What is Run Differential? Definition, Formula, and Example
Run differential is a team's total runs scored minus its total runs allowed, a simple stat that predicts future winning percentage better than a team's actual win-loss record.
What is Run Differential?
Run differential is the difference between the total runs a team scores and the total runs it allows over any span of games: Run Differential = Runs Scored − Runs Allowed. A team that has scored 500 runs and allowed 450 has a +50 run differential. It is the single best simple indicator of team quality — better than winning percentage itself — because it strips out the luck embedded in close-game results.
How Run Differential is Calculated and Used
The raw math is subtraction, but run differential's power comes from its connection to Pythagorean winning percentage, Bill James's formula:
Expected Win% = RS² / (RS² + RA²)
(Modern analysts use an exponent of ~1.83 instead of 2 for better accuracy.) Plug a team's runs scored and allowed into the formula and you get the win total the team "deserves." Comparing that to actual wins reveals over- and under-performers:
- A team winning more games than its run differential predicts is getting lucky in close games and is a regression candidate.
- A team with a strong differential but a mediocre record is unlucky — and typically a smart bet to improve.
One-run games are the key mechanism. One-run record is heavily random — even great teams hover near .500 in one-run contests — so a team going 30–15 in one-run games is enjoying variance, not demonstrating skill.
Worked Example: The 2021 Seattle Mariners
The canonical modern case is the 2021 Mariners. Seattle finished 90–72, nearly making the playoffs, despite being outscored 697–748 — a −51 run differential. Their Pythagorean record was roughly 75–87. The gap came from an absurd 33–19 record in one-run games. Analysts predicted steep regression, and while Seattle stayed competitive in 2022 (this time with a positive differential backing it up), the 2021 profile is the textbook illustration of why front offices trust run differential over standings. On the flip side, the 2023 Padres went 82–80 with a +104 differential — the profile of a team far better than its record, undermined by a 9–23 one-run record.
Why Run Differential Matters
Front offices use run differential to evaluate their own rosters honestly: a .500 team with a +60 differential buys at the trade deadline, while a .500 team at −40 sells. Bettors and projection systems (ZiPS, PECOTA, FanGraphs) build offseason forecasts on runs scored and allowed, not wins. For fantasy players, team run differential signals which lineups and pitching staffs to target — hitters on high-scoring teams get more plate appearances and RBI chances; pitchers on run-preventing teams get more wins and quality starts.
Limitations and Common Misconceptions
Run differential can be skewed by blowouts. A team that wins 15–2 and loses five 3–2 games posts a misleading differential relative to its competitiveness — which is why some analysts prefer "smoothed" versions that cap single-game margins. It also says nothing about *when* runs come; bullpen quality genuinely influences close-game outcomes at the margins, so elite-relief teams can modestly and repeatably beat their Pythagorean record. And run differential is descriptive, not causal: a midseason differential includes players who may no longer be on the roster after the deadline.
Related Terms
In Legends Deck: Run differential logic underpins the simulation engine's team-strength calibration — when you pit two card lineups against each other, expected outcomes are driven by underlying run production and prevention ratings, not surface win-loss records, so genuinely strong cards win out over lucky ones across long sims.