I’m not sure what to make of the first inning. It makes sense that the first inning would feature more scoring, and thus fewer draws, since the first inning is the one time when each team always puts forth its optimal batting order. (Or at least the order it thinks is optimal.) It makes less sense that so many of the non-drawn first innings are won by the home team.
It’s because the home team is also at its best. Fresh pitcher, warmed up fielders, and the visiting batters have to get an idea of where the strike zone is. Whereas, the home team, being in the field, can see where the strike zone is being called.
SFC Schwartz
I understand that. But I think it buttresses my point. There is no special home-team advantage in extra innings; they just have essentially the regular home-team advantage for one inning, which is of course smaller than the home-team advantage for a 9-inning game. This is a step down from the home-team advantage for the game as a whole; we agree on this.
As I said before, there is no “advantage in and of itself for the home team” in extra innings.
Sure, this is a slight advantage for the home batters, but wouldn’t this also be an advantage for the visiting pitcher? (And in any case decline dramatically to near nothing after the first inning or two).
Back to extra innings: Has anyone checked to see how statistically significant is the difference between actual home-team-win rates per inning in extra innings versus the expected rate based on win rates per inning in regular-length games?
I mean, we’re looking at pretty small numbers once you back out the regular per-inning home team advantage, right?
The “expected rate” strikes me as a difficult calculation. You’d need not only the win-lose-draw results from earlier innings, but the distribution of wins by number of runs.
And, I’m not sure it’s necessary. You can measure the same thing directly via the percentages.
Throwing out the draws, and throwing out the anomalous first and ninth innings, the home team wins by inning are:
2: 52.2%
3: 53.1%
4: 51.9%
5: 52.6%
6: 51.9%
7: 52.1%
8: 51.9%
10: 52.5%
11: 52.5%
12: 52.6%
The average home wins in innings two through eight are 52.2% The average home wins in innings 10 through 12 are 52.4%.
If there is an additional home advantage in extra innings, over and above ordinary home advantage, due to foreknowledge of how many runs are needed to win, it’s microsopic. It’s at the very edge of observability.
Anyway, I’m not that surprised at the result. As FreddythePig pointed out, the defense actually makes more decisions than the offense (intentional walks, semi-intentional walks, whether to pay attention to a baserunner trying to steal, positioning the infield, positioning the outfield, whether to throw home to stop a run or go for an easier out, etc). In contrast the only decision the offense really makes is whether to sacrifice (bunt or sac fly), whether to steal and whether to stretch for an extra base.
So even knowing nothing about baseball, it’s plausible that the side making more decisions (the defense) actually has the advantage going last. There is, I think, a psychological advantage to batting last, but I’m not sure how much difference that makes to major league athletes.
And before anyone nitpicks my math, the 10th-inning percentage should be 52.2%. :smack:
Contrast with college football, where overtime features alternating possessions from the 25-yard-line. (Home advantage doesn’t come into play, because the team going last is random and alternates.) Again, the team going last knows how many points it needs, and the team going first knows how many points it can give up.
Curiously enough, the team going last also wins 52% of college football overtimes. But this is versus an obvious baseline of 50%, so going last offers a clear incremental advantage.
I think you found it. It may be only 2% but that is what casinos live on.
It’s nonexistent, from a statistical point of view. See my post #19 above … The percentages that change in extra innings are the winning % for the visiting team (noticeably decreases), and the draw percentage (for a given inning, which noticeably increases).