How 20-0 Player Ratings Work

A 20-0 rating is a draft value, not a verdict on a player's career. That distinction matters. The number has one job: help twelve picks from different positions and eras behave coherently inside this game. It does not come from the NFL, a team, a Hall of Fame ballot, or another site's proprietary rating.

Historical stat, editorial rating, simulated result

20-0 shows three kinds of numbers that should never be blended together.

• A historical stat records something attributed to a real season, such as passing yards or interceptions. It can be fact-checked and corrected.
• A player rating is our editorial 0-100 game value. Reasonable people can disagree with it.
• A simulated result is produced by the 20-0 model from your completed roster and seed. It did not happen in real life.

The rating does not claim to calculate itself mechanically from the displayed stat line. One compact line cannot capture blocking, coverage, era rules, role, postseason performance, or how a position creates value.

What the rating considers

We start with the player's identified peak season and ask five questions.

1. How dominant was the player relative to others at the same position and time?
2. Did the season combine production with efficiency, or only volume?
3. How much did awards, All-Pro recognition, and postseason performance support the case?
4. Did the player's role create value that basic box-score fields miss?
5. Where should the player sit among the other choices that can appear in the same draft?

The fifth question is the game-design layer. A rating can be defensible in isolation and still produce bad choices if every famous quarterback is a 98 while every interior defender sits in the low 80s. Ratings are reviewed within position first, then across the complete player pool.

Why cross-era normalization is judgment, not science

A 1970s cornerback and a modern quarterback did not play the same sport in every practical sense. Schedule length, passing rules, stat definitions, training, equipment, and positional assignments changed. Raw totals therefore cannot be placed on one line and compared without context.

20-0 handles this by judging dominance inside an era before comparing players across eras. That makes the draft playable, but it does not erase uncertainty. A one-point gap is a balancing choice, not proof that one historical player was objectively better. The useful signal is the band: replacement-level, strong starter, elite season, or rare all-time peak.

How a rating reaches the season model

Your six offensive contributions are averaged into an offense rating; the six defensive contributions become a defense rating. Empty slots count as 50. The model averages the two sides to face a generated opponent, then converts the rating difference into win probability with a logistic curve.

This is why balance matters. Raising a weak defense from 75 to 85 usually changes the combined roster more than raising an already excellent offense from 93 to 96. The model does not know that one player is famous. It only sees the contributions in the slots you actually filled.

The final schedule is deterministic for a saved lineup and seed. Replaying the same saved roster reproduces the same result. The projected record shown during a draft is different: it averages 200 quick simulated seasons so a single random week does not make the preview jump around.

Ratings do not determine every result

If the better-rated roster won every time, there would be no season to simulate. A rating advantage raises win probability; it does not guarantee a win. Playoff opponents receive progressively larger boosts, so a roster that cruises through the regular season can still lose late.

That randomness is bounded by the model and seed. It is not a claim that the historical players themselves would produce the generated score. Opponent names, weekly point totals, and season stories are game output.

What can be corrected

Historical fields have a right answer when a reliable record exists. If a stat, season, team, or accolade is wrong, send the player, field, proposed value, and public reference to [email protected]. Confirmed fixes go into the public correction log.

Ratings are editorial, so disagreement is handled differently. Tell us which comparable players make the number inconsistent and why. We review the position group and game impact rather than changing one famous player in isolation.

Use the number for the decision it supports

In Classic mode, the rating helps you compare candidates while preserving position and roster context. In Gridiron IQ, it stays hidden and becomes the reference behind the Scout Report. In neither mode should it replace the decision: a 96 at a position you already filled can be less useful than an 88 who closes the only hole left on defense.

That is the point of the system. The rating makes choices comparable; your roster construction determines whether those choices fit together. Open the full methodology for every published constant, or start a draft and test the tradeoffs directly.