4th & Data
Receipts

The scoreboard

Every SMASH and FADE on the matchup board gets frozen before the first kickoff of the week. The next Tuesday we grade each one against what actually happened and publish the hit rate. We cannot edit a frozen call, and you can check the timestamp.

Weeks on the board

This week's locked calls

Start/Sit report card

The whole weekly start/sit board is frozen on Tuesday and graded against actual points the next Tuesday. The question we grade is simple: did our matchup adjustment rank players better than the plain season projection?

Pre-Snap Read report card

The Pre-Snap Read makes a dozen claims about every slate before lock: which chalk is restrictive, who benefits if it fails, which home-favorite backs are under-owned, which games and pairings carry the ceiling, which players earn a core spot. Each one is graded against real points the next Tuesday.

Week 1first grades land the Tuesday after the games

Lineup autopsy

Our own DraftKings entries, scored after the slate against what the engine believed at lock: each lineup's simulated range, its predicted chance to cash, and where the real score landed. Below the 10th percentile of its own range is variance. Inside the range and still short is process. Over a season the table is the calibration curve for the simulator, which is the only honest scorecard a simulator gets.

Showdown lanes, graded

When a Showdown standings export is banked, every model lineup gets its real score and finish against that field, by lane (win / unfound) and by game script, plus whether the game went under its line.

No Showdown export banked yet.

Top-10 archive

The top ten lineups of every contest export we bank, tagged with the shape (quarterback stack size, bring-back, FLEX position), projected ownership at lock and the sub-5% and 20%+ counts. Study the winners backward.

Track record

Every graded week this season, plus last season replayed week by week with the same engine and only the information known each Tuesday. One rule for a hit: a SMASH group scored more than its own average, a FADE group scored less.

How a call gets graded

SMASH

We said that position on that team should beat its own average this week. Hit means the group's fantasy points came in above its average (this season's games blended with last season). Miss means they did not.

FADE

We said the opposite. Hit means the group came in below its average.

The bar

Groups beat their own average about 46% of the time (big weeks pull averages up), so 46% is the coin flip here, not 50%. Anything we cannot beat that on, we stop publishing. That rule applies to us the same as it applies to anyone we audit.