Chapter 2 of 11
The sacrifice bunt
Trading an out for a base almost always costs runs. It can still win games — the difference is which question you are asking.
The short version
Bunt when one run is the whole game and the hitter at the plate is unlikely to do better; otherwise swing.
KBO 2026 season data through Oct 13, 2026
01
The idea
The sacrifice bunt sells an out to buy a base. Chapter one priced both: an out early in an inning is the most expensive thing an offence owns, and a base is worth much less. So the bunt, judged by expected runs, is almost always a losing trade.
But expected runs answers the question “how many runs will we score?” A manager in the bottom of the ninth of a tied game is asking a different one: “will we score at all?” Bunting trades away big innings — which no longer matter — for a slightly better shot at one run. That distinction, between playing for one run and playing for many, is the whole of bunt strategy.
02
The math
V(bunt) = h·V(hit) + (1 − h)·[s·V(advanced, +1 out) + (1 − s)·V(same bases, +1 out)]
s is the chance the runners move up and only the batter is out; h the chance the bunt becomes a hit; V any yardstick — runs, the chance of a run, or win expectancy. Bunting pays when s clears the break-even.
| Situation | Early: runs | Chance of a run | Tied, bottom 9th | Tied, bottom 11th |
|---|---|---|---|---|
| Runner on first | never | 90% | 90% | 90% |
| Runner on second | never | 80% | 80% | 80% |
| First and second | never | 78% | 78% | 78% |
Read the first row across. Early in a game, with a runner on first, the bunt costs runs at any realistic success rate: 0.97 expected runs swinging away against 0.74 after a perfect sacrifice. The chance of scoring barely moves (44% against 44%). Tied in the bottom of the ninth, the perfect sacrifice takes the home side from 71.9% to 72.0% — and real sacrifices are not perfect.
The second row is different. With a runner already on second and nobody out, a sacrifice puts him ninety feet away with two chances to score him on a fly ball, a grounder, a wild pitch. Early on it still costs runs (1.20 to 0.97), but the chance of a run rises from 64% to 67%, and in a tied ninth the home side's win value goes from 81.9% to 83.6%. That is the classic bunt situation, and the numbers agree with the dugout — provided the sacrifice works at least 80% of the time. In the bottom of the ninth or later, win value moves one for one with the chance of scoring, which is why the last three columns of the table agree.
03
Try it
Pick the runners, the outs and the game. The win column appears once there is a scoreboard to win against.
| Yardstick | Swing away | Bunt | Change | Break-even success rate |
|---|---|---|---|---|
| Expected runs | 0.97 | 0.75 | −0.22 | never pays |
| Chance of a run | 43.7% | 41.6% | −2.2 | 90% |
Success: batter out, every runner up a base. Failure: batter out, runners stay. Six percent of attempts are assumed to become hits. Win value is a win plus half a tie.
04
This season in the KBO
Sacrifice bunts
590
Per team game
0.41
Per game, 2003
0.71
Change since 2003
−42.3%
KBO clubs laid down 590 sacrifices this season, 0.41 a game — down from 0.71 in 2003. The league has absorbed the argument. Who still bunts says the rest: the six hitters with the most sacrifices have a combined wOBA of .335 against a league mark of .346. Managers bunt with the hitters least likely to do better swinging, which is exactly when the break-even falls.
| Most sacrifices | SH | PA | wOBA |
|---|---|---|---|
| Park Hae-min LG | 18 | 591 | .344 |
| Han Seung-taek KT | 17 | 365 | .284 |
| Koo Bon-hyuk LG | 16 | 390 | .315 |
| Kim Sang-su KT | 15 | 489 | .337 |
| Kim Tae-yeon Hanwha | 15 | 452 | .357 |
| Kim Ji-chan Samsung | 14 | 530 | .356 |
05
When the conventional wisdom is wrong
01
The first-inning bunt
Playing for one run in the first inning assumes the game will be decided by one run. Most are not, and the bunt gives back the big inning that would have made the rest easy.
02
Runner on second, nobody out, late
The one situation where the sacrifice holds up on every yardstick but runs. If one run wins or ties, and the hitter is not one of your best, bunt.
03
The hitter matters more than the situation
The tables assume a league-average hitter. A .250-wOBA bat swinging away is worth far less than that, so the break-even drops; for a cleanup hitter it climbs out of reach. Never bunt with your best hitter.
04
The eleventh inning is its own game
In the bottom of the eleventh the home side cannot lose by failing; it can only win or tie. One run is literally everything. That is the purest bunt situation in the KBO, and chapter eleven explains why.
The KBO records successful sacrifices but not attempts, so this season's success rate cannot be computed; the tool lets you set it. Win values use the eleven-inning win tables built from league-average clubs, and treat a tie as half a win — exactly right for a .500 club, which the last chapter qualifies.