Week 8 Under the Lights: Three Home Favorites, Three Different Tests

Green Bay, Washington and Seattle are our Week 8 primetime winner picks, and none tops 59%. The market is giving us three different stories: Green Bay’s spread has returned to its original number while its total keeps falling, Washington is becoming a larger favorite, and Seattle is back on a three-point handicap.

This week, we’re putting Thursday, Sunday night and Monday night in one preview. The forecasts remain those issued Wednesday, October 28, at 2:30:49 p.m. CDT, using games through Week 7 and Tuesday’s saved lines. Thursday operations refreshed prices without changing those predictions. The full Week 8 preview covers the rest of the slate and the delayed weekly run.

Market snapshot: Thursday, October 29, at 12:54:22 p.m. CDT. Availability checked at approximately 1:27 p.m. CDT. Sunday’s and Monday’s sections are early previews from this Thursday vantage point; their final injury designations and game-day inactives remain ahead.

Green identifies the outright winner pick. It does not identify the spread pick or a qualifying bet.

The shared prediction board

These are raw, uncalibrated probabilities. Model agreement counts the five base winner models, which share information rather than supplying five independent votes. Every spread and total probability belongs to the exact original line displayed.

Last week’s primetime results were 2–1 on winners, 2–1 against the spread and 1–2 on totals: Chicago swept all three selections Thursday, Seattle’s win over Kansas City missed all three Sunday, and Dallas delivered the winner and cover Monday but missed the under. Chicago’s win also moved our published Thursday winner record from 3–3 to 4–3. Those are forecast records, not betting returns. Previous Thursday record, Week 7 recap.

Thursday: Carolina at Green Bay

October 29, 7:15 p.m. CDT, Lambeau Field. The Packers are our 58.9% winner pick, but the five-model agreement hides a wide confidence range. Elo is at 73.4%; the other four are only 54.8%–55.9%. Agreement on Green Bay winning is not agreement on a comfortable margin.

The separate point model projects Green Bay 23.17–Carolina 21.65, a 1.52-point margin and 44.82 total. That margin is smaller than the 3.5-point spread even though the separate cover blend narrowly selects Green Bay −3.5 at 51.3%. The scoring estimate and market probabilities come from different models; they need not tell the same story.

The matchup starts with protection. Green Bay is coming off a 20–17 loss to Detroit, while Carolina lost 43–13 to Tampa Bay. Both results are already included in the forecast. Can Green Bay finish drives despite its offensive-line absences, and can Carolina protect Bryce Young well enough to keep pace?

Green Bay has ruled out Sean Rhyan, Travis Glover and Zaire Franklin. Matthew Golden, Devonte Wyatt, Brenton Cox Jr. and Anthony Campbell are questionable, and the Packers added Keisean Nixon (shoulder) to that category in Thursday’s update. Aaron Banks has no game designation. Carolina is without Damien Lewis, Torricelli Simpkins III, Jaelan Phillips and Akayleb Evans, with Bobby Brown III questionable. Bryce Young and Jalen Coker have no game designation. Thursday’s complete inactive lists were not yet available at this check. Official injury report, Packers’ October 29 update.

Those absences cut in different directions. Missing blockers can hurt drive sustainability; Carolina losing Phillips may ease some pressure on Green Bay’s protection. Nixon’s status adds a question in the Packers’ secondary. The replacement players and matchups matter more than simply counting names on the report.

The price has changed, even with the spread back where it started. BetMGM went GB −3.5 → −4 → −3.5, so the saved cover probability again matches its handicap. At −110, however, the break-even rate is 52.4%, above the model’s 51.3%. Green Bay’s −200 moneyline requires 66.7%, also above the 58.9% winner forecast. Neither comparison supplies a qualifying selection in the current portfolio.

The total has fallen 48 → 47.5 → 47. The point projection remains below all three thresholds, but 53.6% Under 48 is not a forecast for Under 47. Exactly 47 points would win the original under and push the current one; exactly 48 would push the original and lose the current one.

Weather offers no obvious strong-wind explanation for that decline. The National Weather Service’s Thursday morning forecast calls for mostly clear skies and northwest wind of 5–7 mph tonight. Its low around 38°F is an overnight low, not a kickoff temperature. That outlook and the injury news are context for watching the game, not proof of what caused the market move. NWS Green Bay forecast.

Sunday night: Philadelphia at Washington

November 1, 7:20 p.m. CST. Washington’s 55.5% winner forecast has three of five models behind it; Philadelphia retains a substantial 44.5% upset probability. The Eagles enter after a 38–21 loss to Dallas, while Washington comes off its Week 7 bye.

The point projection is Washington 22.85–Philadelphia 17.20: a 5.66-point margin and 40.05 total, calculated before rounding the team scores. That margin clears the original five-point handicap but falls just short of Thursday’s six. The saved Washington −5 cover probability is 53.1%; it cannot be relabeled as the probability of covering −6. Similarly, Over 40 at 52.1% is a small lean, and the point projection sits almost exactly on the total.

Philadelphia’s skill-player availability is the main personnel question. Its Wednesday report was a walkthrough estimate: Saquon Barkley and DeVonta Smith were projected nonparticipants, while Nolan Smith was estimated limited. Washington’s Wednesday report listed Jayden Daniels as full and Stefon Diggs as a nonparticipant. Thursday updates were not yet available on those report pages when checked. These are practice labels, not final rulings for Sunday. Eagles report, Commanders report.

If Philadelphia loses important rushing or receiving options, sustaining drives and protecting the defense from short fields could become harder. Clearances could ease those concerns. Washington has its own receiving question around Diggs. We need later reports to resolve those possibilities; Thursday’s price move alone cannot tell us which expectation the market changed.

Washington is becoming more expensive to back. BetMGM’s spread progressed −5 → −5.5 → −6, with the moneyline now −275, up from Tuesday’s −235. Philadelphia moved from +195 to +220, while the total stayed at 40. Washington remains the favorite, and the model still picks Washington to win. A separate price comparison can nevertheless prefer Philadelphia’s upset chance without making the Eagles the favorite.

That distinction is the only primetime recommendation produced by Thursday’s maintained portfolio: Philadelphia moneyline +230 at FanDuel. Its unadjusted break-even probability is 30.30%, compared with the frozen model estimate of 44.48%, a recorded gap of 14.18 percentage points. The BetMGM table below shows +220 because it tracks one book consistently; FanDuel supplied the selected +230 offer. This is a model-versus-price disagreement under YELLOW monitoring, not a validated advantage or a placed bet. The quoted price was observed Thursday at 12:54 p.m. CDT and may change before Sunday.

Monday night: Chicago at Seattle

November 2, 7:15 p.m. CST, Lumen Field. Seattle is the 56.0% winner lean, with four of five models agreeing. The separate spread selection is Chicago +3 at 52.3%. Seattle can win while Chicago covers; a Seattle win by exactly three would push that original spread. Bears game guide.

The point model projects Seattle 25.93–Chicago 23.87, a 2.06-point margin and 49.80 combined points. That fits a close-game story and sits above the saved 46.5 total, where the separate over probability is 54.7%. Seattle’s 20–3 win over Kansas City and Chicago’s 30–24 win over New England last week are already part of the model’s inputs; the projection does not simply repeat either result.

Chicago has a concrete pass-rush storyline: Montez Sweat has been reinstated after a two-game suspension, according to the Bears’ October 28 report. His return, alongside Daniel Hardy’s recent expanded role, puts Seattle’s protection in focus. Can Chicago disrupt possessions often enough to keep the game inside a field goal? Reinstatement supports discussing Sweat’s return; it does not replace Monday’s final availability check. Bears report.

Seattle’s spread has come back across the three-point margin. BetMGM moved −3 (−115) → −3.5 (−105) → −3 (−112). Chicago is now +3 (−105), after Wednesday’s +3.5 (−125). Chicago bettors lost the half-point that would turn a three-point Seattle win from a push into a win, but received a less expensive price. Seattle bettors regained the possibility of a push at that same margin.

The current handicap again matches the original spread forecast, while Wednesday’s did not. The total remains 46.5, and Seattle’s moneyline is still −175. Neither Chicago–Seattle nor Carolina–Green Bay has a qualifying recommendation in this capture. Their winner and market leans belong on the prediction board, without being recast as portfolio bets.

Three snapshots, one consistent bookmaker

This table uses BetMGM throughout. Tuesday was captured October 27 at 1:10:39 p.m. CDT, Wednesday at 2:32:29 p.m. CDT, and Thursday at 12:54:22 p.m. CDT. These are saved observations, not verified opening or closing lines. Parentheses show American prices; moneyline pairs are away / home.

Injuries and weather can change expectations for scoring or a particular matchup. Wagering activity, a sportsbook’s potential payouts and adjustments by competing books can also change the number or price. Our snapshots contain no ticket counts, stake totals or bettor identities, and cannot isolate those causes. In particular, Washington’s move toward −6 is not evidence by itself that a specific Eagle will miss Sunday.

Monitoring remains YELLOW, with winner log loss of 0.70046 over 42 binary outcomes from original Weeks 5–7, just above the project’s 0.70 warning threshold. The separate recent accuracy window is 15/28 (53.6%). These probabilities remain raw and unvalidated; spread and total blends retain proxy components alongside trained models, and betting ROI is unavailable.

Our original picks stay fixed for grading. Green Bay’s protection, Philadelphia’s evolving availability and Chicago’s returning pass rush give each game its own test. As Sunday and Monday approach, later injury reports, weather and prices may change the context—but choosing a winner, covering a handicap and finding an attractive price remain different questions.

Forecast and market figures come from the project’s frozen Week 8 forecasts and October 29 combined primetime handoff. Team reports and the weather outlook are linked above. Thursday uses CDT; Sunday and Monday use CST after the November 1 clock change. All prices and availability in this article are dated snapshots, not live updates.