How it works

Every weekly projection for week W is built using only data available before week W kicked off. Trailing form and usage features are computed over prior games and explicitly exclude the week being projected; the game environment comes from pre-kickoff market lines. When projections are evaluated over past seasons, it is a walk-forward backtest: each season is projected by a model trained only on earlier seasons, exactly as it would have run live.

That means nothing on the site is a model graded on games it already saw. When you browse a past week, you're seeing what the model would have projected before those games were played.

One deliberately simple model per position: a ridge regression with learned (not hand-set) weights over point-in-time features, trained walk-forward so each season is projected using only earlier seasons. Players are ranked within their position by projected PPR points for the week. The features:

• Recent form: trailing PPR over the last few games
• Target share: trailing share of team targets
• Opportunities: trailing carries + targets per game
• Snap share: trailing share of offensive snaps
• Game environment: the team's implied point total from pre-kickoff market lines
• Defense vs. position: trailing PPR the opposing defense allows to the position

Every trailing feature is shifted so the week being projected is never part of its own inputs. Players with too little history get no projection rather than a made-up one.

The site has two kinds of numbers, and we keep them separate on purpose:

• The weekly projections. These look forward, so they are strictly point-in-time: built before kickoff, never retrofitted after the fact.
• The Stat Explorer leaderboards. These describe what already happened (usage, efficiency, expected-TD gaps). They are useful context, but a leaderboard rank is a description of the past, not a prediction.

If a number on this site looks forward, it was computed point-in-time with no hindsight. If it doesn't, it's describing the past.

Everything is built from a single source: official NFL box scores, snap counts, play-by-play, betting odds, and injury status. One source keeps the pipeline reproducible and the point-in-time guarantee auditable.

It also sets honest limits. That source has no air-yards feed and no official EPA, so the model does not use them, and metrics we derive ourselves from the play-by-play (like the expected-points cuts in the Stat Explorer) are labeled "our EP/EPA" to make clear they are our reconstruction, not the official figures. When a stat can't be built cleanly from this source, we'd rather not show it than fake it.