Patriots at Bears: Chicago Is the Pick, but the Price Falls Short

Chicago is our winner pick at 57.7%, but Thursday’s report produces no qualifying bet for Patriots–Bears. All five winner models favor the Bears, with estimates ranging from 53.3% to 61.9%. That is a shared, moderate lean. Our published Thursday winner record enters tonight at 3–3.

Both teams are 4–2, with kickoff at 7:15 p.m. CDT on Thursday, October 22, at Soldier Field. Prime Video carries the national broadcast. Bears game preview, Patriots viewing guide.

The forecast remains the one issued Tuesday, October 20, at 3:07 p.m. CDT. Today’s prices and final injury designations add context; they do not replace the prediction published in the full Week 7 preview.

Green identifies the predicted outright winner. The spread, total and price analysis below answer separate questions.

Our Thursday record is 3–3

Denver’s 17–14 overtime win last week evened the record. These are the teams selected in our original published previews, with results linked to the follow-ups.

That is three correct outright picks in six games. Dates follow the U.S. Central calendar, including the Melbourne game; the Wednesday, September 9 season opener is excluded. Both versions of the Week 4 forecast selected Cleveland. Later retraining does not rewrite these picks, and this winner record says nothing about spread or total performance.

Denver’s close win also illustrates the difference between picking a winner and forecasting the scoring environment: the winner selection landed, but the original Over 40.5 missed. Six games are too few to establish a special Thursday strength.

Chicago’s model advantage is modest

All five base models choose Chicago, but none assigns it even a two-in-three chance of winning.

The blend leaves 42.3% for New England. These models share information, so agreement is not five independent confirmations. Their probabilities remain raw and have not passed calibration validation.

The separate XGBoost margin and total models project:

The projected margin is 0.67 points below Chicago −3, and the projected total is 0.67 below 50.5. Yet the separately fitted market models lean Chicago −3 at 54.0% and Over 50.5 at 52.2%. The point estimates and probability models disagree about these thresholds; they should not be presented as one consistent score distribution. Calculations use the full-precision values before rounding.

A three-point Chicago win would push the original spread selection. The saved forecast does not include a separate push probability, so the 54.0% figure is not enough to construct a push-adjusted fair price.

The matchup questions behind the numbers

Can New England protect Drake Maye with two replacement guards? Alijah Vera-Tucker and Ben Brown are out. Mike Vrabel named Walter Rouse and Matt Pryor as the starting replacements, while right tackle Morgan Moses remains questionable. Protection communication and avoiding third-and-long are central to New England’s chances. Patriots personnel analysis.

The Patriots’ tactical preview points to early-down play-action as a possible way to attack Chicago and avoid its stronger third-down defense. That creates a useful viewing test: can New England gain manageable yardage early, or will its changed line face repeated long dropbacks? Chicago also has personnel limits: Montez Sweat remains suspended and Dayo Odeyingbo is on injured reserve. The protection matchup should not be reduced to New England’s absences alone. Patriots gameplan, Bears preview.

Can Chicago sustain drives without short fields? Caleb Williams completed 30 of 36 passes in his previous outing, while New England has a minus-six turnover margin through six games. Limiting giveaways would force Chicago to build longer possessions. Bears guard Jonah Jackson has no game designation, but running back Kyle Monangai is out, making the distribution of rushing work another watch point. Patriots matchup analysis, Bears personnel update.

Can New England carry its recent quarterback efficiency into primetime? Maye has 10 touchdown passes and one interception during the Patriots’ three-game winning streak, according to Chicago’s game preview. The Bears, meanwhile, begin three consecutive scheduled primetime games tonight. That gives this game significance beyond a single Thursday result, while leaving the saved probabilities unchanged. Bears game preview.

New England enters 15–8 all-time in Thursday games, a category broader than the Thursday Night Football broadcast. Chicago won their previous regular-season primetime meeting 33–14 on Monday Night Football in 2022. Patriots historical preview.

Final injury designations

The Wednesday final report supersedes the participation-only information in our weekly preview:

• New England out: Alijah Vera-Tucker, Ben Brown, Carlton Davis III, Channing Canada and Karon Prunty. Questionable: Morgan Moses, Kevin Byard III, Christian Elliss and Craig Woodson. Drake Maye and A.J. Brown have no game designation.
• Chicago out: Kyle Monangai and Jack Sanborn. Doubtful: T.J. Edwards. Questionable: Tyrique Stevenson. Jonah Jackson has no game designation.

These designations were rechecked Thursday at 12:54 p.m. CDT; the final game-day inactive lists were not yet verified. Edwards remains doubtful here, not confirmed inactive, and a questionable player is not automatically out. Official injury report.

Thursday’s prices and the missing betting edge

The BetMGM comparison covers three saved observations. All times below are CDT. Parentheses show American odds; spread prices are for Chicago. These are snapshots, not verified market openers or closing lines.

Chicago costs less to back outright than in Tuesday’s capture. New England also offers a smaller payoff than Wednesday. Because those prices include the bookmaker’s margin, the two moves alone do not establish a change in a fair win probability.

The handicap and total remain Chicago −3 and 50.5 throughout. Chicago’s spread price improved from −108 to −102 on Wednesday, while New England +3 became more expensive at −118 instead of −110; both spread prices then held steady Thursday. No cause for the movement is established by these observations.

Tonight has zero qualifying offers and zero recommendations in the saved Thursday screen across eleven captured bookmakers. At Chicago −150, the simple break-even rate is 60.0%, above the model’s 57.7%. Using full precision, the model-minus-implied difference is −2.28 percentage points. This uses the displayed price directly, not a no-vig estimate.

The project’s screen also requires an edge of at least five percentage points, a selected-side probability at or below 45% or at or above 55%, and positive quarter-Kelly sizing. Chicago −3 at 54.0% falls inside the excluded probability band; the opposite side is 46.0%. Over 50.5 at 52.2% and Under at 47.8% fail that gate too. A winner pick can therefore coexist with no recommended bet, and different lines cannot inherit these probabilities.

What goes on the record tonight

The frozen selections remain Chicago to win, Chicago −3 and Over 50.5. Friday’s follow-up will grade those exact lines, compare Chicago 26.08–New England 23.75 with the final score, and update the 3–3 Thursday winner record. Any spread push stays separate.

Monitoring remains GREEN, with 0.6802 winner log loss averaged equally across the three weekly scores from Weeks 4–6. Its separate Weeks 5–6 accuracy window is 18 of 28, or 64.3%. GREEN means the configured warning thresholds are not currently breached; it does not validate calibration. Verified recommendation ROI remains unavailable because the monitoring inputs contain no usable settled bets.

The spread blend uses three trained market models. The total blend combines three trained models with two neutral 50% components, and these market models do not have the winner ensemble’s full weekly validation coverage. Today’s football reporting does not manually adjust any of those outputs.

Tonight opens the slate in the full Week 7 preview. Sunday starts with Pittsburgh–New Orleans in France and ends with Kansas City–Seattle; Dallas–Philadelphia follows Monday. Each matchup has its own winner, spread and total questions.

Source note: The original Tuesday forecast and Thursday market report come from prediction-repository revision f6a080a. The forecast was issued October 20 at 20:07:52 UTC; Thursday’s market response was observed October 22 at 17:00:04 UTC. Official team reporting and the original published Thursday picks were checked separately. Companion editorial evidence preserves exact values, quote timestamps, source hashes and the historical record. No forecast was regenerated for this article.