Marshall at Missouri State. Lean: Missouri State +4.5. Lean: Under 52.5.

Spread.My fair line is Marshall -1.5 to -2.5, conditional on Carlos Del Rio-Wilson’s availability, so Missouri State +4.5 offers roughly two points of value, but far less than the model’s 6.8-point gap. Marshall produced only 310 yards and went 4-for-12 on third down against Middle Tennessee before Del Rio-Wilson’s shoulder/collarbone injury forced three quarterbacks into action. Missouri State’s 584-yard performance against FCS Lindenwood is easy to discount, but Skyler Locklear’s accuracy and a defense that has limited explosive plays make the Bears competitive at home. The counterargument is substantial: Marshall has the stronger returning core, while Missouri State may be missing its starting center, kicker and a receiver, with two linebackers already lost for the season. Movement toward Marshall likely reflects that imbalance and an expectation Del Rio-Wilson plays. Nonetheless, his status remained unverified Wednesday, and +4.5 covers the most likely one-score outcomes. Public-split feeds conflict, so they add no conviction.

Total.The over deserves respect: Edge Labs projects 54.38, Missouri State generated 584 yards against Lindenwood, and the hot, dry, light-wind forecast presents no conventional weather obstacle. Missouri State’s vertical concepts and multiple mobile quarterbacks also create explosive-play and broken-play upside. The under case is stronger at 52.5. Marshall managed only 310 yards against Middle Tennessee, with one touchdown effectively supplied by a 54-yard interception return to the 2. Del Rio-Wilson’s unverified shoulder status threatens both efficiency and Marshall’s quarterback-run game. Missouri State could also be without its starting center and kicker, reducing drive consistency and field-goal reliability. The totals model is largely informed by last season and cannot see Missouri State’s new quarterback/coaching structure or current injuries. I project approximately 50 points; under 52.5 is a modest position, not a full-size bet.

Where our number looks wrong.Missouri State -2.4 is not defensible as a true fair line. The two-game margin refit appears to overreward a 48-14 rout of FCS Lindenwood and insufficiently opponent-adjust the teams’ blowout losses at Texas A&M and Penn State. It also misses Missouri State’s offensive-line and defensive attrition and Marshall’s quarterback injury. I make Marshall a small favorite, not Missouri State a favorite.

Spread.Market has Marshall as a 4.5-point road favorite (home +4.5), and it has moved TOWARD Marshall from a 3.5-4 open, even though Marshall's starting QB Carlos Del Rio-Wilson (shoulder/collarbone) left the MTSU win at halftime and is uncertain for Saturday (RotoWire 9/13-14). That is the key disagreement: our number actually leans Missouri State (model has these teams near-even, -5.94 vs -5.87) and the biggest injury on the board hits the favorite, both of which argue for the home dog. Marshall would lean on Zion Turner/Khalil Wilkins if CDRW can't go, and its offense already managed only 14 ppg with two shutout-heavy results. Missouri State is 1-1, hammered FCS Lindenwood 48-14 at home, and gets a full home crowd. I'll take the points. Confidence is held to a lean because the line drifting toward Marshall despite the QB news is a real sharp-side caution, and both teams are volatile.

Total.The live news cuts the other way: Marshall's offense is already the weaker unit (14 ppg) and faces starting either a hobbled Del Rio-Wilson or a backup QB, which typically shortens and stalls drives. Both teams have a shutout on their ledger this year, so neither offense is proven against real competition. Kickoff is hot (low-90s) with only light 5-11 mph winds and a slight evening storm chance (NWS Springfield), heat plus a backup-QB game nudges toward a grind rather than a track meet. Missouri State's soft red-zone defense is the one over argument, but the QB uncertainty outweighs it. Lean under, modestly.

Where our number looks wrong.The rating sees no injuries and rates Marshall (-5.87) essentially level with Missouri State (-5.94) despite Marshall's far higher returning production (0.988) and stronger prior SP+, that looks like the in-season refit over-discounting Marshall on one blowout loss. The saving grace is that the unseen Del Rio-Wilson injury pushes the true number back toward our home lean, so the conclusion survives even if the rating's path to it is shaky. The model also can't explain why the market moved toward the injured favorite.

Spread.Our model's home lean (Missouri State -2.4 vs a 4.4-point market dog, gap 6.8) is context only, it has no proven ATS edge in any gap bucket per its own disclosed backtest, and this early it is still mostly preseason prior plus two results (a Missouri State blowout of FCS-caliber Lindenwood, a Marshall struggle to beat Middle Tennessee). The real, verifiable edge is news the model can't see: Marshall starting QB Carlos Del Rio-Wilson left the second half of the 9/13 win over MTSU with a shoulder/collarbone injury and was still unconfirmed for this game as of the last update, needing a full practice week to play (RotoWire, 9/13). If backup Zion Turner starts or plays extended snaps on the road at a place Marshall lost by 1 last season, that's a live downgrade to Marshall's offense that the market has only partially priced, the line ticked UP toward Marshall (roughly 3.5-4 to 4.5) after the injury broke, suggesting the market is leaning on Marshall's talent gap rather than resolving the QB question. Missouri State also produced 584 total yards/264 rushing against Lindenwood, a real (if soft-competition) signal its offense is functional under a new staff. I take the home dog for the points, moderate conviction given the QB outcome could break either way.

Total.Two defenses that rank in the bottom third of FBS (Missouri State 121st in points allowed at 32.0/g, Marshall 119th in yards allowed at 35.5 PPG) meet at a total the market has only nudged from the low-52s to 52.5 all week, not a market that sees a blowout-shootout or a defensive struggle coming, just a mid-tier total. Missouri State's offense looked functional (584 yards) last week even discounting the opponent, and if Marshall is forced to lean on backup QB Zion Turner or a limited Del Rio-Wilson, Missouri State's defense (which just allowed 50 to Texas A&M) should still be able to move the ball against a Marshall defense that has already allowed 35.5 PPG. 93F heat at a 6:30pm kickoff is a mild fatigue factor late but not severe enough to flip this to a lean-under call.

Where our number looks wrong.The model's home lean looks directionally right by coincidence, not by insight, it can't see the Marshall QB injury (its single biggest input this week) and its own backtest shows zero ATS edge at any gap size, so the 'Lean' label shouldn't be read as a real edge. The model also treats Missouri State's 48-14 win as a straightforward in-season signal without adjusting for Lindenwood being a well-below-FBS-average opponent, which likely overstates how much Missouri State's offense has actually improved.