What's a Caspar Score?
The Caspar Score is the one number we rank every team by. The higher the score, the higher the rank. It starts at zero and adds up three things.
- 1
Did you win? (Record)
Every win adds 1 point, and every loss takes away 3. A loss costs three times what a win earns, so a 7-1 team starts with 4 points (7 minus 3).
- 2
Who did you play? (Schedule strength)
We rank all 138 FBS schedules from toughest to easiest. The toughest schedule earns about 14 bonus points, and the easiest earns almost none. Playing good teams counts, even in a loss.
More on schedule strength - 3
Who did you beat? (Quality wins)
Beating a strong team by a lot earns big points. Beating a weak team barely moves the needle, and losing to a weak team really hurts. This part counts double, because good wins are the best proof a team is good.
More on quality wins
Then the tiebreaker that settles close calls. If you beat a team and your scores are close (within 1.5 points), you rank ahead of that team. You won on the field. How head-to-head works
Why "Pre"? Early in the season we can't judge opponents by their records yet, so we measure how strong each opponent looked before the season: recruiting, returning players, transfers, and the head coach's track record. That preseason rating matches end-of-season power ratings pretty closely, which is why we trust it. About the preseason rating
For the number fans
Caspar Score = W - 3L + (139 - SOS Rank)/10 + 2 x SOW
Then a head-to-head check settles close calls.
Schedule strength: who did you play?
Before the season, we give every FBS team a strength rating (more on that below). A team's schedule strength is the average rating of the FBS opponents it has played so far. Then we line up all 138 schedules from toughest to easiest. The toughest earns about 14 bonus points, and each step down the list costs a tenth of a point, so the easiest earns almost nothing. Games against FCS teams don't count, so padding a schedule doesn't help.
Quality wins: who did you beat, and how?
Every win earns points based on two things. First is how good the opponent was. Strong teams are worth far more than weak ones, because we square their rating, so beating a top team counts several times more than beating an average one. Second is how much you won by. A bigger margin earns a little more, up to 21 points, and anything past that earns nothing extra, so there's no reward for running up the score. Losses take points away. A loss to a weak team costs the most, a close loss to a great team costs very little, and a blowout costs more than a close game. This part counts double in the Caspar Score.
Head-to-head: you won on the field
Numbers can't see everything, so we let actual games settle close calls. If you beat a team and it's no more than 1.5 points ahead of you, you move directly above it, no matter how far apart you were. We check the oldest games first and repeat until the order stops changing.
The preseason rating: how strong was each team going in?
Early in the season, records don't tell us much about opponents, so we measure strength with four things known before kickoff:
- Recruiting (25%): the 247Sports Composite recruiting class rankings over the last four years.
- Returning players (25%): how much of last year's production is back, from Bill Connelly's returning production numbers.
- Transfers (20%): the 247Sports transfer portal rankings.
- Head coach (30%): see below.
We put all four on the same scale and blend them. The rating lines up closely with the major end-of-season power ratings, so it's a fair way to measure who your opponents really were.
Head coach (30%): why the coach matters so much
Players come and go every year, but the head coach stays. The coach decides who gets recruited, who gets developed, which transfers come in, and how the team plays every Saturday. That's why a coach's career record is one of the best clues to how good a team will be, and why it's the biggest single piece of our preseason rating.
Here's how we measure it:
- Career winning percentage. We start with every game the coach has led as a head coach, not just at the current school. A coach who has won everywhere tends to keep winning.
- Recent seasons count double. Coaches change over time, so the last five years tell us more than what happened 15 years ago.
- Level of competition. Winning in a power conference is harder, so those seasons get a small boost, and seasons outside the power conferences get a small cut.
- Protection against flukes. A coach with only one or two seasons might have had one lucky year. So we pull short track records toward average until there are enough games to trust them. Long track records barely move.
- New head coaches start at average. With no record to judge, we don't guess high or low.
The bottom line is that a proven coach with a long winning record lifts a team's rating, and an unproven coach starts in the middle until the results show otherwise.
Why this is a sound way to rank teams right now
- Record comes first, as fans and committees expect, but schedule and quality of wins can move teams across record lines.
- Margin is capped at 21 to reward winning without rewarding running up the score.
- Preseason opponent strength is independent of in-season results, which avoids circular reasoning. With the final weights it tested at about a 0.88 rank correlation with the average of current ESPN FPI and TeamRankings power ratings.
- Nothing is forced. Miami came out number one with no tuning.
Honest limits
- Miami leads UCLA by only 0.18 points.
- Army, Navy, and Air Force transfer portal data was not available, so those teams were given the average.
- Only FBS vs FBS games count.
Sources
ESPN (standings, FPI, schedules), TeamRankings, NCAA.com, 247Sports, Bill Connelly/ESPN returning production, and the Wikipedia list of head coaches. All data as of November 4, 2026.