Ratings & Simulation

You Draft Player-Seasons, Not Players

The draft pool isn't one entry per player — it's one entry per player per season. A quarterback with a ten-year career appears ten times, and each of those seasons is rated on its own numbers: the MVP year is a different pick from the rookie year. That's what makes the pool enormous — more than 100,000 player-seasons, covering roughly 24,000 players across five-plus decades of pro football — and it's why two drafts never feel the same.

Where Ratings Come From

Each player-season's ratings are built from historical performance — what that player actually did in that specific year — weighted by position. A quarterback is judged mostly on what quarterbacks control; a lineman on line play. Eras are handled on their own terms, so a 1970s great isn't punished for playing in a lower-scoring league.

An overall rating is only the beginning. Perfect 20 includes more than 30,000 season specific profiles. Every complete profile starts with six shared traits, then adds skills built for the player's position. Some positions use as many as 18 traits, and cornerbacks and safeties have separate models because they are asked to do different jobs.

Some traits affect only the simulation. They can change how a player behaves during Hardcore play by play without increasing the overall rating. Browse the Ratings Lab, or learn how the ratings work.

Two Engines, Two Kinds of Game

The classic modes (Normal, Knowledge, the Gauntlets, and friends) resolve each game whole: the engine weighs your roster against that week's opponent — strengths, weaknesses, coaching — plus the honest dose of luck that makes football football, and a full season simulates in moments. The better team wins more often, not always. That's by design: if favorites always won, 20-0 would be arithmetic instead of drama.

Hardcore runs on a different engine entirely. Games there aren't decided — they're played, drive by drive and play by play. Individual matchups matter (your line against their front), players get hurt, backups get thrown in, and each week produces its own story with key plays and turning points. Same honest dice, far more moving parts — see the Hardcore guide.

Fairness & Repeatability

The engine is deterministic under the hood: the same roster, opponent, and dice produce the same result, and every run's outcome is recorded exactly as simulated. Nothing is nudged after the fact — no pity wins, no rubber-banding.

Drafting With This In Mind

Ratings tell you who was great; the simulation punishes rosters with a hole in them. The winning habit is drafting for the whole team — see the Normal Mode guide for how that plays out.