Green Bay Packers at Minnesota Vikings. Lean: Minnesota Vikings -1.5. Lean: Under 46.5.

Spread.Minnesota -1.5 is my pick. Edge Labs makes Minnesota -0.8, only 0.7 points short of this market; that disagreement is negligible for an unproven spread model. More consequential is Green Bay's personnel: Josh Jacobs was placed on the exempt list after the model freeze, while Micah Parsons remains in knee rehabilitation. Those losses weaken both offensive reliability and the pressure available against Murray. Minnesota's current depth chart confirms Murray with Jefferson, Addison and Hockenson, although a depth chart does not establish medical clearance. The counterargument is Love's continuity against a quarterback making his Minnesota debut, plus Gannon's familiarity with Murray from Arizona. Much of the personnel news is already public: FanDuel reports movement from Minnesota +1.5 to -1.5, beginning earlier than the dossier's capture window. I am following a defensible adjustment, not claiming undiscovered information. The modest spread remains playable, but incomplete Minnesota practice information prevents stronger conviction.

Total.Under 46.5 is a small lean. Applying the dossier's stated 1.2-point historical overprediction adjustment takes 42.97 to 41.77, but I would shrink that apparent advantage substantially toward the market. The strongest over case is serious: indoor conditions, Murray's receiving options, Parsons' absence and Green Bay's defensive transition could produce efficient passing. FanDuel's rise from 45.5 to 46.5 already reflects a more optimistic scoring outlook. Against that, Jacobs' unavailable status reduces Green Bay's established rushing and finishing options, while Murray's first regular-season outing in this offense introduces execution uncertainty. No verified pace data justify assuming a fast game. DraftKings reports disproportionate under handle, but that is one book's sample, not proof of informed action. The extra point above the opener makes under narrowly preferable.

Where our number looks wrong.The frozen rating overstates Green Bay's available offensive personnel by missing Jacobs' subsequent exempt-list placement. Injury blindness also misses Parsons' absence, while unchanged head-coach flags conceal a defensive coordinator transition. The totals model may understate Minnesota's scoring ceiling by anchoring to its 2025 offense despite Murray's arrival. These are directional concerns, not validated point adjustments.

Spread.Our v3 rating actually has Green Bay as the better team (2.05 vs 1.53) and still lands on Minnesota -0.8 once home field is applied; the market is -1.4. That is agreement, and the model has no ATS edge (47.6% held out), so any case has to come from news it cannot see. It is frozen at Aug 19 and blind to injuries, and the injury ledger here is one-sided. Green Bay opens without Micah Parsons (ACL, reserve/PUP, out at least four games) and without Josh Jacobs (commissioner exempt list, court date moved up to Sept 10), and on Wednesday listed guard Aaron Banks and tackle Zach Bako-Bewele questionable with knees. Minnesota's report is depth pieces and one illness. A healthy home division favorite in a dome, with Kyler Murray upgrading an offense that posted -0.12 EPA/play in 2025, should be worth more than 1.5. The counter is real and is why I am not higher: Green Bay has quarterback and scheme continuity in Week 1, Minnesota is installing around a brand-new starter, and ESPN's FPI still makes Green Bay a 56.6% favorite.

Total.Fixed roof, so weather is out of it. Our totals model says 42.97 against a 46.2 consensus, and its documented ~1.2-point high bias would push the fair number lower still, but it shrinks toward the league mean, so it will always lean under on a high number, and its holdout MAE (10.46) is worse than the market's (10.19). That is a nudge, not a thesis. The football case: Minnesota had the better defense in 2025 (-0.087 EPA/play allowed) and is healthy, while Green Bay is without its lead back, has two offensive linemen questionable, and is starting a committee backfield, that is how drives stall into punts. Murray is making his first start in a new system. Against that: no Parsons means Minnesota should move the ball, and the total has been bought up a full point since Aug 30. A small under, not a big one.

Where our number looks wrong.Two places. First, the v3 rating still has Green Bay ahead of Minnesota by half a point, which is a preseason number frozen 8/19 that cannot see that Green Bay lost Micah Parsons to an ACL and reserve/PUP, lost Josh Jacobs to the exempt list, and is patching two offensive line spots, the rating is pricing a roster the Packers will not field on Sunday. Second, the dossier's own gap reading calls this a 'pass' because |gap| is 0.6, but that framing is only meaningful if the gap carried signal, and the model's own held-out record (47.6% ATS, 10.0 margin error against a 9.5 closing spread) says it never did. The honest statement is not 'no edge because the gap is small' but 'no edge from this model at any gap', the injuries table holding zero NFL rows is the binding constraint here, not the size of the disagreement.

Spread.What moves me off the model's 'pass' is news the frozen (8/19) rating can't see: Micah Parsons opened the season on Reserve/PUP with a knee injury (ESPN injury report, 9/1; Packers.com profile on his rehab, 9/8), which removes Green Bay's best pass rusher for at least the first four weeks. That is a real, dated negative for the away team that argues in the market's direction, not against it, it undercuts the model's slight lean toward Packers rather than confirming a fade. Green Bay's RB1 Josh Jacobs also carries a Reserve/Commissioner-Exempt designation with a 9/10 court date per ESPN's injury tool (single-sourced, could not corroborate elsewhere; treat cautiously). Offsetting factors, Kyler Murray's first start as a Viking and some O-line questions (C Jurgens IR-R, OT Tiernan questionable), keep this a lean, not a confident play. I take the freshest number, Minnesota -1.5.

Total.Fixed-roof dome, so weather is a non-factor. Our totals model reads well under the market (42.97 vs 46.2; even adjusted for its known ~1.2-point upward bias, roughly 44.2 vs 46-46.5), though it's blind to injuries and shrinks toward the mean, so only partial evidence. Fresh injury news cuts both ways: Parsons' absence eases pressure on Vikings QB Kyler Murray (mild over catalyst), while Jacobs' uncertain/possible absence removes Green Bay's lead back (under catalyst), these largely offset. With no weather boost, a division opener, and Murray's first start with this team (typical early-week timing friction), I don't see a catalyst strong enough to back the number the market bid up to 46-46.5. Under, at modest confidence given the offsetting injury news.

Where our number looks wrong.The model itself flags the real problem: it is frozen from 8/19 and carries zero injury rows, so it cannot see that Green Bay's best defender (Parsons) and possibly its starting RB (Jacobs) are out. Both are negatives for the side (Green Bay) the model likes relatively more than the market does, meaning the model's small lean toward Green Bay is arguably stale rather than sharp. Given the model's own 47.6% held-out ATS record, I would not trust its 'pass' as evidence of an efficient line here, I'd trust the fresher injury news over the frozen rating.