New Orleans Saints at Baltimore Ravens. Lean: Baltimore Ravens -8.5. Lean: Under 47.

Spread.Take Saints +8.5 at FanDuel (-115). Edge Labs makes Baltimore -9.6, but its raw margin is only 7.7; the unproven calibration creates the apparent favorite value. My judgment is closer to Baltimore -7.5, with substantial injury uncertainty. Monday's receiver news reduces Baltimore's explosive-play ceiling without establishing that Flowers will miss Sunday. Shough's comeback offers a plausible route to a late Saints cover, although his turnovers make an outright offensive upgrade premature. Baltimore deserves favoritism: Minter and Doyle's debut produced 506 yards, and the revamped rushing attack looked effective. Those positives are public, however, and the captured DraftKings spread already moved from -7.5 to -8.5. I cannot establish that the market missed the injuries; I think the remaining receiver uncertainty favors taking the points. Kamara's reported potential return adds modest upside, not a guaranteed adjustment. Conflicting betting-split snapshots provide no dependable sharp-money signal. This is a small price-sensitive lean, not a strong position.

Total.Take under 47.5 at FanDuel (-115). Applying the dossier's specified 1.2-point bias correction yields 46.57, only 0.93 below this price. The stronger under argument is Baltimore's receiver uncertainty and misleading opener scorelines: Saints-Lions was 24-24 after regulation, with two Detroit touchdown drives following short-field turnovers. Baltimore allowed just 251 yards despite conceding 23 points. The over case remains credible: Doyle's offense generated 506 yards, Shough demonstrated explosive passing, and healthy Flowers participation would weaken this under. A Henry-led clock-control script is an inference; I verified no neutral-situation pace metric. FanDuel's rise from 46.5 contrasts with BetMGM's captured decline, so there is no uniform market signal. Forecast winds are modest; uncertain showers warrant no material weather deduction. Prefer the half-point above 47, with restrained conviction.

Where our number looks wrong.The spread calibration stretches Baltimore's raw 7.7-point advantage to 9.6 despite no demonstrated ATS benefit. Final-margin learning misses receiver availability and can misread turnover-driven results. Baltimore's new coaching staff and personnel also require more than a binary coaching flag. For totals, subtracting the specified 1.2-point historical bias changes 47.77 to 46.57; the unchanged QB flag misses Flowers and Lane entirely. Neither adjusted number establishes an independently validated betting edge.

Spread.Our rating (Ravens -9.6) sits a point above the -8.5 market, but it has no proven ATS edge and a week in is still mostly the preseason prior. The case for Baltimore stands on its own: the Ravens blew out Indianapolis 41-23 (Fox 9/13) with Henry (144, 3 TD) and Lamar (324 pass), while New Orleans lost 31-30 in OT at Detroit, and most of Shough's career-high 410 yards came while trailing 21-0 in a soft comeback script (Saints.com 9/13). Baltimore is far more talented, healthy at QB, at home, and faces a Saints front missing Cam Jordan (hamstring) and Godchaux. The hesitation is that 8.5 is a live backdoor number: big Ravens leads invite garbage points and Shough can move the ball. I lay the points as a modest lean, not a hammer.

Total.The model (47.77) barely separates from the 47 market and is admittedly worse than the market on holdout, so this is close and I keep the conviction low. I lean under. Baltimore's formula as a big favorite is Derrick Henry and clock control; with a lead they grind and bleed possessions, which caps scoring. Weather adds a mild drag: rain showers are possible before 2 p.m. for the 1 p.m. ET kickoff before skies clear (AccuWeather/NWS, warm ~87F, not windy). New Orleans' 30 points came in a specific OT comeback script that will not repeat here, and they face a solid Ravens defense. I'd rather have the under 47, but only as a light lean given how unreliable our totals model is.

Where our number looks wrong.The v3 rating's -9.6 is inflated by the calibration map stretching a large raw margin and is still mostly the preseason prior one week into the season. It also cannot see that a big home favorite laying 8.5+ carries genuine backdoor risk against an offense that just posted 410 passing yards.

Spread.Both teams showed defensive cracks in week 1, Baltimore allowed 23 to a lesser Colts team while its offense (Henry 144/3 TD, Lamar 324 yards) did the heavy lifting, and New Orleans spotted Detroit a 21-0 hole before a garbage-time rally forced OT, a script that flatters Shough's box score more than the Saints defense. New Orleans gets Kamara back but Cameron Jordan (hamstring, no preseason reps) is a real question for a pass rush that already couldn't touch Goff. I'll lay the -8.5 with the market rather than fade a public number the sharp side is already backing.

Total.Both offenses looked live in week 1 and both defenses looked beatable: Baltimore hung 41 on Indianapolis, and New Orleans put up 30 (with a missed 2-point try) on a good Detroit team despite Shough's three giveaways. That argues for a competitive-scoring floor even if Baltimore controls the game. Our totals model (47.77) sits a bit above the current market (46.5-47), and the model's own bias note says it should be nudged down about 1.2 points against the market historically, which still lands it essentially on top of the current number rather than clearly under it. The total has bounced between 46.5 and 48.5 all week without settling decisively low, and a full-strength Kamara plus Olave coming off a career game gives New Orleans more paths to points than a pure blowout script would suggest, even in a likely loss.

Where our number looks wrong.The model's own documentation says it has zero proven ATS edge historically and is still running mostly on the preseason prior with only one game of 2026 data folded in, it doesn't know that New Orleans' defense was gashed for 21 first-half points by Detroit or that Baltimore's offense looked explosive in week 1 beyond what a single-game rating update captures. The 1.1-point gap it flags as a pass is coincidentally in the right direction, but that's the market/injury picture doing the work, not the rating.