Panthers vs. Eagles

Logged 15 Oct 2026, 19:02 UTC · Brier 0.1620

• Jalen Hurts is ruled out for the Eagles, which moves this game from a normal home-favorite setup into the backup-QB class, the single biggest swing factor in any NFL game and worth roughly 10-15 points of win probability off Philadelphia.
• The injury hit is broader than quarterback: Saquon Barkley and DeVonta Smith both missed practice with hamstring issues and center Drew Kendall also sat, meaning the Eagles' offensive downgrade around the backup QB is compounding rather than isolated.
• Carolina enters relatively stable at 2-2 off a win over Detroit with only one notable injury (right tackle Monroe Freeling), so the matchup compares a functioning team against a decimated one; a 2-2 vs 2-3 record gap normally means little, but here it is overlaid on the QB absence.
• Home field at Lincoln Financial Field still counts for something (roughly 2-2.5 points of spread historically), and the Eagles' 9-5 all-time edge in the series is stale pre-2025 roster data and carries little weight; these keep me from pushing the Panthers estimate higher.
• I treat the injury reports as genuine signal (an official rule-out plus DNP designations) rather than narrative; the series-history fact is narrative and is given minimal weight.

What would change it: News before kickoff that the Panthers' own starting quarterback or multiple offensive starters are also ruled out, or that the market context changes (e.g., Hurts' rule-out being reversed), would push this back toward a coin flip.

-3.7 probability points from the market at generation · med confidence · Polymarket

Base rate: Historically NFL home teams win roughly 57% of games, but teams forced to start a backup quarterback win only about 35-40% of the time; the intersection (home team minus its QB1) has won roughly 40-45%, a rate I state from knowledge of the class rather than any retrieved page, with a cited source establishing that this game belongs to that class.

Chosen by Crossbook's published selection rule (auto-v1), not by hand.

Formed blind at 56.0%, without sight of the price; published at 59.8%, the market’s 63.5% moved 50% of the way toward it by fixed rule. None of the cited facts could be shown to be one the market had not already priced. That check is recorded, not enforced.