Key takeaways
- Expect a routine sacrifice bunt to lower your expected runs for the inning; in FanGraphs' run expectancy table, a successful bunt from first base with nobody out costs about 0.19 runs.
- Consider the bunt only where the table makes it nearly free, such as runners on first and second with nobody out, or where one run matters more than the inning's total.
- Attempt a steal only when the runner's realistic success rate is clearly above the break-even, which works out at roughly 70 to 74 percent in the table used here.
- Treat an intentional walk as a gift of expected runs to the opponent, justified only by a much better matchup, a force play or the specific game situation.
- Use run expectancy for 'how many runs' questions and win expectancy for 'how likely are we to win' questions, because they can disagree late in close games.
Usually not — at least if the goal is to score as many runs as possible. Run expectancy, the average number of runs that score from a given combination of outs and baserunners to the end of the inning, says a routine sacrifice bunt gives away more than it gains. In the run expectancy table in FanGraphs' library, a runner on first with nobody out is worth 0.831 runs; a successful bunt that moves him to second at the cost of an out leaves 0.644. The bunt "worked" and the inning got worse.
That is not the whole answer, because baseball is not always about the inning's average. Late in a close game, one run can matter more than the chance of a big inning, and that is a win expectancy question rather than a run expectancy one. The same framework also prices the stolen base and the intentional walk, which is where it is most useful.
Run expectancy in brief
FanGraphs' RE24 entry describes a run expectancy matrix: the expected runs from each of the 24 base-out states (eight arrangements of runners times zero, one or two outs) to the end of the inning. Its example matrix is set to a run environment of 4.15 runs per game, which it describes as roughly neutral for 2014. This article quotes a value from that matrix only where an argument needs it; every other figure below is our own arithmetic on it, and the full matrix is on FanGraphs' RE24 page.
To price a play, subtract the starting value from the value after the play and add any runs that scored. Two caveats travel with every number here. The matrix describes an average hitter in an average situation, and run scoring has changed since 2014 — Baseball-Reference lists 4.48 runs per team game for the 2026 regular season, so today's values would all be somewhat higher. The comparisons matter more than the decimals.
The sacrifice bunt, priced
MLB.com defines a sacrifice bunt as a bunt that advances one or more runners at the cost of the batter's own out. Assuming it succeeds exactly as planned, the change in expected runs, worked out from FanGraphs' matrix, is:
| Situation before the bunt | After a successful sacrifice | Change in expected runs |
|---|---|---|
| Runner on first, nobody out | Runner on second, one out | −0.19 |
| Runner on second, nobody out | Runner on third, one out | −0.20 |
| Runner on first, one out | Runner on second, two out | −0.18 |
| Runners on first and second, nobody out | Runners on second and third, one out | −0.02 |
Three of the four classic bunting situations lose close to a fifth of a run even when the bunt goes perfectly. The fourth — runners on first and second, nobody out — is nearly break-even, which is why it is the most defensible routine bunt.
And bunts do not always go perfectly. The lead runner can be thrown out or the bunt popped up, and every failure mode moves the result further below those figures.
Who is batting changes the price
The matrix assumes an average hitter. A sacrifice gives up that hitter's plate appearance, so the weaker the hitter, the less the bunt costs. That is the historical logic of the pitcher's bunt: MLB.com's sacrifice-bunt entry notes that pitchers in National League parks were frequently asked to bunt. Since the National League adopted the designated hitter in 2022, under the 2022–26 collective bargaining agreement, pitchers no longer have to bat in National League games. Baseball-Reference's league totals show sacrifice bunts down from 1,667 in 2011 to 766 in 2021, then 390 in 2022. They have risen in each season since, to 682 in 2026, but remain well under half the 2011 figure.
When one run is the goal
Run expectancy averages over every way an inning can play out, including the big innings. FanGraphs' RE24 entry is explicit that it makes no adjustment for the score or the inning — a bases-loaded situation is the same tied in the ninth as seven runs apart in the fourth.
Win expectancy fills that gap. MLB.com describes it as a team's chance of winning from a specific point in a game, calculated from the score, inning, outs, runners and run environment by comparison with similar historical situations. In the bottom of the ninth of a tie game, the question is not how many runs you will average but whether you will score at least one. Those are different measures, and a play that lowers the first can be judged on the second. That is the strongest case for the sacrifice: a specific inning, not a general strategy.
The squeeze: a bunt with a runner on third
MLB.com defines the squeeze play as a bunt with a runner on third base, scored as a sacrifice if it succeeds. The run scores, the batter is out, and the matrix gives a striking result:
- With one out: the run scores and the inning is left with the bases empty and two out. Counting the run, the batting team gains about 0.23 expected runs.
- With nobody out: the same trade leaves the bases empty with one out, and the batting team loses about 0.18 expected runs.
The same play is good with one out and bad with none. With nobody out, a runner on third is already worth more than one run in the matrix, so trading an out for a single guaranteed run gives up more than it buys. A failed squeeze is costly either way: with one out, if the runner is tagged out and the batter reaches first, the inning loses about 0.65 expected runs — almost three times what a successful squeeze gains.
The stolen base: the break-even rate
A steal is a trade-off with a known payoff on each side. With nobody out, FanGraphs' matrix values a runner on first at 0.831 runs. A successful steal of second raises that to 1.068, a gain of 0.237; a caught stealing leaves the bases empty with one out, worth 0.243, a loss of 0.588. The attempt pays only if the runner succeeds more often than 0.588 ÷ (0.237 + 0.588), about 71 percent of the time. The same arithmetic for a steal of second at each number of outs:
| Outs | Gain if safe | Loss if caught | Break-even success rate |
|---|---|---|---|
| 0 | +0.237 | 0.588 | about 71% |
| 1 | +0.155 | 0.394 | about 72% |
| 2 | +0.091 | 0.214 | about 70% |
Steals of third run from about 70 percent with nobody out to about 74 percent with two out in the same matrix. MLB.com's stolen-base entry gives a simpler rule of thumb — a stealer at 75 percent or better is typically helping his team — and makes the underlying point: a caught stealing hurts more than a steal helps.
The average stolen base success rate
Baseball-Reference's league totals put the success rate for the 2026 regular season at 3,289 steals in 4,273 attempts, about 77 percent. The big change came three seasons earlier: attempts jumped from 3,297 in 2022 to 4,369 in 2023, and the success rate rose from about 75 percent to about 80 percent. It has eased in each season since, to that 77 percent in 2026.
The 2023 jump came in the season the bases grew from 15 to 18 inches square, shortening the distance between first and second, and between second and third, by four and a half inches. MLB.com's base-sizes entry predicted the change would likely encourage more stolen-base attempts. The same season brought the limit of two disengagements — pickoff throws or step-offs — per plate appearance, with an unsuccessful third pickoff attempt sending the runner up a base, according to MLB.com's pitch-timer entry. League-wide, base stealers since then have been succeeding comfortably above the break-even rates worked out above.
The intentional walk
Since the 2017 season, MLB.com notes, a manager can signal an intentional walk from the dugout and the batter is awarded first base without a pitch being thrown. Its purposes, per MLB.com: setting up a possible double play, and avoiding an excellent hitter when a much weaker hitter, or a better matchup for the pitcher, is on deck.
FanGraphs' matrix shows what that costs:
- Runner on second, one out, walked to first and second, one out: about 0.26 more expected runs for the batting team.
- Runners on second and third, one out, walked to bases loaded, one out: about 0.17 more.
In run expectancy terms, the walk is a gift. It is justified only when the next hitter is far enough below average, or the force play valuable enough, to win the gift back — and, late in a close game, when win expectancy rather than run expectancy is the question being asked.
A checklist for any of these decisions
- What is the base-out state? Look up the before and after values; most "small ball" plays move you backward.
- Who is batting, and who is next? The matrix's average hitter is a baseline, not the player in front of you.
- What is the inning and score? If one run decides the game, think in win expectancy.
- What is the realistic success rate? A play that wins at 80 percent can lose at 65.
- What happens when it fails? The failure cases — a caught stealing, a lead runner forced out, a squeeze runner trapped — are where most of the cost lives.
Situations in our games
Baseball Career Sim 2026 resolves games through a simulator built around innings, outs, counts, runners on base and the batting order, and its own description makes the same point this article does: the count, the runners and the score all change the right answer. Its published feature list does not single out bunting or base-stealing decisions, so take this article as a guide to reading situations rather than to specific controls. Its league, teams and players are fictional.
On the management side, 162-0 Baseball Draft & GM Team counts speed among the inputs its season simulator weighs, and Franchise GM Mode has you set the batting order; how a batting order works covers why order matters less than roster quality, and the baseball draft simulator guide covers drafting for speed and defence. That game's pool is built from real players' careers, with ratings that are its own estimates, and its projections are a model's output rather than forecasts. For the probability side of all this, understanding probability in sports simulations is the companion piece; baseball stats explained for beginners covers the stolen-base rate in the context of other statistics, and our baseball hub collects the rest.
Neither game is affiliated with or endorsed by any league, team, players' association or player. MLB.com, FanGraphs and Baseball-Reference are named only to identify the rules and figures cited. Our other sports titles are listed under sports career and GM games.
Good to know
Frequently asked questions
Is bunting a good strategy in baseball?
What is the difference between a sacrifice bunt and a squeeze?
What is the average stolen base success rate?
What success rate does a base stealer need to break even?
How does the intentional walk rule work?
Sources
- RE24 | Sabermetrics Library (opens in a new tab)FanGraphs — accessed
- Sacrifice Bunt (SH) | Glossary (opens in a new tab)MLB.com — accessed
- Stolen-base Percentage (SB%) | Glossary (opens in a new tab)MLB.com — accessed
- Intentional Walk (IBB) | Glossary (opens in a new tab)MLB.com — accessed
- Win Expectancy (WE) | Glossary (opens in a new tab)MLB.com — accessed
- Base Sizes (2023 rule change) | Glossary (opens in a new tab)MLB.com — accessed
- Pitch Timer (2023 rule change) | Glossary (opens in a new tab)MLB.com — accessed
- Designated Hitter Rule | Glossary (opens in a new tab)MLB.com — accessed
- Major League Batting Year-by-Year Averages (opens in a new tab)Baseball-Reference.com — accessed
League, team and athlete names appear here only to identify the rules, records and history being discussed. Reign Creative LLC's sports games are not affiliated with, endorsed by or licensed by any league, team, players' association or athlete, and all names and marks belong to their owners.
About the author
Reign Creative LLC — The independent app studio that builds the apps covered on this site and publishes every article here. On Google Play it publishes as Reign Collective Apps.
We use AI tools to help research, draft and edit our articles. Each article lists the sources its factual claims rely on, and details about our apps come from their live Google Play listings. No article on this site has been reviewed by an independent expert unless the reviewer is named on it.
No person at Reign Creative LLC has read this article in full. It was published after automated checks and an AI editorial review, which are described in our editorial policy and are not the same as a person reading every sentence.
Spotted an error? Tell us — how corrections are handled is set out in our editorial policy.
