Playing for One Run

Question of the day: What does a "Moneyball" team do differently strategically than a traditional team?

Answer: Well, one thing they do is take seriously Earl Weaver’s axoim that a team’s 27 outs are their most valuable possessions.

One way this can be seen is by looking at how often American League teams successfully sacrificed in the first three innings of games in 2004.

CHA 25
ANA 15
KCA 15
SEA 14
DET 13
BAL 11
NYA 11
TBA 11
TEX 9
CLE 9
MIN 6
OAK 5
TOR 3
BOS 1

As you’ll note the teams run by sabermetrically-minded Bill Beane, J.P. Riccaldi, and Theo Epstein are at the bottom of this list. Why? Because bunting is inherently a strategy that increases the probability of scoring a single run while decreasing the chances of scoring multiple runs. This is easily seen by looking at simple run expectancy and scoring probability tables. When you do the math to calculate the break-even percentage you find that it is never a good idea to sacrifice when you’re trying to maximize runs. And of course early in the game a team should be attempting to maximize runs, which incidentally, leads to another of Weaver’s axioms; “Don’t play for one run unless you know that run will win a ballgame”.

Clearly the White Sox under Ozzie Guillen don’t follow this approach by sacrificing 25 times in the first three innings. When you examine those 25 you find that fully 12 of them were by the first or second spot in the order although one of those was actually a squeeze play.

When you look at successful sacrifices more generally you find the following for the major leagues as a whole:

By Lineup Pos

1 11%
2 18%
3 2%
4 0%
5 3%
6 4%
7 7%
8 9%
9 46%

By Outs

0 78%
1 22%

By Inning

1 7%
2 11%
3 13%
4 10%
5 13%
6 10%
7 10%
8 11%
9 8%
10 3%
11 1%
12 1%
13 1%
14 >1%
15 >1%
16 >1%
17 >1%

By Score Differential

-8 >1%
-7 >1%
-6 >1%
-5 >1%
-4 1%
-3 2%
-2 6%
-1 14%
0 35%
1 16%
2 11%
3 7%
4 4%
5 2%
6 1%
7 >1%
8 >1%
9 >1%