The Victory Index is one score from 0 to 100. Five components go in, each carrying a published weight that changes by league. Every ranking row on this site shows its own breakdown, so you can always see which part of the formula you'd argue with.
The five inputs are described in full below. How each one is weighted is proprietary and varies by league — it is the part of this system that took the longest to get right.
Win percentage, with ties counted as half a win, plus a bonus of up to 15 points for beating teams that are themselves highly rated. Beating good teams is the point; running up a record against a soft slate is what the schedule component exists to catch.
points, goals, runs scored per game, converted to a league-relative percentile rather than a raw number. A team averaging 4.9 runs isn't compared to a team averaging 28 points — each is compared only to the rest of its own league.
The same scaling applied to what a team allows, inverted so that giving up less scores higher.
The last ten results, with each older game weighted 15% less than the one after it, and margin folded in so a one-point escape counts for less than a blowout. A team that won early and has since collapsed will carry a low momentum score even with a strong record.
The average winning percentage of every opponent faced. A .500 schedule is neutral; a .650 slate approaches the maximum. This is the component that separates a 10–2 team in a power conference from a 12–0 team that hasn't left its own division.
The formula is the same everywhere; the weights are not. College football leans hardest on schedule because the gap between the easiest and hardest slates is enormous. The NBA leans on efficiency, because 82 games leave very little schedule variance to price in.
| League | Winning | Offense | Defense | Momentum | Schedule |
|---|---|---|---|---|---|
| 🏈 National Football League | 34% | 24% | 22% | 13% | 7% |
| 🏀 National Basketball Association | 32% | 26% | 24% | 13% | 5% |
| 🏀 Women's National Basketball Association | 33% | 25% | 24% | 13% | 5% |
| 🏒 National Hockey League | 36% | 22% | 24% | 13% | 5% |
| ⚾ Major League Baseball | 30% | 27% | 25% | 13% | 5% |
| ⚽ Major League Soccer | 34% | 23% | 23% | 14% | 6% |
| 🏈 College Football | 33% | 23% | 19% | 10% | 15% |
| 🏀 NCAA Men's Basketball | 30% | 24% | 21% | 11% | 14% |
| 🏀 NCAA Women's Basketball | 30% | 24% | 21% | 11% | 14% |
| ⚾ NCAA Baseball | 31% | 26% | 22% | 11% | 10% |
| 🥎 NCAA Softball | 31% | 26% | 22% | 11% | 10% |
A lead is not a win, so we only credit part of it. The live adjustment takes the current margin, normalises it against a typical winning margin for that sport, and multiplies by how much of the game has been played. A two-goal lead in the first period barely moves the number. The same lead with four minutes left moves it a lot. When the game ends the adjustment disappears and the result enters the season totals properly.
Small samples produce silly rankings. Until a team has played roughly a third of its season, its score is pulled toward the league average — up to 30% at the very start, tapering to nothing. A 2–0 team will not sit above a 9–1 team on this site.
Scores, standings and schedules come from live sports feeds and refresh continuously: every 20 seconds while a game is in progress, every few minutes otherwise, and a full rebuild of every league's ratings runs overnight. Feeds occasionally disagree with the official record; when they do, the official league record wins and the rating is corrected on the next rebuild.
The Victory Index is a description of what has already happened, not a prediction and not betting advice. It has no prior for reputation, recruiting rank, preseason polls or last year's result. A team starts each season at zero games played and is rated on those games alone.
Spotted something that looks wrong? Tell us — the correction and the reasoning both get published.