Tennessee State vs. Georgia: Final Score & Recap

What we projected, and where it missed

Our margin model made Georgia the pick and gave it a 68 percent win probability at a projected margin of about nine points. The side was right. The size was off by 51 points, and the reason is worth stating plainly rather than dressing up after the fact. The model earns its keep on games where both teams have a measurable record against shared competition. Week 1 against an FCS opponent is the opposite of that: there is almost nothing on Tennessee State's side of the ledger to price, so the model retreats to a cautious edge and a coin-flip-plus win probability. Nine points against a team that would finish with 10 rushing yards is not a read, it is a shrug with a number attached.

The market did far better and still missed badly. Georgia closed at 46.5 with a total of 56.5, and the final margin was 60; Georgia's 63 points alone cleared the total by itself. The pregame file records our call and our confidence but not the ingredient list behind it, so there are no decomposition drivers to name here. What the model had right in direction was the mechanism it usually leans on, and the box confirmed it at scale: Georgia ran 0.746 expected points added per play to Tennessee State's minus 0.098, and scored on 75 percent of its drives to Tennessee State's 8.3 percent. Directionally correct, catastrophically undersized. Treat the Georgia number as a data point about the model's Week 1 blind spot, not as a forecast that landed.

Five possessions, five touchdowns

Nate Frazier settled it before the crowd finished sitting down: 48 yards for a touchdown on the game's opening drive, then 14 more later in the first quarter. Stockton added three touchdown passes, 17 yards to Elyiss Williams, 11 to Talyn Taylor, and a 14-yarder early in the second quarter. Five possessions, five touchdowns, 35-0. Stockton was gone by the middle of the second quarter at 15-of-16 for 202 yards, and Frazier's 72 yards on four carries with three scores amounted to a full evening's work compressed into about eleven minutes of football.

I'm excited for Nate. He's worked really hard. He just hit right where some big holes opened up and he did a nice job being explosive through them.

The team that runs on first down decided to throw

Nothing in the Georgia box score says this outright: against the most overmatched opponent on its schedule, Georgia stopped being a run-first early-down team. Of its 60 offensive snaps, 35 came on first down, and on those 35 it threw 21 times, 60 percent. In the first half the rate was 73.7 percent. It went explosive, 15 yards or more, on 37.1 percent of first downs and averaged 0.724 EPA per play there. Georgia in 2025 ran on 60.3 percent of its first downs across 930 snaps and threw on 39.5. That is a 20.5-point swing away from the team's own habit.

The reversal tracks the depth chart precisely. With Stockton or Ryan Puglisi on the field, 47 of the 60 snaps, Georgia ran 38.3 percent of the time. With Ryan Montgomery or Hezekiah Millender, 12 snaps, it ran 75 percent, and the fourth quarter alone was 81.8 percent run. First-down pass rate fell from 73.7 percent in the first half to 43.8 in the second. Georgia threw while the starters were in and ran once the game belonged to the third and fourth quarterbacks.

Blowout favorites as a class pass less on first down than their own norm, not more: 1.1 points less at spreads of 40-plus, 1.7 points less at 24 to 40, across eleven seasons. Georgia's plus 20.5 sits at the 98.3rd percentile of 1,615 blowout-favorite team-games, and only 28 of the 1,615 cleared a plus-20 flip with a first-down pass rate of 55 percent or better. So it is off-grain everywhere except in Georgia's own file. In games favored by 35 or more from 2015 to 2025, this program threw on 42.4 percent of 695 first downs against 37.7 percent everywhere else, and 2023 already carried a plus-16.7 blowout lean. This was the biggest version of something Georgia does.

On the 11-season record the flip bought Georgia nothing measurable, and it cost nothing either. Restrict the comparison to neutral-script first downs, meaning first half with the offense not trailing, which is exactly the window Georgia's flip lived in, and 183 team-games from 89 programs survive since 2015. Those teams won 98.9 percent of the time by an average of 35.2 and covered 50.8 percent, indistinguishable from all 1,415 blowout favorites in the same span. Their next game went 61.7 percent and 48.3 against the spread versus a 64.2 and 49.4 baseline. And it does not stick: across 77 analogs with enough games on either side, the pass lean drifted back 1.8 points afterward, the identical drift every other blowout favorite shows. Do not read Georgia as a pass-first offense now; teams that throw like this in a body-bag game are the same team the following week.

The closest tier, nine games, is the interesting one and the thinnest. Hold the same flip at a spread of 40 or more and nine team-games survive since 2017. They went 9-0 by an average of 43.1 points and 9-0 again the next week, but covered only two of nine against a 43.9 percent baseline, which is the shape of a team that stops trying once the reps are banked. Georgia's own second half is that shape exactly, and Georgia owns three of those nine names: UT Martin and UAB in 2023, Tennessee Tech in 2024. Nine games is too thin to treat as destiny, and no play-call log proves intent. What the flip plausibly bought is the thing the box cannot price: Stockton throwing on first down against live coverage with nothing at stake, at a cost of zero, since first-down passes averaged 11.76 yards and first-down runs 11.79.

Nine touchdowns, one goal-line snap

Georgia snapped the ball inside the Tennessee State 10 exactly once all night, on Frazier's 3-yard touchdown run with 10:56 left in the third quarter, and once inside the 5. Last season the Bulldogs averaged 6.85 snaps inside the 10 per game. This was not an offense that never got close, either: red-zone volume was ordinary, 11 snaps inside the 20 against a 2025 average of 11.69 and a 2024 average of 11.62. The difference is that seven of the 11 were touchdowns. That is a 63.6 percent touchdown rate per red-zone snap, against 27.0 percent last season and 27.8 the season before.

The nine scoring plays were snapped from an average of 20.9 yards out, and every one came on first or second down. Georgia never faced a third down inside the red zone, and faced only six third downs in the entire game.

Scan all 18,941 FBS team-games from 2015 through 2025 and this structural signature, 10 or more red-zone snaps with at most one inside the 10, turns up 142 times, and it normally means the exact opposite of what it meant for Georgia. Those 142 offenses averaged 2.93 touchdowns, 27.4 points, and a 16.7 percent red-zone-snap touchdown rate, well below the 22.5 percent league norm. Reaching the red zone ten times and never snapping inside the 10 is ordinarily the fingerprint of an offense stalling around the 12 and sending on the kicker. Only eight of the 142 inverted it with six or more touchdowns, and nobody has matched Georgia's conversion rate: across 9,285 team-games with 10 or more red-zone snaps since 2015, the best anyone posted was 50.0 percent.

That finishing rate is the least sticky number in football, and 80 team-games since 2015 show it. Those 80 scored six or more touchdowns with at most one snap inside the 10. They won 82.5 percent by an average of 25 and finished 44.2 percent of red-zone snaps in the end zone. The very next week that finishing rate fell to 23.9 percent, landing within a point of the league norm and within a point of their own rest-of-season figure, and the goal-line situation came straight back: 5.27 snaps inside the 10, up from 0.6. Only 10 of the 80 ever repeated the feat that season. Against the spread the following week they went 49.3 percent, so the market neither over-adjusted nor under-adjusted.

So a 63-3 night bought Georgia no information about the part of its offense that decides SEC games. One condensed-field rep, on a 3-yard run. The risk is not that scoring from 20 yards out is fragile; it is that a team can play a full game and learn nothing about whether its line can move a pile at the 2, with Western Kentucky next and then Arkansas, Oklahoma, Alabama, and a November that will not be settled from distance.

The one drive that went backwards

Georgia ran 11 real offensive possessions, plus a closing kneel-down, and scored touchdowns on nine of them. The nine touchdown drives began an average of 66.4 yards from the end zone, and none started closer than the Tennessee State 52. The best field position of the night produced the only negative possession of the night, and the Tennessee State defense had nothing to do with it.

Sacovie White-Helton's punt return set Georgia up at the Tennessee State 40 midway through the second quarter, its shortest field of the game by 12 yards. Georgia gained 20 yards on five snaps and posted minus 2.89 EPA, the only negative-EPA possession it ran all night; every other drive fell between plus 0.29 and plus 7.12. Then came the sequence that ended it. Peyton Woodring made a 38-yard field goal, and it was wiped out by a penalty for Georgia having two players wearing No. 51 on the field for the same play. On the 43-yard re-kick, holder Drew Miller never got the snap down, and the play-by-play reads "#90 D.Miller pass incomplete short right to #10 E.Williams thrown to TSU33, TURNOVER ON DOWNS." There were roughly 7:37 left in the second quarter, and a made kick would have made it 38-0.

That is the whole causal chain to Tennessee State's only points. The Tigers took over on downs and drove to Preston Hurless's 37-yard field goal with 2:08 left in the half, 35-3. Nine touchdowns and one special-teams operation failure account for every point scored in this game.

Which brings up the other number worth discounting. Georgia ran 0.746 EPA per play, 11.00 yards a snap, and went explosive on 30 percent of its plays. Across 24,716 FBS team-games since 2014 that grade ranks 104th, the 99.6th percentile, and it is Georgia's best against an FCS opponent in twelve years by 0.18. Tennessee State's defense had allowed 0.104 EPA per play across 12 games in 2025, with a worst game of 0.275; Georgia ran about seven times that. A top-1 percent grade against an FCS team is an opponent-quality artifact, not a forecast. In the 52 comparable games since 2014, the next week came in at 0.271, the correlation between the explosion and the following week's grade was minus 0.147, and the rest of the season decayed 72 percent to 0.224. Not one of the 52 sustained it.

The saving caveat is who those 52 teams were: median pregame Elo 1,556, only three above 1,700. Georgia entered at 1,871, and the six elite-tier programs who did something similar all won their next game and went 42-7 the rest of the way. Six games is a hint, not evidence. Discount the efficiency. The field-goal unit is the one new data point of the night, and it is a negative one.

What Tennessee State actually did

"Ten rushing yards on 30 carries" reads like annihilation, and the snaps underneath it say something quieter. Tennessee State's success rate was 29.8 percent on 57 snaps against its own 2025 season average of 31.8, and its stuff rate was 34.5 percent against a season figure of 36.7. On the two measures that describe whether an offense is staying on schedule, this was a normal Tennessee State night, delivered against a top-three defense in 101-degree heat. The official number needs a footnote too: the NCAA charges sacks to rushing, and the play-by-play separates 29 designed runs for 47 yards from two sacks for minus 27.

What Georgia erased was the ceiling. Tennessee State's longest play of the game was 17 yards. Zero plays of 20 or more on 57 snaps, and only two of 10 or more. Georgia had nine of 20-plus and 28 of 10-plus.

The self-inflicted half is the first-down run. Tennessee State ran on 71.4 percent of its first downs, 15 of 21, for 24 total yards. That is 1.60 a carry, with seven of the 15 stopped at or behind the line, leaving an average second-and-8.2 and third-and-7.8. The result: 28.1 percent of their snaps came on third down at an average of 7.75 to go, nine of them third-and-7 or longer, against Georgia's 10.0 percent. They converted six of 16. The one drive that produced anything was the one where they abandoned the run entirely: 13 plays, nine of 13 passes against 43.2 percent on their other 44 snaps, 3-for-4 on third down, helped by a Georgia personal foul on the drive's first snap, ending in Hurless's field goal.

This exact state, with efficiency held, ceiling gone, and the run game met at the line, occurred 51 times in 24,716 team-games across twelve seasons, about one in every 480. The 44 with a final score went 7-37 and scored 11.7 points against a league average of 28.4, and six of the seven that won did it by holding the other team to 16 or fewer. Two control cells split the cause cleanly. Teams stuffed the same way with their explosives intact, 558 of them, went .493 and scored 27.6. Teams that lost the efficiency too, 86 of them, went .128 and scored 6.7. So the ceiling was worth roughly 15.9 points of scoring, and staying on schedule the way Tennessee State did was worth about five.

The honest limit: 38 analogs with a next game say this state does not persist. Explosiveness snapped back to the season norm the following week for 22 of them, and next-game scoring jumped from 11.7 to 23.4. The missing top end was imposed by Georgia rather than lost by Tennessee State, and those analogs faced opponents averaging only 193 Elo points better, nothing close to this mismatch. What travels is the first-down run rate. If it stays near 70 percent into the conference schedule with a stuff rate above .30, the third-down volume follows, against defenses that cannot claim this one's excuse.

The quarterback run game that never showed up

On 27 Georgia rush attempts, exactly two were designed quarterback carries, both by Millender, both in the fourth quarter: 13 yards on third-and-6 and nine on second-and-5. Stockton, Puglisi and Montgomery combined for zero, Montgomery's lone carry being the closing kneel. Across 33 pass plays there was not one scramble and not one sack. Eleven different Bulldogs carried the ball; one was a quarterback, and he was the fourth one to play.

That is a strange number for this team. In 2025 Georgia quarterbacks took 134 of the team's 576 rush attempts, 23.3 percent. Stockton alone carried 129 times for 462 yards and 10 rushing touchdowns, making him Georgia's second-leading rusher and its leading rushing-touchdown scorer on a team that scored 31 rushing touchdowns. Georgia's quarterbacks carried at least five times in every one of its 14 games, averaging 9.57. Two is below the floor of all fourteen.

In 47 comparable games this looks like a scoreboard artifact rather than a schematic retreat, and the betting market is what sorts it. Teams whose quarterbacks took 20 percent or more of the carries the prior season and then got two or fewer in a game did it 47 times in 1,216 chances, 3.9 percent. Those 47 went 28-19 overall, but 21-2 when favored and 7-17 when not. As favorites of 14 points or more they went 11-0 by 26.2, while the same teams went 93 percent by 26.8 across all 208 of their games as 14-point favorites. The vanished quarterback runs added nothing to the outcome. Nor did they linger: the next game averaged 5.55 quarterback carries, and 28 of 42 were back above five.

Georgia's exact shape is rarer than that. Of 3,478 FBS team-games in 2024 and 2025, only 41 used four or more passers, and those games averaged more quarterback carries, not fewer: 10.49 against 9.05 in one-quarterback games, because deep rotations happen in routs where the backups run. Exactly one of the 41 paired four passers with two or fewer quarterback carries. Widen it to three or more passers and there are ten in two seasons, 9-1 with an average margin of 24.1. Two seasons of box coverage and a closest match of ten games means this should be read as unmeasured rather than reassuring. Stockton was lifted in the second quarter at 15-of-16, the defense never forced a scramble, and Georgia was a 46.5-point favorite. Those are three reasons for a 2 that have nothing to do with abandoning a play type worth 134 carries. The tell arrives the first time Georgia meets a defense that can generate pressure, because quarterback runs are also the sack-exposed snaps, and 129 of them sat on Stockton last year.

What Smart said

On the conditions, Smart was unmoved, telling reporters afterward, per CBS Sports: "We did beat the heat. I don't think the heat impacted us." On the reps his backup quarterbacks banked in the second half, he was more equivocal than a 60-point margin might suggest.

I think they get better work in practice, honestly. But I'm glad for them and their families and the work they do and all the work they put in that they got to play and see a little bit of a reward for what they do.

He spent a longer stretch of the postgame on offensive lineman Jaden Reddell, and the answer was about development rather than anything in the box score. "When it comes to development, this guy is the embodiment of development, and he's still getting better," Smart said, before describing where Reddell started: "He wasn't ready to block. He wasn't physically ready. He wasn't mentally ready. It was a shock to go play at this speed, but he never pouted. He never whined." Then the closing line, which is the one worth keeping: "I still think his best is ahead."

What to watch

Three things carry forward, and the efficiency grade is not one of them. Watch the first genuine goal-line series of the season, because Georgia got exactly one condensed-field snap here and the history says a 63.6 percent finishing rate reverts to the league's 22.5 almost immediately. Watch whether Stockton's carries get re-exposed or rationed against a defense that can actually rush the passer. And watch the field-goal operation, the only unit that generated new and negative information: a made 38-yarder erased by two players wearing the same number, then a hold that never got down, then the opponent's only points. Tennessee State could not punish either error. Alabama can.

Smart had said it himself in the week before the opener, and the night made his point in a way he probably did not intend: "I'm excited for this team to go out and play their first game. We've got to play somebody else to get better. That's the stage we're in."