Mississippi State at South Carolina. Lean: Mississippi State +3.5. Lean: Under 58.5.

Spread.Edge Labs makes South Carolina -2.2, versus -3.5 available at FanDuel, so its number initially points to Mississippi State. I disagree modestly because the in-season rating appears to have overreacted to two final margins, particularly State’s excellent Minnesota win, while discarding South Carolina’s substantial talent and returning-production advantages. The market has already corrected: early lookaheads of -6.5/-7.5 collapsed after State’s start, meaning -3.5 gives the Bulldogs considerable credit. South Carolina also may regain elite EDGE Dylan Stewart and starting LT Jacarrius Peak; both practiced in full pads last week and were reportedly trending positively, although neither was officially cleared when researched. Their anticipated return is probably partly priced, but Stewart is especially important against Kamario Taylor. Taylor’s 7-to-0 TD/INT start makes laying points uncomfortable, and DraftKings reported overwhelming Mississippi State support while the line held around four. Still, at the available -3.5 rather than -4, I favor South Carolina’s deeper roster, home field and potential trench reinforcements. This is only a modest position pending the SEC report.

Total.Edge Labs projects 54, five points below the available 59, but that gap is well inside the model’s historical error. The better under argument is that both 50-point averages are inflated: South Carolina’s 45 against Towson included two punt-return touchdowns while Sellers passed for only 96 yards, and both defenses have allowed fewer than 3.0 yards per rush. The market has already lifted the total substantially from DraftKings’ early 54.5, pricing Mississippi State’s offensive breakout and both teams’ pace. The over case is legitimate, Taylor has produced 10.4 yards per attempt, and the teams average roughly 72 offensive plays, but 59 requires sustained SEC-level efficiency. Light wind makes weather mostly neutral; heat and a modest late-storm chance add uncertainty rather than a strong under edge.

Where our number looks wrong.The model has made these teams essentially equal on a neutral field after only two games. That likely overweights Mississippi State’s early margins and understates South Carolina’s large recruiting, returning-production and depth advantages. It also cannot account for Stewart and Peak potentially debuting. Conversely, its 1.3-point disagreement with the available -3.5 is too small to establish a betting edge given its 12.7-point historical margin error.

Spread.The market has already de-graded South Carolina hard: DraftKings opened this at SC -7.5 and FanDuel at -6.5, and consensus has bled all the way to -3.5/-4 by 9/16 (dossier line_history; SportsGrid/BetMGM). That move is respect for a 2-0 Mississippi State that won 38-13 at Minnesota, a real road result, not a cupcake. Our own number (SC -2.2 vs market -3.8) also leans to the dog, and it agrees with the news: South Carolina lost two receivers for the season, DJ Black (WR) is questionable, and LaNorris Sellers threw a career-low 96 yards vs Towson, so the Gamecocks' passing offense is thin and unproven. The offset is that EDGE Dylan Stewart and LT Jacarrius Peak are both trending to make their season debuts, which strengthens both SC lines. That is why this is a lean, not a hammer, take the points at +4 with Williams-Brice and an improving SC front knocking it down.

Total.Both offenses have piled up points (SC 51 ppg, MSU 50 ppg) but against Kent State, Towson and UL Monroe, plus MSU's one real game at Minnesota. This is the first legitimate defensive test for either side, and the total has been bid up from DraftKings' 54.5 open to 58.5-59. Several things cut against that climb: rain in the forecast (55% precip, dossier), South Carolina's passing game is compromised (Sellers' 96-yard game, two WRs out for the year, Black questionable), and SC's pass rush gets Dylan Stewart back. Two capable QBs and shaky 2025 defenses keep the over live, so this is a light lean under 58.5.

Where our number looks wrong.The rating still leans mostly on preseason priors (SC 11.87 vs MSU 3.78) and cannot see that MSU's offense has jumped (38 at Minnesota) or that SC's WR room is gutted and its QB struggled passing. Its flat 2.5 home field and calibration also do not know Stewart/Peak are only now debuting. The -2.2 number happens to land near the truth, but for the wrong reasons.

Spread.Our model reads this almost as a coin flip: after two games, EL's in-season rating actually has Mississippi State (10.568) fractionally ahead of South Carolina (10.342), yet the market has moved to South Carolina -3.5/-4, total 58.5 (SportsGrid market preview, 9/16). South Carolina's unbeaten start came against Kent State and Towson, two of the weakest schedules in FBS, inflating its scoring margin without testing the O-line or the questionable Jacarrius Peak (LT)/Dylan Stewart (EDGE) duo, neither of whom has played a snap in 2026 (Garnet & Cocky, 9/16). Mississippi State's 38-13 road win at Minnesota was against real competition, and sophomore QB Kamario Taylor looked like a legitimate dual-threat difference-maker (16/22-227-3 TD, 86 rush yards; WLOX/Saturday Down South, 9/13) - a development the model, built on preseason priors and box scores, has not fully absorbed. I take the road SEC underdog getting points against a home team whose sample is soft and whose front is still banged up.

Total.Both offenses have scored at will (South Carolina 51 ppg, Mississippi State 50 ppg) but against feeble competition; this is the first real defensive test either side has faced, plus each team's first SEC/road environment of the season. Our captured forecast shows a 55% precipitation chance at kickoff, though a separate general-area 10-day check found a lower ~24% rain chance for Saturday - genuinely uncertain, but any rain at Williams-Brice caps a pass-heavy shootout. I lean under on a total set for two offenses whose numbers are mostly garbage-time padding.

Where our number looks wrong.The spread model has no demonstrated ATS edge and is still mostly a preseason/box-score prior after only two games; it cannot see Kamario Taylor's emergence at quarterback for Mississippi State, nor that South Carolina's spotless record and gaudy per-game scoring came almost entirely against Kent State and Towson. It also has no way to weigh Jacarrius Peak's and Dylan Stewart's still-unresolved availability up front for South Carolina, which matters more than a flat home-field constant.