Chapter 7 of 11
Bullpen leverage and the closer
Some innings decide games and some do not. The save rule spends the best reliever on a three-run ninth and saves him from a tied eighth.
The short version
Use your best reliever in the highest-leverage inning you are likely to face, whatever it is called.
KBO 2026 season data through Oct 13, 2026
01
The idea
Not every inning matters equally. A clean inning with a seven-run lead changes almost nothing; a clean inning in a tied eighth can decide the game. Leverage is the name for that difference: how much the outcome of this inning moves the probability of winning, relative to an average inning.
A manager has one best reliever and a fixed number of his innings. The question is where to spend them. The save rule answers “the ninth inning with a lead of three or fewer.” Leverage answers “wherever the game is most likely to turn” — and the two answers are often different innings.
02
The math
leverage = expected |win expectancy after the half-inning − before| ÷ the same for an average half-inning
Win expectancy is computed for the KBO's nine-inning game with extra innings to eleven. The swing is averaged over every number of runs the half-inning might produce, then divided by the average swing over all the half-innings a game actually reaches. It is the definition of Tom Tango's leverage index, applied to a whole half-inning.
03
Where the game turns
Leverage of a full half-inning, by inning and the fielding side's lead.
| Inning | −3 | −2 | −1 | tied | +1 | +2 | +3 | +4 |
|---|---|---|---|---|---|---|---|---|
| 1 | 0.6 | 0.8 | 0.9 | 0.9 | 1.0 | 0.9 | 0.8 | 0.7 |
| 2 | 0.6 | 0.8 | 0.9 | 1.0 | 1.0 | 1.0 | 0.9 | 0.7 |
| 3 | 0.6 | 0.8 | 1.0 | 1.1 | 1.1 | 1.0 | 0.9 | 0.7 |
| 4 | 0.6 | 0.8 | 1.0 | 1.2 | 1.2 | 1.1 | 1.0 | 0.8 |
| 5 | 0.6 | 0.8 | 1.1 | 1.3 | 1.4 | 1.2 | 1.0 | 0.8 |
| 6 | 0.6 | 0.8 | 1.1 | 1.5 | 1.6 | 1.3 | 1.0 | 0.7 |
| 7 | 0.5 | 0.8 | 1.2 | 1.7 | 1.9 | 1.5 | 1.0 | 0.7 |
| 8 | 0.4 | 0.7 | 1.1 | 2.1 | 2.4 | 1.6 | 1.0 | 0.6 |
| 9 | 0.2 | 0.4 | 0.8 | 2.8 | 3.4 | 2.0 | 1.0 | 0.5 |
| 10 | · | · | · | 2.8 | · | · | · | · |
| 11 | · | · | · | 2.8 | · | · | · | · |
Columns are the fielding side's lead. 1.0 is an average half-inning; 2.0 moves the game twice as much. Dots are situations the game cannot reach.
Read the ninth row. Defending a three-run lead in the ninth — a save situation — has a leverage of 1.0. Defending a one-run lead there is 3.4. A tied eighth is 2.1, and a tied bottom of the ninth for the visitors is 2.8: one run there ends the game. The save rule sends the closer to the first of those and keeps him away from the last two.
04
This season in the KBO
Saves
317
Holds
746
Blown saves
201
Saves per blown
1.6
KBO bullpens recorded 317 saves and 746 holds this season, against 201 blown saves. The blown figure covers every lead a reliever surrendered in a save situation, including those of set-up men, so 61% is a floor on the league's real conversion rate rather than a closer's number.
| Reliever | SV | HLD | BSV | IP | ERA | FIP |
|---|---|---|---|---|---|---|
| Kim Jae-yoon Samsung | 35 | 0 | 5 | 64 1/3 | 3.22 | 3.75 |
| Son Ju-young LG | 31 | 0 | 5 | 50 1/3 | 3.22 | 4.23 |
| Park Young-hyun KT | 30 | 0 | 4 | 61 2/3 | 2.48 | 4.18 |
| Lee Young-ha Doosan | 26 | 0 | 4 | 68 | 3.18 | 3.20 |
| Cho Byeong-hyun SSG | 24 | 0 | 2 | 54 | 2.33 | 3.85 |
| Choi Joon-yong Lotte | 15 | 10 | 7 | 60 2/3 | 3.71 | 4.06 |
05
When the conventional wisdom is wrong
01
The three-run save
A leverage of 1.0 is close to an ordinary inning. An average reliever holds a three-run ninth most nights; the closer's best innings are wasted there.
02
The road closer in a tie
Visiting managers save their closer for a lead that has not happened yet, and lose the game in the bottom of the ninth with someone else on the mound. At 2.8, that half-inning is among the most important in baseball.
03
The seventh can be the ninth
A one-run lead in the seventh carries a leverage of 1.9; with the heart of the order due up it is higher still. The fireman — the best arm, used when the fire starts — is an older idea than the closer, and a better one.
04
Eleven innings means no tomorrow
In the KBO the game stops after eleven. In a tied tenth or eleventh there is no later inning to save anyone for; every reliever not used is simply unused.
This leverage is computed for whole half-innings from the start, bases empty, using league-average clubs. Tom Tango's leverage index works plate appearance by plate appearance, base-out state included, and needs play-by-play to calibrate, which the KBO does not publish. The platform's pitcher WAR does not weight relief innings by leverage, so it understates closers.