Analyst · Week 12
The stats model is the foundation. This layer reads the week's news and forecasts, reports only what the model is not already counting, and moves the number by a capped amount — every claim linked to its source, every claim graded after the final.
Is the news layer helping?
Since the layer went live, over 403 graded claims:
- The adjusted forecast has been worse than the stats model alone — net change in squared error (Brier) of +0.121, where negative is better.
- 202 of 403 claims pushed the number toward the actual result.
- It changed the final pick in 5 games, and those flips went 4–1.
This is the whole test. News can't be backtested the way the base model was (2010–25), so the layer earns its place only forward, and only if it lowers the error over many games — not on any one result. Its job is to move the number toward the truth, not to rescue a pick: when it cut a heavy home pick by 10 points and the team still lost, the move was correct (it lowered the error) even though the pick was wrong — the base model had simply started too far above the market. Picks where we disagree with the market's final price are our weaker spot (about 44% in the backtest, versus 68% when we agree) — they're flagged “we differ” on the picks table so you can watch them.
Tune-up, Nov 10, 2026: what we switched off
These claim types made the forecast worse this season (positive Brier change in the scorecard below), so from Nov 10, 2026 they are still read and graded but no longer move the number: key starter out, not on the injury report; key starter returning; other (in-week, concrete, high confidence only); suspension; locker-room or organizational turmoil. Past picks are not changed.
| Same 148 graded games | Record | Brier (lower is better) |
|---|---|---|
| Stats model alone | 100–48 | 0.2093 |
| As published (every type) | 103–45 | 0.2105 |
| Without the switched-off types | 101–47 | 0.2081 |
A replay with the same stored claims and the same caps, those types left out (4 picks would have changed). It is a look back, not a promise: the switched-off types keep being graded and come back if they start lowering the error.
What we do with the news, step by step
This page is the news layer's public notebook. In plain terms:
- Before kickoff, every run pulls the feeds and attaches new items to the games they concern. Items published after a game has started are never read: in-game news is not a prediction input.
- The reader turns new items into claims: a type (QB change, key starter out, resting starters…), a team, a player, a direction, and a confidence.
- The filters decide, in the same run, whether a claim is applied or declined. Declined if it is below the confidence floor for its type, if the model already counts it (the player is on the injury report, the QB is already the named starter), or if it is the kind of story that has no place (offseason hires, turf, "limited in practice").
- Applied claims move the probability right then, by a fixed number of points for the type, scaled by confidence, capped. That adjusted number is the pick. "Model 74% → analyst 76%" on a game below means exactly that.
- After the final, every claim is graded, applied or declined: did it point the right way, and did it lower the error? Declined claims are graded as if they had been applied, so we learn whether the filter was right to say no.
- Once a month, with enough graded cases, the rules change for future weeks: a type that keeps hurting loses weight; a rejection reason that keeps "would have helped" gets loosened. Past picks are never touched. Every rule change is dated in the changelog.
A claim only affects the game it is attached to. It does not carry into next week's model: a starter who misses Week 2 is caught by the injury report by then, and the model counts him there.
How it works
Sources: public news feeds; the National Weather Service forecast at kickoff for outdoor stadiums; the nflverse injury report and named starters are what the model already counts. The reader (one language-model call per game, only when there is new news) returns typed claims about things not already counted. A fixed table turns each claim into points, scaled by the reader's confidence, capped per type and ±10 in total. Weather is logged at zero weight until a season of grading says otherwise.
| Claim type | Max points |
|---|---|
| starting QB change not yet in the feed | 6.0 |
| key starter out, not on the injury report | 2.0 |
| key starter returning | 1.5 |
| resting starters / playing backups | 8.0 |
| suspension | 2.0 |
| coaching change | 2.0 |
| motivation / eliminated / nothing to play for | 1.5 |
| locker-room or organizational turmoil | 1.0 |
| unusual travel or schedule disruption | 1.0 |
| weather at kickoff (logged, zero weight) | 0.0 |
| other (in-week, concrete, high confidence only) | 0.5 |
Scorecard by claim type
After each final: did the move go the right way, and did it lower the squared error (Brier)? Negative Brier change is good. Types that don't earn their weight get zeroed at season's end.
| Type | Claims | Graded | Avg pts | Helped | Brier Δ |
|---|---|---|---|---|---|
| key starter out, not on the injury report | 254 | 228 | 1.8 | 111/228 | +0.240 |
| other (in-week, concrete, high confidence only) | 85 | 78 | 0.4 | 40/78 | +0.032 |
| motivation / eliminated / nothing to play for | 34 | 31 | 1.1 | 18/31 | +0.008 |
| starting QB change not yet in the feed | 28 | 27 | 5.2 | 16/27 | -0.159 |
| locker-room or organizational turmoil | 15 | 11 | 0.8 | 5/11 | +0.026 |
| unusual travel or schedule disruption | 14 | 12 | 0.8 | 6/12 | -0.005 |
| coaching change | 7 | 6 | 1.7 | 5/6 | -0.041 |
| weather at kickoff (logged, zero weight) | 5 | 5 | 0.0 | 0/5 | +0.000 |
| suspension | 4 | 4 | 1.9 | 1/4 | +0.014 |
| key starter returning | 1 | 1 | 1.1 | 0/1 | +0.006 |
Shadow scorecard — what we declined
Rejected claims are kept at zero weight and graded as if applied. If a rejection reason keeps "helping", the rule is too strict; if "other" keeps hurting, it stays out. This is how the layer learns.
| Rejected because | Type | Claims | Graded | Would have helped | Brier Δ if applied |
|---|---|---|---|---|---|
| no ruled-out language (practice status is not 'out') | key player out | 341 | 307 | 150/307 | +0.191 |
| retired: now on the injury report | key player out | 208 | 196 | 100/196 | +0.060 |
| return of a player the model was not counting out | key player back | 163 | 147 | 84/147 | -0.115 |
| confidence 0.75 below floor for other | other | 87 | 77 | 37/77 | +0.021 |
| no ruled-out language (practice status is not 'out') | QB change | 79 | 69 | 34/69 | +0.442 |
| confidence 0.65 below floor for other | other | 39 | 36 | 19/36 | -0.009 |
| retired: read before this team's previous game (old news) | key player out | 37 | 32 | 17/32 | +0.067 |
| soft status (limited/questionable/cleared) | key player back | 28 | 25 | 16/25 | -0.036 |
| retired: read before this team's previous game (old news) | QB change | 26 | 25 | 19/25 | -0.491 |
| already on the injury report | key player out | 22 | 22 | 11/22 | +0.090 |
| confidence 0.70 below floor for other | other | 22 | 20 | 10/20 | +0.004 |
| confidence 0.30 below floor for key player out | key player out | 21 | 19 | 10/19 | +0.000 |
| confidence 0.65 below floor for motivation | motivation | 17 | 16 | 7/16 | +0.010 |
| confidence 0.75 below floor for turmoil | turmoil | 17 | 17 | 8/17 | +0.036 |
| confidence 0.40 below floor for key player out | key player out | 15 | 13 | 4/13 | +0.027 |
| confidence 0.70 below floor for turmoil | turmoil | 15 | 13 | 6/13 | +0.022 |
| no ruled-out language (practice status is not 'out') | suspension | 14 | 13 | 4/13 | +0.084 |
| about an earlier game, not this one | key player out | 13 | 10 | 6/10 | -0.019 |
| confidence 0.45 below floor for key player out | key player out | 12 | 12 | 7/12 | -0.018 |
| confidence 0.60 below floor for other | other | 12 | 11 | 7/11 | -0.004 |
| retired: read before this team's previous game (old news) | other | 12 | 11 | 8/11 | -0.016 |
| not an in-week coaching change | coaching change | 11 | 10 | 7/10 | -0.050 |
| QB out: the model already uses the named starter | key player out | 10 | 10 | 5/10 | +0.036 |
| retired: read before this team's previous game (old news) | travel | 10 | 5 | 3/5 | -0.010 |
| retired: read before this team's previous game (old news) | turmoil | 9 | 5 | 4/5 | -0.010 |
| confidence 0.65 below floor for travel | travel | 8 | 7 | 3/7 | +0.009 |
| confidence 0.35 below floor for key player out | key player out | 7 | 7 | 2/7 | +0.011 |
| retired: same player already counted in this game | key player out | 7 | 7 | 0/7 | +0.087 |
| QB is already the named starter | QB change | 6 | 6 | 3/6 | +0.070 |
| confidence 0.00 below floor for key player out | key player out | 6 | 6 | 0/6 | +0.000 |
| retired: schedule feed now names this QB | QB change | 6 | 5 | 3/5 | -0.023 |
| confidence 0.60 below floor for turmoil | turmoil | 5 | 5 | 3/5 | +0.002 |
| confidence 0.65 below floor for coaching change | coaching change | 5 | 4 | 3/4 | -0.017 |
| offseason / logistics story, already counted | motivation | 5 | 5 | 1/5 | +0.039 |
| offseason / logistics story, already counted | other | 5 | 4 | 1/4 | +0.004 |
| retired: read before this team's previous game (old news) | motivation | 5 | 5 | 3/5 | +0.009 |
| retired: same player already counted in this game | key player out | 5 | 5 | 5/5 | -0.088 |
| confidence 0.55 below floor for other | other | 3 | 3 | 2/3 | -0.004 |
| confidence 0.60 below floor for coaching change | coaching change | 3 | 3 | 1/3 | +0.009 |
| confidence 0.60 below floor for motivation | motivation | 3 | 3 | 2/3 | -0.014 |
| confidence 0.65 below floor for turmoil | turmoil | 3 | 2 | 2/2 | -0.013 |
| retired: read before this team's previous game (old news) | coaching change | 3 | 0 | – | – |
| retired: same player already counted in this game | other | 3 | 3 | 3/3 | -0.016 |
| confidence 0.00 below floor for QB change | QB change | 2 | 2 | 0/2 | +0.000 |
| confidence 0.30 below floor for QB change | QB change | 2 | 1 | 0/1 | +0.015 |
| confidence 0.50 below floor for other | other | 2 | 2 | 1/2 | -0.000 |
| retired: same player already counted in this game | key player out | 2 | 2 | 2/2 | -0.067 |
| retired: same player already counted in this game | other | 2 | 2 | 0/2 | +0.006 |
| retired: same player already counted in this game | key player out | 2 | 2 | 2/2 | -0.034 |
| retired: same player already counted in this game | other | 2 | 2 | 0/2 | +0.011 |
| retired: same player already counted in this game | turmoil | 2 | 2 | 2/2 | -0.007 |
| retired: same player already counted in this game | other | 2 | 2 | 0/2 | +0.005 |
| retired: same player already counted in this game | other | 2 | 0 | – | – |
| confidence 0.00 below floor for coaching change | coaching change | 1 | 1 | 0/1 | +0.000 |
| confidence 0.25 below floor for key player out | key player out | 1 | 0 | – | – |
| confidence 0.30 below floor for other | other | 1 | 1 | 0/1 | +0.002 |
| confidence 0.50 below floor for turmoil | turmoil | 1 | 1 | 0/1 | +0.003 |
| confidence 0.55 below floor for motivation | motivation | 1 | 1 | 1/1 | -0.009 |
| confidence 0.55 below floor for turmoil | turmoil | 1 | 1 | 1/1 | -0.006 |
| confidence 0.60 below floor for travel | travel | 1 | 1 | 1/1 | -0.005 |
| offseason / logistics story, already counted | coaching change | 1 | 1 | 1/1 | -0.010 |
| retired: same player already counted in this game | turmoil | 1 | 1 | 0/1 | +0.014 |
| retired: same player already counted in this game | suspension | 1 | 1 | 0/1 | +0.008 |
| retired: same player already counted in this game | key player out | 1 | 1 | 0/1 | +0.008 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.007 |
| retired: same player already counted in this game | other | 1 | 1 | 0/1 | +0.007 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.007 |
| retired: same player already counted in this game | key player out | 1 | 1 | 0/1 | +0.024 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.006 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.007 |
| retired: same player already counted in this game | other | 1 | 1 | 0/1 | +0.004 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.003 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.018 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.018 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.001 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.002 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.002 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.006 |
| retired: same player already counted in this game | QB change | 1 | 1 | 1/1 | -0.030 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.011 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.003 |
| retired: same player already counted in this game | QB change | 1 | 1 | 1/1 | -0.029 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.002 |
| retired: same player already counted in this game | motivation | 1 | 1 | 1/1 | -0.008 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.003 |
| retired: same player already counted in this game | QB change | 1 | 1 | 1/1 | -0.020 |
| retired: same player already counted in this game | QB change | 1 | 1 | 1/1 | -0.020 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.008 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.019 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.003 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.003 |
| retired: same player already counted in this game | other | 1 | 1 | 1/1 | -0.001 |
| retired: same player already counted in this game | key player out | 1 | 1 | 0/1 | +0.007 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.013 |
| retired: same player already counted in this game | key player out | 1 | 1 | 1/1 | -0.013 |
| retired: same player already counted in this game | key player out | 1 | 1 | 0/1 | +0.015 |
| retired: same player already counted in this game | suspension | 1 | 0 | – | – |
| retired: same player already counted in this game | turmoil | 1 | 0 | – | – |
Season: stats model 100–48, model + analyst 103–45; 117 games moved, 5 picks flipped (4 won).
GB @ LA Wed 8:00 PM
est. model 74% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report graded, not applied (type switched off) WR ruled out for season following knee surgery; 4-5 month recovery | LA | Puka Nacua | 0.95 | -1.9 | after the final |
| key starter out, not on the injury report graded, not applied (type switched off) Center placed on IR with hamstring injury, out 4-6 weeks | LA | Coleman Shelton | 0.95 | -1.9 | after the final |
| suspension declined: retired: same player already counted in this game Three-game suspension for personal conduct violation; serving suspension while on IR, eligible to return after Week 14 vs 49ers | LA | Puka Nacua | 0.95 | -1.9 | after the final |
| suspension declined: no ruled-out language (practice status is not 'out') Three-game suspension for personal conduct policy violation; suspension served during IR stint, eligible to return after Week 14 vs 49ers | LA | Puka Nacua | 0.95 | -1.9 | after the final |
| key starter out, not on the injury report declined: no ruled-out language (practice status is not 'out') Rams starting center out 4-6 weeks with hamstring strain | LA | Coleman Shelton | 0.95 | -1.9 | after the final |
| key starter returning declined: return of a player the model was not counting out Edge rusher Micah Parsons confirmed to return Week 12 vs Rams after reconstructive knee surgery | GB | Micah Parsons | 0.95 | -1.4 | after the final |
| key starter returning declined: return of a player the model was not counting out DT set to return after bye week for Week 12 game vs Packers | LA | Kobie Turner | 0.90 | +1.4 | after the final |
2 other items read and graded, not shown (we publish roster and schedule facts only).
CHI @ DET Thu 1:00 PM
est. model 66% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report graded, not applied (type switched off) Center placed on injured reserve; ineligible to return until Week 15 at earliest | DET | Cade Mays | 0.95 | -1.9 | after the final |
| key starter out, not on the injury report graded, not applied (type switched off) Defensive end suffered torn ACL while celebrating sack; season-ending injury | DET | Aidan Hutchinson | 0.95 | -1.9 | after the final |
| key starter returning declined: return of a player the model was not counting out Safety expected to return this week after missing entire season with knee injury | DET | Kerby Joseph | 0.90 | +1.4 | after the final |
PHI @ DAL Thu 4:30 PM
est. model 75% home · no adjustment
· forecast at kickoff: 56.0°F, wind 5.0 mph, Mostly Sunny
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report graded, not applied (type switched off) CB placed on injured reserve with ankle injury from 49ers game | DAL | DaRon Bland | 0.95 | -1.9 | after the final |
| coaching change declined: retired: read before this team's previous game (old news) Mike Pettine takes over defensive play-calling from Christian Parker; Parker remains DC in title only | DAL | 0.70 | -1.4 | after the final | |
| coaching change declined: retired: read before this team's previous game (old news) Mike Pettine takes over defensive play-calling duties from Christian Parker (remains DC in title only) | DAL | 0.70 | -1.4 | after the final | |
| coaching change declined: retired: read before this team's previous game (old news) Mike Pettine takes over defensive play-calling from Christian Parker; Parker remains DC in title only | DAL | 0.70 | -1.4 | after the final |
3 other items read and graded, not shown (we publish roster and schedule facts only).
KC @ BUF Thu 8:20 PM
est. model 69% home · no adjustment
· forecast at kickoff: 39.0°F, wind 5.0 mph, Rain Showers Likely
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report declined: about an earlier game, not this one Offensive lineman ruled out for Week 11 vs. Bills | KC | Jaylon Moore | 0.95 | +1.9 | after the final |
| key starter out, not on the injury report declined: about an earlier game, not this one Offensive lineman ruled out for Week 11 vs. Bills | KC | Trey Smith | 0.95 | +1.9 | after the final |
| key starter out, not on the injury report declined: about an earlier game, not this one Running back ruled out for Week 11 vs. Bills | KC | Kenneth Walker | 0.95 | +1.9 | after the final |
DEN @ PIT Fri 3:00 PM
est. model 46% home · no adjustment
· forecast at kickoff: 45.0°F, wind 13.0 mph, Slight Chance Light Rain
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report graded, not applied (type switched off) Left guard dislocated ankle in Week 10 loss to Bengals, out for season | PIT | Mason McCormick | 0.95 | -1.9 | after the final |
NO @ CIN Sun 1:00 PM
est. model 85% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report declined: retired: read before this team's previous game (old news) Guard ruled out for Monday night game | CIN | Dalton Risner | 0.95 | -1.9 | after the final |
| key starter out, not on the injury report declined: no ruled-out language (practice status is not 'out') Bengals rule out starting guard Dalton Risner for Monday night | CIN | Dalton Risner | 0.95 | -1.9 | after the final |
BAL @ HOU Sun 1:00 PM
est. model 77% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report declined: no ruled-out language (practice status is not 'out') DE carted off with serious leg injury on opening kickoff vs. Colts; status for Ravens game unclear | HOU | Chris Braswell | 0.75 | -1.5 | after the final |
| key starter out, not on the injury report declined: no ruled-out language (practice status is not 'out') WR injured hip vs. Colts, unable to return; status for Ravens game unclear | HOU | Kayshon Boutte | 0.65 | -1.3 | after the final |
NYG @ IND Sun 1:00 PM
est. model 43% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| starting QB change not yet in the feed declined: no ruled-out language (practice status is not 'out') Colts released Daniel Jones on Nov 20; Anthony Richardson now confirmed starter | IND | 0.85 | -5.1 | after the final |
NYJ @ MIA Sun 1:00 PM
est. model 57% home · no adjustment
No claims shown: nothing concrete in the news beyond what the model already counts.
ATL @ MIN Sun 1:00 PM
est. model 76% home · no adjustment
No claims shown: nothing concrete in the news beyond what the model already counts.
TEN @ JAX Sun 1:00 PM
est. model 91% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter out, not on the injury report graded, not applied (type switched off) Running back placed on injured reserve | TEN | Tyjae Spears | 0.95 | +1.9 | after the final |
| key starter out, not on the injury report declined: no ruled-out language (practice status is not 'out') Edge defender hospitalized after serious car accident Friday night in Tampa; status unclear for Sunday game | TEN | Jaylen Harrell | 0.75 | +1.5 | after the final |
3 other items read and graded, not shown (we publish roster and schedule facts only).
NE @ LAC Sun 4:05 PM
est. model 35% home → analyst 40% (+5.7 pts)
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| starting QB change not yet in the feed Drake Maye ruled out indefinitely with fractured fibula; backup QB needed for Week 12 vs LAC | NE | 0.95 | +5.7 | after the final | |
| key starter out, not on the injury report graded, not applied (type switched off) WR expected to miss multiple weeks with hamstring injury | NE | A.J. Brown | 0.90 | +1.8 | after the final |
WAS @ ARI Sun 4:25 PM
est. model 38% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| starting QB change not yet in the feed declined: no ruled-out language (practice status is not 'out') Gardner Minshew replaced by Josh Johnson (40-year-old, 15th team in career) due to Carson Beck back spasms; fourth starting QB in four weeks | ARI | Josh Johnson | 0.95 | -5.7 | after the final |
SEA @ SF Sun 4:25 PM
est. model 60% home · no adjustment
| Claim | Team | Player | Conf. | Points (home) | Graded |
|---|---|---|---|---|---|
| key starter returning declined: soft status (limited/questionable/cleared) WR set for return Sunday after Achilles injury (last played Week 6); listed questionable but optimism growing | SF | Mike Evans | 0.75 | +1.1 | after the final |
LV @ CLE Sun 8:20 PM
est. model 74% home · no adjustment
No claims shown: nothing concrete in the news beyond what the model already counts.
CAR @ TB Mon 8:15 PM
no prediction stored yet
No claims shown: nothing concrete in the news beyond what the model already counts.
Follow your team. It's free.
Your team on top of every page, with its next game and our pick. Want more? Plus from US$7/mo, 7 days free.