Comparing fantasy teams across leagues and scoring formats

You scored 142 in your 10-team PPR league. Your friend scored 97 in a 14-team standard league. Who had the better week? The honest answer is "it depends on everything except those two numbers", and that is the problem this guide is about.

Why raw points don't travel

• Scoring rules. PPR, half PPR and standard give very different totals for the same players. Six-point passing touchdowns and tight end premium push them further apart.
• League size. In a 10-team league every starting lineup is stacked; in a 14-team league talent is spread thinner, so the same performance produces fewer points.
• Starting slots. Superflex, 2 QB and extra flex spots add points that a one-QB league never sees.
• The NFL week. Some weeks every offense scores; some weeks it rains in four stadiums at once. A 120 can be great one Sunday and ordinary the next.

Percentages of the league average help a little, but they still inherit the strength of your particular league. You need a reference that is the same for every league with the same rules.

The idea behind a format-neutral score

Instead of comparing teams with each other directly, compare each team with what an ordinary team would score in an equivalent league that same NFL week, then put that gap on a standard scale. Two teams in completely different formats can then be compared by how far above or below ordinary each one was.

How Matrix Score works

Each week FantasyMatrix scores recorded NFL production under your league's exact rules and works out the player supply available for the legal starting slots across your league. A frozen statistical model, learned from external redraft leagues, turns that into an estimated ordinary mean and spread for your format that week. Then:

Matrix Score = 100 + 15 × (actual starter points − reference mean) ÷ reference spread

Example (illustrative numbers): Your starters score 120 points. The reference mean for your format that week is 100 and its spread is 20. Matrix Score = 100 + 15 × (120 − 100) ÷ 20 = 115.00. A team in a different format with a different points total can land on the same 115 if it was the same distance above ordinary.

So 100 is a typical team in an equivalent league that week, and 15 points is one standard step. Season scores average the same completed weeks. Wins, schedules and lineup efficiency are not part of the headline score: it measures what your starters produced.

Percentiles: how many teams you beat

FantasyMatrix also shows an estimated percentile: the share of teams in equivalent formats a score would beat, based on the model's normal curve. A season average spreads less than a single week, so the estimate accounts for the number of weeks. It is a model estimate, not a rank among FantasyMatrix members; the Leaderboard is the actual rank.

The honest limitations

• It is an estimate. Several points of difference can be noise, especially in small leagues.
• The reference is learned from a sample of external leagues and inherits their biases.
• Formats from 4 to 16 teams with supported rules are scored; unusual formats can be unavailable rather than guessed.
• A score of 120 is 20 index points above the reference, not "20% better" and not proof of manager skill.

The full detail, including validation results that did not pass, is on the methodology page.

Try it on your team

Create a free account and connect your team to get your Matrix Score and join the cross-league Leaderboard.