This dashboard updates independently. The Weekly Brief preserves our picks, reasoning and decision-time odds.
Reloads the last published snapshot. It does not pull a live feed from the source.
Scores & Schedule
Final Results · In-Progress Snapshots · Upcoming Games
Choose a week, then what to show. Tap a card to open that game’s detail page. In-progress scores are periodic snapshots, not a live play-by-play.
Tap a game card to open its detail page. Market, spread, Adj. EPA, ratings, and injuries live there — the card stays a compact score. Final scores appear when the scoreboard confirms completion. In-progress scores show the last checked game clock; they are periodic snapshots, not live play-by-play. Named-book WP on the Weekly Brief is separate from anonymous schedule estimates here.
Records & Ratings
One Row Per Team · Descriptive + Predictive
Sort the same table another way. Open a team name for its sample and EPA numbers. None of these ranks rewrite our pick odds.
Tap a team name to open the detail panel (record, raw + Adj. EPA, next opponent / anonymous WP, and nfelo/FPI cross-checks); tap again to close. Records cover settled 2026 games. Efficiency ranks use n = games in the stats sample. Adj. EPA is opponent-adjusted and sample-shrunk; nfelo and FPI are dated power ratings. None of them are blended into market win probabilities or pool PS.
How to read this table Records, EPA ranks, and power ratings
Open this if a column name is new to you. The table itself does not change our pick odds.
Records count settled 2026 results only (wins–losses–ties). n is the efficiency sample — often smaller than the record sample early in a week while team stats catch up, and tiny in Week 1. Off / Def / Overall are raw EPA ranks among teams that have stats: #1 is best. They describe completed plays against the opponents actually faced. A small early-season sample is a snapshot, not a ranking of true talent.
Adj Off / Adj Def / Adj Overall are opponent-adjusted EPA ranks (Adj. EPA). They start from the same team-game EPA/play, then a ridge offense/defense split credits or debits opponent quality and home field. Estimates shrink toward league average (0) while n is small (weight n/(n+4) on the sample). Compare Adj. Overall with raw Overall to see whether a hot EPA start came against a soft slate. Adj. EPA is still descriptive efficiency — not a pick probability and not blended into market win chances or pool Pick Strength.
nfelo and ESPN FPI are predictive power ratings from dated public sources. Use them to challenge the chalk — where a market favorite looks weaker, or an underdog stronger, than the models. They are not pick probabilities and they are not blended into market win chances or pool Pick Strength.
|nfelo − FPI| highlights disagreement between the two models. A gap is a research cue: look at the matchup, sample, and availability. It is not betting-value proof and it does not mean the market is wrong.
The Weekly Brief remains the numerical Survivor benchmark. Market win probabilities and frozen picks there stay independent of this dashboard. Ratings never rewrite those numbers.
Player Leaders by Team
Choose a team to see its settled passing, rushing, and receiving leaders. This is production so far, not a depth chart.
Usage rates (Att/G, Tgt/G) come from settled production already in this feed. Games played versus the team sample is participation, not an official injury designation. Missing values show —.
Team Outlook
Full Schedule · Results & Estimates
Choose a team and whether to see the rest of the season or every week. ~ estimates come from anonymous schedule prices, not our named-book board.
~ probabilities come from anonymous schedule moneylines after removing the margin, conditional on no tie. The source does not identify a bookmaker or quote time. These are reference estimates; named-book evidence for our selections stays in the Weekly Brief.
Sources & Update Method Freshness, Coverage and Corrections
Make a Suggestion or Request Specific Data Help Shape the Next Report
A matchup stat we missed? A better way to compare teams? Tell us what would make this more useful.