Gardner-Webb at Marshall. Lean: Gardner-Webb +24.5. Lean: Under 51.5.

Spread.The strongest explanation for Marshall laying approximately 24 is its FBS depth, home field and improving rushing offense against an FCS team operating under an interim coach. Marshall rushed for 267 yards last week, and Carlos Del Rio-Wilson's productive return reduces concern about his earlier injury. Nevertheless, Gardner-Webb's passing ability attacks a specific weakness: this week's beat reporting identifies Marshall's ineffective pressure and missed tackles after nine Missouri State gains of at least 21 yards. Pennington has three 300-yard passing games; the underdog need not win the trenches consistently to stay within 24.5. Our spread model supplies no number, so this is entirely a matchup-and-price judgment. The captured FanDuel move from 26.5 to 24.5 has already removed value, but retaining more than 24 remains sufficient for a standard bet. The Liberty near-upset deserves discounting because Gardner-Webb scored defensively and Pennington threw two interceptions. Even allowing for a healthier Marshall offense, I prefer the remaining cushion.

Total.The strongest explanation for 50.5 to 51.5 is Marshall controlling the game on the ground while limiting an FCS offense; Gardner-Webb's Liberty performance included only 16 offensive points. That under case is substantial, especially because defensive touchdowns boosted Gardner-Webb's last two results. The over case rests on Marshall's improving offense, its vulnerability to explosive plays and Pennington's demonstrated passing production in favorable weather. Our 60.11 projection is 9.61 above DraftKings' 50.5, but that gap is smaller than the model's historical error and its prior-heavy treatment misses Gardner-Webb's coaching changes. I lean over 50.5, without a normal-unit bet: the matchup supports scoring, but verified possession-adjusted pace and sustainable offensive efficiency are insufficient to establish a robust edge.

Where our number looks wrong.There is no spread projection to challenge because Gardner-Webb is unrated. Marshall's rating alone cannot price this matchup, and the model omits the opposing quarterback's fit against Marshall's pressure and tackling problems. The totals model's 60.11 also deserves discounting for coaching turnover and recent defensive scores.

Spread.Our spread model produces no number here (Gardner-Webb is an unrated FCS side), so it offers nothing to lean on, and its listed weaknesses (no QB/injury awareness, unreliable vs FCS) mean I would ignore it even if it had one. The market's roughly -24 is the standard FBS-over-decent-FCS price and is well supported: Marshall fields a full FBS roster with preseason All-Sun Belt QB Carlos Del Rio-Wilson (healthy, returned at Missouri State) at home for homecoming. What nudges me toward the points is Marshall's own body of work: two-point and six-point wins over FBS Middle Tennessee and Missouri State, hardly a blowout profile, plus FanDuel drifting from -26.5 open to -24.5, toward the dog. Against that, Gardner-Webb was beaten 35-17 by FCS Austin Peay, and FBS depth usually wears an FCS team down late, so -24 is defensible. I have no price-sensitive edge at 23.5/24.5. This is a pass with a mild lean to the points, not a bet.

Total.Market is 50.5/51.5; our totals model's 60.11 is not credible here. It carries no FCS adjustment and cannot see that Gardner-Webb's road offense has managed just 13, 17 and 23 points this season. Both external projections land near 51 (Marshall ~31, GW ~20). Marshall's two competitive games each totaled exactly 54, which argues for the over, but those points came largely from FBS opponents scoring 24-26; an FCS road team should struggle to match that behind a comfortable favorite likely to run clock in the second half. Gardner-Webb's weak road scoring tilts me slightly under, but it is thin, a mild lean and not a bet, with no weather help (clear, ~72F, ~6 mph wind).

Where our number looks wrong.The spread model returns no number because the FCS side is unrated, so it is silent rather than wrong; I would not trust it against an FCS opponent regardless. It adds nothing to the read.

Spread.The market's current number (Marshall -23.5 DK / -24.5 FD) already moved 2 points off FanDuel's open of -26.5 while DraftKings sat static at -23.5, so some money has already leaned Gardner-Webb, but the books still disagree by a full point, no consensus repricing. The strongest explanation for a ~24-point line is simple: Marshall is FBS/Sun Belt at home for Homecoming against a 1-3 FCS program with a turnover-prone QB (Cole Pennington, 1 TD/4 INT through 3 games). But Marshall's own results undercut a blowout price, its two 2026 wins were 28-26 over Middle Tennessee and 30-24 at Missouri State, margins of 2 and 6, sandwiched around a 0-45 loss at Penn State. A team that hasn't cleared a 6-point margin against comparable-or-weaker competition is being priced to win by 24-plus. Gardner-Webb played Liberty (FBS) to a 1-point game and Wofford to a 4-point game, then hung 44 on NC Central. That gap between Marshall's actual scoring margins and the number is real incremental evidence, not a generic dog narrative, but the FBS/FCS talent gap and an unverified GW health picture keep this a lean, not a bet.

Total.No real read either way. Both books' totals are unchanged from open (DraftKings 50.5, FanDuel 51.5), unlike the spread which has moved, the market sees nothing new worth pricing on points. Marshall's offense has trended up the last two weeks (28, then 30 points) after being shut out 0-45 by Penn State, and Gardner-Webb can move the ball (Cole Pennington ~290 yards/game passing) despite 4 interceptions in 3 games, which could cut either way, extending Marshall's short-field scoring or simply killing Gardner-Webb's own drives. A likely running-clock second half in a ~24-point mismatch argues under; Marshall's defense allowing 31.7 ppg this season and Gardner-Webb's functional passing game argue over. Neither case clears the bar for incremental, sourced evidence, so this is a pass with no view rather than a manufactured lean in either direction.