Florida · Weeks 4–6

Which Florida?A DAWG win over No. 4 Ole Miss, then two Certified SOFTs in a row.

Florida beat No. 4 Ole Miss by 24, then lost its next two games by a combined 47. Since 2019, no other Power 4 team has followed a DAWG game with two straight Certified SOFTs. Here's what the scoreboard and the snaps actually show.

Sept. 26–Oct. 10, 2026 · Gainesville and Columbia, Mo. · Data through Week 6, 2026 · DAWG methodology v3.0

Two weeks ago, Florida was the best story in the SEC.

On Sept. 26, in front of 90,683 at the Swamp, the then-No. 21 Gators beat No. 4 Ole Miss 52–28 as four-point favorites. They scored seven touchdowns, didn't turn it over, and climbed to No. 8 in the next poll.

Then they lost 45–17 at Missouri. Then, on Saturday, they lost 38–19 at home to South Carolina as 10.5-point favorites. On Sunday, Florida was out of the AP Top 25.

So which one is the real Florida: the team that hung 52 on a top-five opponent, or the team that lost its next two by a combined 47? We dug into all three games. The short answer is that the high was a little less high than it looked, and the lows were every bit as low.

Three games, 15 days

From a DAWG win to two Certified SOFTs

Sat., Sept. 26 · Gainesville
vs. No. 4 Ole Miss
W 52–28
Line: Florida −4
+19.6DAWG
PLAY +3.3
Florida entering: No. 21
Sat., Oct. 3 · Columbia, Mo.
at No. 25 Missouri
L 17–45
Line: Florida −5.5
−34.0Certified SOFT
PLAY −20.6
Florida entering: No. 8
Sat., Oct. 10 · Gainesville
vs. South Carolina
L 19–38
Line: Florida −10.5
−36.2Certified SOFT
PLAY −23.8
Florida entering: No. 16
AP rank 21→8→16→Unranked

DAWG Index v3.0 · ranks are the AP poll in effect at each kickoff, then the poll of Oct. 11

Nobody had done this

DAWG grades every game against what the closing line expected. Florida's three-game run went DAWG, Certified SOFT, Certified SOFT: +19.6, then −34.0, then −36.2.

Since 2019, no other Power 4 team has followed a DAWG game with two straight Certified SOFTs. Back-to-back Certified SOFTs alone have happened 10 times, Florida's included. And the 53.6-point drop from the Ole Miss game to the Missouri game is one of the 19 biggest game-to-game drops among 5,642 pairs of back-to-back games in that span.

The high: Ole Miss

This wasn't a fluke win. Florida outgained Ole Miss 498 yards to 426, had zero turnovers to Ole Miss's two, and led 17–6 at halftime. It was also closer than the final score: Ole Miss scored early in the third quarter and again late in it, and it was 24–21 with about five minutes left in the third.

From there, Florida outscored Ole Miss 28–7. One of those touchdowns came right after Florida sacked Ole Miss's quarterback and recovered the fumble at the Florida 43.

Here's the wrinkle. On the scoreboard, Florida beat its line by almost 20 points. Snap for snap, it was only a little better than a four-point favorite usually is: PLAY +3.3. Ole Miss moved the ball well too, at about two-tenths of an expected point per snap. Florida was better, and it finished drives. But the snaps say "good win," not "24-point statement."

The scoreboard and offense + defense

The win was bigger on the scoreboard. The losses weren't.

DAWG · the scoreboardPLAY · offense + defense
Ole Miss
Missouri
South Carolina
← Worse than expectedBetter →

Both grades are in points against what the closing line expected. DAWG grades the final margin with garbage time out; PLAY grades every competitive snap with expected points.

DAWG Index v3.0

The first low: Missouri

No turnovers by either team, no fluky bounces. Missouri just beat Florida, 560 yards to 325. It scored on eight of its first 10 possessions, led 20–10 at the half and 45–10 early in the fourth quarter. Florida's last touchdown came in garbage time, which DAWG doesn't count.

Florida was a 5.5-point favorite. That's a −34.0 on the scoreboard and a −20.6 on the snaps. In fairness to the Gators, Missouri looks a lot better now than it did at kickoff. The Tigers were No. 25 that day; they're 5–1, No. 13 in the new poll and fifth among 68 Power 4 teams in season DAWG.

The second low: South Carolina

Saturday looked different on paper. South Carolina outgained Florida by just six yards, 393 to 387. The difference was what happened at the ends of drives.

Florida led 7–0 and 10–7. At 10–10 late in the second quarter, South Carolina intercepted Aaron Philo and returned it to the Florida 26, and LaNorris Sellers threw a touchdown pass five plays later. Florida missed a 43-yard field goal at the end of the half. From 10–10 until Florida's touchdown with a minute left, South Carolina outscored Florida 28–3. Sellers threw four touchdown passes and ran for another.

Florida had chances to make it a game. Down 31–13 early in the fourth quarter, Philo fumbled at the South Carolina 15. A few minutes later, Florida turned it over on downs at its own 27, and South Carolina scored three plays after that.

Turnovers decided the margin, but they weren't the whole story. Leave Florida's giveaways out and South Carolina was still better per snap, by about a quarter of an expected point to under a tenth. That's −36.2 on the scoreboard and −23.8 on the snaps. Against the 2026 Power 4 field, only one game has graded worse than that −36.2. Florida's loss at Missouri is third.

Season DAWG rank among 68 Power 4 teams

5th to 60th in two weeks

DAWG Index v3.0 · rank by season-average DAWG after each week's games · lower is better

Haven't we seen this before?

Florida fans might. The third-worst two-game stretch since 2019 also belongs to Florida: in 2021, a loss to Georgia followed by a loss at South Carolina. That one was a SOFT and then a Certified SOFT, not two Certified SOFTs, so this year's version is a little worse. South Carolina was the second game both times.

Worst back-to-back games, 2019–2026

Florida's two weeks, in context

Sum of DAWG in consecutive graded games, same season, for the 68 current Power 4 teams. 5,642 pairs since 2019.

DAWG Index v3.0 · every game in the public CSV

So which Florida is real?

Three games can't answer that, and we'd be guessing if we pretended they could. Here's what the grades do establish.

The Ole Miss win was real, and the margin outran the snaps. Florida beat what its line expected by 19.6 points on the scoreboard. Snap for snap, it was +3.3, about what a four-point favorite usually does. It played well and finished drives. It didn't dominate the way 52–28 suggests.

Both losses were bad on both measures. −34.0 and −20.6 at Missouri, −36.2 and −23.8 against South Carolina. Neither was a scoreboard fluke. Missouri outgained Florida by 235 yards with no turnovers on either side, and South Carolina was better per snap even with Florida's giveaways set aside.

Over six games, Florida has played below its lines. It's −5.8 in season DAWG, 60th of 68, and −7.9 in PLAY over five graded games, after being favored in all six. That's the season so far. It isn't a verdict on the team: random swings alone put a six-game average anywhere within about ±11 points of zero (a 95% range), and −5.8 is inside it.

That's as far as the evidence goes. One very good afternoon, two very bad ones, all within 15 days.

Next up is a trip to No. 7 Texas on Saturday. We're not going to tell you what happens there. We're just going to grade it.

The numbers

Ole MissMissouriS. Carolina
ScoreW 52–28L 17–45L 19–38
Closing line−4−5.5−10.5
Total yards498–426325–560387–393
Turnovers0–20–03–0
DAWG+19.6−34.0−36.2
PLAY+3.3−20.6−23.8

Yards and turnovers are Florida's first.

How the grades work. DAWG compares the final margin, with garbage time removed, to the margin teams usually produce at that closing spread. PLAY asks the same question of the offense and defense, using expected points added (EPA) from the open-source cfbfastR model; special teams aren't in it. Neither is a power rating, and neither predicts anything. How it works · Florida's season · Every game (CSV)