Stonehill at Ohio. Our picks: Stonehill +33.5 (medium). Over 52.5 (low).

Spread. The strongest explanation for Ohio laying roughly 33 is the subdivision mismatch: Stonehill generated only 161 yards against UMass, and Ohio's replacement quarterbacks both produced against South Alabama. That supports a substantial favorite even with Poulos uncertain. Our spread model supplies no number, so there is no quantitative disagreement to exploit. My modest preference is Stonehill +33.5 at -105: Ohio's defensive problems extend beyond the scoreboard, with 266 rushing yards conceded last week and additional defensive availability concerns. That makes sustaining a five-touchdown margin less comfortable than the talent gap suggests. However, the captured move from Ohio -34.5 to -33.5 at DraftKings, and -37.5 to -32.5 at FanDuel, has already reduced the underdog's cushion; its cause is unverified. Stonehill's offensive collapse against its first FBS opponent is the decisive counterweight. I would accept the best captured underdog price as a LOW lean, but the remaining advantage is too uncertain for a normal-unit bet.

Total. The market's roughly 43-10 implied score sensibly combines Ohio's talent advantage with limited Stonehill production. The strongest under case is Stonehill's 161-yard FBS debut, plus Ohio's unsettled quarterback situation. Conversely, Ohio's tackling failures, Stonehill's 380-yard passing performance at New Hampshire and productive relief work from Levi Davis give the over credible routes. Sunny kickoff conditions offer no weather-based suppression. The model's 61.07 is 8.57 above DraftKings 52.5, but its FCS translation and roster blindness make that disagreement unreliable; the gap is also smaller than its 12.86-point holdout error. Both captured books have already raised their totals two points. I cannot establish which competing effect is mispriced at 52.5 to 53.5. Under is only a conditional direction at 56.5, not a lean supporting today's captured numbers.

Spread. Our spread model produces no number here (Stonehill is an unrated FCS side) and has no proven ATS edge, so this is a market read. The strongest case for Ohio near -33 is a talent chasm: a MAC team at home against a bottom-tier NEC program that just lost 36-14 at UMass. But the number has been bought down hard, FanDuel opened -37.5 and sits -32.5, DK -34.5 to -33.5, and Ohio's own situation is messy: first-year staff, a three-way QB rotation after Poulos's injury, RB Brune and WR Perry listed questionable, slow starts in back-to-back games, and a run defense that yielded 266 rush yards to South Alabama. That argues Stonehill can hang around long enough to backdoor a 33-point number, and the market is drifting that way. It is a real lean on the points, not a bet: FCS blowout margins are high-variance, garbage-time scores swing both ways, and I have no verified injury edge. I would take Stonehill at the current best number as a lean only.

Total. Our totals model (61.1 vs a ~53 market) is 8.1 over the number, below its own 12.86-point noise threshold, so it names no side on its own and is unreliable this early against an FCS opponent. The stronger, independent signal is Ohio's actual profile: its three games have produced 70, 56 and 77 combined points (avg ~68), it scores 28.7 per game while allowing 39, and it has surrendered chunk plays and 266 rush yards in a single game. The number has also climbed at both books (DK 50.5 to 52.5, FanDuel 51.5 to 53.5), money leaning over. Against that, a lopsided FCS blowout invites a second-half running clock and backups that can strangle scoring, and SportsGrid's model actually projects the under. Weather is benign (clear, 71F, 6 mph). Real over lean at DK 52.5, not a bet given the blowout-script risk and no proven totals edge.

Spread. The market's -32.5/-33.5 number already reflects a steep discount from open (FanDuel moved -37.5 to -32.5, a 5-point shift toward Stonehill) driven by verified facts: Ohio has allowed 39.0 ppg through three games, including 266 rush yards and a blown 24-7 halftime lead in last week's 41-36 loss at South Alabama (Athens Messenger, 9/19). Starting QB Nick Poulos has missed two straight games after a helmet-to-helmet hit vs Jacksonville State, backup Matt Vezza started Week 3, and Hauser's 9/22 presser gave no clarity on who starts Saturday. RB Duncan Brune (hand) and OL starter CJ Dawson II have also been out. Our own db still lists Poulos, Brune and WR Perry questionable as of 9/24. Even discounting for Stonehill's own offensive futility (19 ppg, held to 6 and 14 points twice), getting 32.5-33.5 against a favorite that can't stop the run and has real, unresolved lineup uncertainty is value the market hasn't fully closed. I hold this as a lean, not a bet, because Saturday's QB/RB availability is unresolved.

Total. The total moved from 50.5/51.5 at open to 52.5/53.5 now, tracking the same defensive-weakness narrative as the spread. Both defenses have been porous: Ohio allows 39.0 ppg (266 rush yards last week alone), and Stonehill allows 26.3 ppg, including 36 to UMass and 42 to New Hampshire. Ohio's offense (28.7 ppg) should move the ball on Stonehill's defense regardless of who plays QB, and a leaky Ohio defense gives Stonehill's punchless offense (19 ppg) more room than usual. That argues for the over. Counter-case: Stonehill has been shut out for long stretches this season (6 points vs Bryant, 14 vs UMass), and a first-year HC managing a 30+ point lead may pull starters and shorten the game, capping scoring below the number. I lean over but hold it as a lean given real game-script uncertainty.