Understanding CalPreps Rating Methodology

How computer power ratings provide objective, unbiased team rankings

The Foundation: 100% Objective

The most important principle of the CalPreps computer power rating system is that it is completely objective. The system considers only game results—nothing else.

The system DOES NOT consider:

School size
Division placement
League affiliation
Historical performance
Geographic location
School popularity

Just teams with results—cold, hard, unfeeling, but as accurate, objective, and fair as possible.

Early Season Adjustment (11-Man Football Only)

For 11-man football ratings, pre-season projections based on returning players are factored into early-season ratings to improve accuracy. This factor is completely eliminated as the season progresses, ensuring mid-to-late season ratings are purely objective.

Two Calculation Methods

CalPreps ratings can be calculated in two ways. We'll explain the version with margin of victory first, as it's easier to understand.

With Margins

Rating differences roughly equal point spreads. An 80-rated team should beat a 60-rated team by about 20 points.

  • Considers margin of victory
  • Win/loss still most important
  • Diminishing returns on blowouts
  • Cutoff point for margin counting

Without Margins

All wins are counted equally regardless of score. Only who you beat matters.

  • A win is a win
  • No margin consideration
  • Focus purely on wins/losses
  • Opponent quality is key factor

How the System Works (With Margins)

Example Starting Ratings

Team A:+10
Team B:0
Team C:-5
Team D:-8
Team E:-10

1 System Compares Actual vs. Expected Results

The system sorts through all season results and compares what happened to what "should have" happened based on the ratings. For example, if Team A (rated 10) played Team C (rated -5), Team A should win by about 15 points.

Example outcomes for Team A:

  • vs. C (lost or narrow win): Rating hurt 📉
  • vs. D (won by 15): As expected, no change ➡️
  • vs. B (won by 22): Better than expected, rating helped 📈
  • vs. D (won by 10): Worse than expected, slight ding 📉

2 Aggregate Performance Determines Adjustment

The system calculates how much better or worse the team performed than expected across all games, divides by games played, and adjusts the rating accordingly.

Example: If Team A performed an average of 2 points worse than expected, their rating drops from 10 to 8.

3 Iterative Process Until Stability

After adjusting all teams, the system starts over with the new ratings and repeats the process. This continues until ratings stop moving—when they've settled at their proper levels.

4 All Teams Start at Zero

Despite our example above, all teams actually start at 0 with no bias. The system runs continuously from this baseline until ratings stabilize at their correct values.

Important Nuances (With Margins)

Win/Loss is Most Important: Even with margins, the actual result (win or loss) always carries the most weight.

Diminishing Returns: Teams aren't fully credited for massive blowouts against weak opponents.

Margin Cutoff: There's a maximum margin beyond which additional points don't count.

Win Minimum: A 1-point win is valued much higher than a 1-point loss (not barely better).

Example: Ratings Without Margins

When margins aren't used, the process is the same but simpler—all that matters is who you beat and who you lost to.

Season Data

Team W-L Games
A 3-0 wins: C, D, E
B 2-1 wins: F, F / loss: E
C 2-1 wins: D, E / loss: A
D 1-2 win: F / losses: A, C
E 1-2 win: B / losses: A, C
F 0-3 losses: B, B, D

Analysis

  • Team A is clearly the best (3-0 with quality wins)
  • Team C (2-1) beat stronger opponents than Team B, whose wins came against 0-3 Team F
  • Team E (1-2) should rank above Team D because E beat B, and both have identical losses
  • Team B (2-1) actually belongs below the 1-2 teams—their record overstates their performance

Final CalPreps Ratings

Team A (3-0) 22.2
Team C (2-1) 13.2
Team E (1-2) 3.2
Team D (1-2) -0.9
Team B (2-1) -8.7
Team F (0-3) -22.1

Note: Team B's 2-1 record is deceptive—their wins against the worst team and loss to a 1-2 team places them 5th overall, below both 1-2 teams.

Early Season Ratings (11-Man Football)

To address early-season inaccuracies and provide ratings from day one, the system uses adjusted starting points for 11-man football teams:

Pre-Season Starting Point: Teams start at adjusted positions based on previous year's performance and returning players (manually researched or calculated from roster data).

Quick Wash-Out: As games are played, the starting point's influence diminishes rapidly. Small adjustments (+3, -1.5) wash out very quickly.

Mid-Season Purity: By mid-season, the pre-season adjustment is completely eliminated, and ratings are purely based on current-year results—exactly like the original system.

For Playoff Selection: States using CalPreps ratings for playoff purposes are not affected by pre-season adjustments, as ratings are purely objective by the time playoffs begin.

Why CalPreps is Superior

This system is vastly more accurate than simplified point systems used by some states:

❌ Simple point systems (e.g., 3 points for beating large school, 2 for medium):

  • • Don't account for actual team strength
  • • Treat all "large schools" the same
  • • Can't adjust for schedule difficulty
  • • No iterative refinement

✅ CalPreps system:

  • • Considers actual strength of every opponent
  • • Self-corrects through iteration
  • • Accounts for full schedule context
  • • Proven highly accurate for predictions

Learn About Other Rating Systems