How NFL Analytics Works
How the BBMI NFL ratings work — and the deliberate decision not to sell NFL picks.
Why no NFL picks?
The NFL is the most efficient, most heavily-bet market in sports. Thousands of sharp models and millions of dollars price every line within hours. We don’t believe we have a durable edge there — and we won’t sell picks we can’t stand behind. So NFL is analytics only: ratings, projections, and playoff odds, with the Vegas line shown for context.
Where we do have an edge — college sports’ thinner markets — we publish picks and grade them publicly. See the public track record.
The BBMI Composite Rating
The BBMI Composite Rating is a single neutral-field number: how many points above or below average an NFL team is. It blends three independent rating systems, each capturing different information about team quality:
- Elo ratings (40%) — from the open-source nfelo project. Score-based power ratings with QB adjustments, home field, and rest days. Captures who’s winning and by how much.
- EPA efficiency (40%) — computed from nflverse play-by-play data. Expected Points Added per play measures the quality of each offensive and defensive snap, independent of the final score.
- Opponent-adjusted box scores (20%) — from Pro Football Reference game logs with iterative opponent-quality adjustment. Captures schedule-adjusted performance.
Each source is normalized and weighted. The composite updates weekly during the NFL season.
What is EPA?
EPA (Expected Points Added) measures the value of each play in terms of expected points. A first down on your own 20-yard line is worth about 0.5 expected points. A touchdown from the 5-yard line is worth about 6.3. Every play either adds or subtracts expected points based on the down, distance, yard line, and result.
EPA/play tells you how efficient a team is per snap. A team with +0.10 offensive EPA/play gains about 0.10 expected points more than average on every play — over 65 plays per game, that’s about 6.5 points better than average. It’s the most granular publicly available measure of team quality.
The inputs
Offensive EPA
Expected points added per play, adjusted for the quality of defenses faced. Passing and rushing are weighted separately.
Defensive EPA
The same, from the defense’s side — how many expected points a unit prevents versus an average offense.
Strength of schedule
Every game is weighted by opponent quality, so a strong record against a hard slate outranks one padded against weak teams.
Situational context
Home field, rest (short weeks, byes), and key-player injury impact adjust the projected margin for a given game.
From ratings to projections
For any matchup, the rating gap plus home field gives a projected margin and a win probability. To project the full season — playoff odds, seeding, Super Bowl probabilities — the model simulates the remaining schedule 10,000 times, resolving tiebreakers and playing out the bracket in each run. Those simulations power the Season dashboard and Playoff Pulse.
Data sources
| Source | Data | Update |
|---|---|---|
| nflverse | Play-by-play with EPA, CPOE, WPA | Weekly |
| nfelo | Elo ratings with QB model | Weekly |
| Pro Football Reference | Team game logs and box scores | Weekly |
| Odds API | Vegas spreads and totals (display only) | Pre-game |
| Open-Meteo | Game-day weather for outdoor stadiums | Game-day |
How we know the edge isn’t there
BBMI conducted an extensive investigation into NFL spread and total prediction during April 2026, testing 12 independent approaches across four seasons of data. We tested our own opponent-adjusted model, nfelo Elo ratings, calibrated EPA projections, three-source blending, ATS performance at high edge thresholds, conditional subsets by week and spread size, and weather effects on totals.
Every test confirmed that NFL closing lines are efficient against all publicly available models. No combination of Elo ratings, EPA metrics, or opponent-adjusted statistics could consistently beat the market. The NFL has 32 teams, enormous public attention, and the sharpest betting lines in sports — the market already incorporates everything these models capture and more.
Rather than offer picks at coinflip accuracy, we provide analytical tools and data that inform your own football analysis. We believe this transparency is what sets BBMI apart.
NFL vs. our other sports
BBMI’s NCAA basketball, football, baseball, and MLB models generate validated betting picks because those markets have exploitable inefficiencies — 130+ teams, thin markets, less sharp money, and information asymmetry the model can capture. The NFL is different: 32 teams, massive public attention, and closing lines that already reflect all available information. Different markets require different honest assessments.