How OLC Digital Ratings Work

This page explains how an OLC rating is calculated, how predictions and championship projections are built from it, and what the numbers can and can’t tell you.

The Rating Scale

Every OLC rating sits on a 0–100 scale. In practice, most teams in a given sport land somewhere between roughly 20 and 85. There’s no fixed number that means “tournament-worthy” across every sport — ratings are relative to the field, and the gap between two teams’ ratings is what predicts a matchup between them, not either number in isolation.

The Four Rating Components

An OLC rating is a weighted blend of four independently-scored components, so no single component can carry — or sink — a team’s rating on its own:

Performance measures how convincingly a team has won or lost, averaged across every game it has played. A blowout win scores higher than a narrow one; a true road result is worth slightly more than the same result at home. This component is capped on both ends, so a single lopsided score can’t separate a team from the pack the way it might in a raw scoring-margin stat.

Strength of Schedule (SOS) measures how difficult a team’s opponents have been, using the same two-layer approach behind the NCAA’s long-standing RPI: the average winning percentage of a team’s opponents, and the average winning percentage of those opponents’ opponents — which rewards a tough schedule even against teams that don’t individually have gaudy records.

Resume Bonus rewards beating good teams and penalizes losing to weak ones — the “good win / bad loss” logic behind modern committee tools like the NCAA’s NET Quadrants. Opponents are tiered using each team’s record and schedule strength to date, computed in a way that doesn’t create a circular dependency on the final rating itself.

Location-Weighted Win % adjusts a team’s winning percentage for where each game was played — a true road win counts for more than a home win, and a home loss counts for more against a team than the same loss on the road.

These four components are combined into the Composite Rating using a fixed weighting — Performance and Strength of Schedule carry the most weight, with Resume Bonus and Location-Weighted Win % adding a smaller adjustment — so a team can’t rank highly purely by running up the score against weak opponents, and can’t rank highly purely by scheduling cautiously. The exact weighting is tunable per sport and reviewed after each season, not changed mid-season.

Game Results

Only the final score and where the game was played feed into a team’s rating — there’s no adjustment for injuries, weather, or other context outside the box score. That keeps the rating auditable: anyone can check a team’s rating against its actual schedule of games and see where each component came from.

Margin of Victory

Margin matters, but only up to a point. Each result is converted to a bounded score rather than an unbounded point differential, specifically so that a lopsided win isn’t worth meaningfully more than a comfortable one. That cap is what keeps a single blowout from letting a team separate itself from the pack, and it discourages running up the score.

Home-Field Advantage

Game predictions account for home-field advantage directly: the home team’s rating gets a fixed bonus before the two teams’ ratings are compared, reflecting the real, measurable edge of playing at home.

Early-Season Ratings

Treat early-season ratings with real caution. With only a handful of games played, a single result can swing a team’s rating sharply — that’s true of every rating system, including the NCAA’s NET, and it isn’t a flaw in the formula so much as a reflection of how little data exists that early. Ratings become more stable, and more meaningful, as the season connects more of the field through common opponents.

Divisions and Leagues

Each sport on this site is rated within its own division or league — for example, FBS and FCS football are calculated as separate pools, each on its own 0–100 scale. A team’s rating reflects its performance against the opponents in that same pool.

Game Predictions

For any matchup between two rated teams, OLC generates a win probability using the difference between the two teams’ ratings, adjusted for home-field advantage, run through the same probability formula used throughout the site. A larger rating gap produces a more lopsided probability; two closely-rated teams produce something close to a coin flip.

Championship Projections

Each covered sport’s projection reflects that sport’s actual qualification structure — automatic bids, at-large selections filled by rating among remaining eligible teams, and bracket seeding — rebuilt fresh from the current ratings and standings each week. The specific rules vary by sport to match how that sport’s actual championship works.

Football sports carry two separate projection views, and they answer two different questions:

Bracketology is the field as it stands today — seeded strictly off results already on the board. It does not know or guess about any game that hasn’t been played yet, so early in the season it can look thin or even empty simply because too few results exist to seed a full field.

Predicted Bracketology is the field as OLC currently projects it will look once the season plays out. Every remaining game on every team’s schedule is run through the same win-probability model used for individual game predictions, those simulated results are folded into a projected final rating, and the field is seeded off that — so it reflects where a team is trending, not just where it sits today. This is the only view available before a sport has played enough games to seed a real-results field, and it keeps updating each week as actual results replace simulated ones.

Soccer and volleyball currently show only the real-results view (labeled simply “Bracketology”) each week — a full remaining-schedule simulation isn’t built for those sports yet.

Updates

Ratings, predictions, and projections recalculate after each week’s results are entered — see the About page for the typical turnaround.

Limitations

OLC ratings measure team performance based on the results available at the time and should not be interpreted as a guarantee of a future outcome. They are one way of quantifying team strength from game results, not the only valid way, and reasonable methodologies can and do disagree, particularly early in a season before enough games have connected the full field. A win probability is exactly that — a probability, not a prediction of the actual outcome, and even a heavy favorite loses some share of the time.