Pythagorean Record Calculator
The Pythagorean formula turns points scored and allowed into an expected win percentage — a "true talent" record based purely on scoring margin. Enter your sport, scored/allowed totals, and games played to get your expected record, then compare it to your actual wins with the luck meter.
Actual vs. expected wins
Luck meter verdict
How far your actual wins sit from your Pythagorean expectation
Why scoring margin predicts wins better than record alone
Over a full season, a team's win total is strongly related to how many more points it scores than it allows — far more so than to how many one-possession or one-run games happened to bounce its way. The Pythagorean formula (named for its resemblance to the Pythagorean theorem) converts total scored and allowed into an expected win percentage, giving a "what the underlying performance says this team's record should be" number that's often a better predictor of future performance than the actual record itself.
How it’s calculated
Expected win% = Scoredexp ÷ (Scoredexp + Allowedexp). This tool uses three sport-specific exponents from established sabermetric and analytics convention: MLB: 1.83 (the refined exponent from Bill James's original baseball Pythagorean method, found by researchers to fit actual results better than a flat exponent of 2), NBA: 14 (popularized by analyst John Hollinger for basketball's much higher, lower-variance scoring), and NFL: 2.37 (a widely used football-analytics convention, sitting between baseball's and basketball's exponents to match football's scoring profile). Projected record splits expected win% across your entered games; the luck meter compares your optional actual-wins entry to the expected win total.
This is a scoring-margin model, not a full simulation — it does not account for strength of schedule, injuries, bullpen or clutch performance, or any single-game context. Treat the expected record as a "true talent" reference point, not a guarantee of what a team will actually do.
Worked example
An MLB team that scores 800 runs and allows 700 over 162 games has an expected win percentage of 8001.83 ÷ (8001.83 + 7001.83) = .561 — a projected record of about 91-71. If that team actually finished 91-71, the luck meter reads "about right"; if it finished 87-75 instead, it outperformed its scoring margin by roughly 4 wins the wrong way (unlucky), while a 96-66 finish would flag as notably lucky.
Common mistakes
- Using the wrong sport's exponent — NBA's 14 and MLB's 1.83 are not interchangeable; each is calibrated to that sport's scoring pattern.
- Treating a big gap between actual and expected wins as permanent skill rather than a signal that often regresses toward the mean the following season.
- Forgetting that a small sample of games (a partial season, a short stretch) makes both the expected and luck-meter numbers much less reliable.
Where it is used
- Identifying playoff teams that got there on scoring margin (sustainable) vs. close-game luck (less sustainable).
- Front-office and analytics discussions about whether a team's record reflects its true quality.
- Fantasy and betting research that looks past raw win-loss record to underlying performance.
Frequently asked questions
Where does the Pythagorean formula come from?
Bill James introduced the original baseball version in the 1980s, noting that a team's win percentage tracks the ratio of runs scored squared to runs scored squared plus runs allowed squared — hence "Pythagorean." Sabermetricians later found that an exponent around 1.83 (rather than exactly 2) fits actual MLB results better. The same idea was adapted for other sports: analyst John Hollinger popularized an NBA version using an exponent of about 14, and football analysts commonly use an exponent around 2.37 for the NFL.
Why do MLB, NBA, and NFL use such different exponents?
The exponent controls how sharply small scoring differences translate into win-percentage differences, and that sensitivity depends on a sport's typical scoring volume and game-to-game variance. Basketball's high, low-variance scoring (both teams routinely in the 100s) needs a much larger exponent (~14) to make small point-differential edges translate into realistic win rates, while baseball's low, high-variance scoring (games often decided by 1-2 runs) fits a small exponent (~1.83), and football sits in between (~2.37).
What does it mean if my actual wins are way above my expected wins?
It usually means a team has won an unusually high share of close games — one-run baseball games, games decided by a single score in football, clutch finishes in basketball — which analysts generally treat as harder to sustain than the underlying scoring margin suggests. A gap of 4+ wins above expected is often flagged as a regression candidate: teams that outperform their Pythagorean record by a wide margin tend to perform closer to their expected win rate the following season.
Does a good Pythagorean record guarantee a good real record?
No — it's a projection based purely on scoring margin, not a guarantee. Bullpen quality, clutch performance, injuries, and plain randomness in close games all cause real records to diverge from the Pythagorean expectation, sometimes significantly and for a full season. It's best used as a "true talent" signal alongside the actual record, not as a replacement for it.