New Mexico at Oklahoma. Lean: New Mexico +22.5. Lean: Under 46.5.

Spread.EL's Oklahoma -18.3 implies 5.2 points of cushion, but I would discount that substantially. New Mexico's six-takeaway demolition of Mercyhurst exaggerates its sustainable advantage, and replacing Jack Layne with a quarterback rotation matters against Oklahoma's defense. Those limitations support much of the market's premium over EL. The counterweight is Oklahoma's offensive uncertainty after scoring ten at Michigan, Michael Fasusi's unresolved availability, and Kip Lewis's knee injury. Lewis potentially missing time improves the Lobos' chances of sustaining enough offense to stay inside this number. Layne's August injury is established information, so it should not receive another full adjustment now. The dossier's opposing spread moves at DraftKings and FanDuel show convergence rather than convincing directional sharp action. I favor the cushion above three touchdowns, with modest conviction because New Mexico's road offense remains largely untested.

Total.EL's 47.54 is only 0.96 lower, providing little independent support. The under case rests on Oklahoma's defensive strength, New Mexico's quarterback transition, and avoiding extrapolation from turnover-assisted scoring against Mercyhurst. Oklahoma's ten-point Michigan performance also raises offensive questions. The over deserves respect: Oklahoma faces a substantially weaker defense, Lewis's injury can help New Mexico, and explosive plays or short fields could quickly defeat the under. I cannot verify an extreme-weather or slower-tempo advantage. The dossier documents substantial early under movement, meaning much of that case was already priced. Public boards lack dependable quote timestamps, so their higher total does not establish a confirmed market reversal. This is a small lean.

Where our number looks wrong.EL likely overstates New Mexico's improvement after two favorable home matchups, particularly a six-takeaway FCS blowout, while missing Layne's season-ending absence and the replacement quarterbacks' uncertainty. Oklahoma -18.3 therefore looks too favorable to the Lobos, although Lewis's injury pushes back in their favor. The injury feed also misidentifies Layne and incorrectly names New Mexico State in Lewis's note. Separately, the totals documentation says weather is absent, but the supplied calculation explicitly subtracts 2.597 points for wind; that adjustment must not be counted twice.

Spread.Our number (OU -18.3) sits 4.6 short of the market's -22.9, and the live news leans the same way. Oklahoma's problem is the side of the ball that has to cover: 345 yards per game, 105th in FBS, 10 points and 3.4 yards per carry at Michigan with the backs given only 13 totes (SI, 9/13). To bury a 9-4 G5 team by 23 the Sooners have to score in the mid-30s, and they have done that once, against UTEP. The defense is the real thing (8.5 points and 230.5 yards allowed, 3rd in pass defense per AP/KRMG 9/13), so the likely script is a controlled 17-to-20-point win with the clock running, not a 45-10 laugher. Add the schedule: OU travels to No. 2 Georgia the following week, and SI's own line was that left tackle Michael Fasusi is not needed here but is vital at Georgia (9/12) - the tell of a team managing bodies. New Mexico is 2-0 with a +8 turnover margin, the national lead, and 246 rushing yards per game. Take the points.

Total.This is the cleaner side of the game. That is a two-to-three point move in a day with no injury news attached, and I will not fade it. The underlying numbers agree: Oklahoma is 105th in total offense and just put up 10 points against a real defense, while New Mexico is playing without its starting quarterback and rotating two passers who have thrown for 198 combined yards into the nation's 3rd-ranked pass defense. The blowout branch that produces an over also produces a running clock and backups. NWS has Norman mostly sunny, high near 93 with a 20 percent chance of showers, and our capture has 13 mph wind - a mild further drag, not a driver.

Where our number looks wrong.The in-season refit is being moved by the wrong results. It cut Oklahoma from a 18.1 preseason rating to 14.5 for losing 17-10 at Michigan - a respectable road result against a top-10 opponent - and raised New Mexico from -4.2 to +1.3 for beating Central Michigan and an FCS team by a combined 101-14. Both moves are schedule artifacts, so our -18.3 is flattering the dog for reasons I do not believe, and the honest gap is smaller than 4.6. That is why this is a 57 and not a 65: I like the side on the offensive-profile and lookahead arguments, not on our number. The totals model is also blind to the fact that New Mexico's 54 points per game came against a MAC team and an FCS team.

Spread.But the model cannot see personnel: New Mexico's starting QB Jack Layne is out for the season after a second elbow surgery, and the Lobos have been rotating three inexperienced arms (Moga, Fa'avae, Laubstein) behind an offense that built its 54 ppg average almost entirely against Central Michigan and FCS Mercyhurst, not a Power 4 defense. Oklahoma's defense has allowed just 8.5 ppg through two games and is coming off a competitive road loss to Michigan (10-17), a game that should sharpen rather than deflate them at home against a lesser opponent. That combination, a shaky, unproven passing game on the road against an elite, healthy defense, argues the market has this priced about right or slightly light, not that -22.5 is unclearable. Confidence is capped in the high-50s because the number is already very large and backdoor-cover risk in a probable blowout is real if Oklahoma plays reserves late.

Total.The case for the under is New Mexico's offense: it has not faced a defense remotely like Oklahoma's (8.5 ppg allowed) and is breaking in a rotating, unproven QB group after Jack Layne's season-ending injury, which should make sustaining long scoring drives against a Power 4 front difficult. The case against a full fade of the over is that Oklahoma's offense leads the FBS in red-zone conversion and could pile up points quickly against a Group of Five defense that has only been tested by weak competition, and New Mexico ranks 1st in FBS turnover margin (+8), which could shorten fields for both sides. On balance we lean under on the strength of the QB downgrade suppressing New Mexico's scoring more than it helps Oklahoma's, but with only moderate conviction given the model itself is flat.

Where our number looks wrong.The model is built on final-margin history and a flat home-field constant; it has no way to know New Mexico lost its starting QB for the year, which is exactly the kind of information that should widen a line beyond what a pure ratings gap suggests. Its 'pass' reading here undersells the roster mismatch.