitchers, broken down by inning:
Notice anything interesting? This graph almost an exact inverse of the run-scoring deficiency by the NL. To me, it's very clear that the more pitchers bat, the fewer runs get scored in the NL. (By the way, the graph above of pitcher PAs does not exclude AL pitchers batting, but obviously they are a small minority of the total pitcher PAs. Also, the above graph is for 2007 only, but I also graphed it for the average of 2003-2007, and the graph looks virtually identical. If I put both lines above, you'd barely notice any difference.)
Also yesterday I said I'd post something about detailed run-scoring in the 2nd or 3rd inning. I looked at that (specifically number of 1-, 2-, 3-, 4-, or 5+ innings by AL teams vs NL teams) but I didn't find anything worth mentioning.
Regarding the difference of run-scoring between the AL and NL, I say case closed.
What do you say?