We called the wrong team, and so did everyone we anchor to. Kentucky won at Kyle Field by 10 as a 16.5-point underdog, the first time this season our call, the market and the model all had the wrong winner. It was 7–7 at the half. Kentucky then scored 21 straight in the third quarter on drives of 75, 83 and 33 yards, the short one handed over by a failed fourth-and-2 at the Aggie 32, and led 28–7 entering the fourth. Kenny Minchey, the quarterback our board said A&M's coverage would smother, went 14 of 18 for 252 yards and two touchdowns without a turnover, with completions of 53, 42 and 32 yards against the big-play problem we had on the page and discounted. Marcel Reed threw 49 times for 236 yards at 4.8 an attempt with two interceptions, the second at the Kentucky 7 with the Aggies down 10. The Aggie run game, the thing we were most worried about, was fine: 187 yards at 5.3. The things we were least worried about lost the game.
We picked the wrong winner, and so did the betting line and our computer. Kentucky won by 10 points even though the people betting on the game expected Texas A&M to win by more than 16. It was tied 7 to 7 at halftime. Then Kentucky scored three straight touchdowns in the third quarter and led 28 to 7 going into the last quarter. Kentucky's quarterback, Kenny Minchey, completed 14 of 18 passes for 252 yards and two touchdowns and never gave the ball away, including three long completions against a Texas A&M pass defense we had said was the better side. A&M's quarterback, Marcel Reed, threw 49 passes for only 236 yards and had two of them caught by Kentucky, one at the Kentucky 7-yard line when A&M was down by 10. A&M ran the ball well, which was the thing we worried about most. The things we worried about least are what lost the game.
16 falsifiable claims graded: the score call, three tells, twelve matchup-board edges. Six storylines graded on whether they affected the game or the read. Full grades live in the season ledger (data/2026/season_ledger.csv), with each pregame claim quoted verbatim.
The call was wrong, and it was wrong for the reason the page named as unlikely. We printed Texas A&M 33, Kentucky 17 and wrote that the way A&M wins "is already written": the pass rush and the passing game. The pass rush produced two sacks and one hurry on 25 dropbacks. The passing game produced 4.8 an attempt and two interceptions. The market's 16.5 was off by 26.5 and the model's 21 by 30.7. The base rate we leaned on, favorites of 14.5 or more at 712-38 since 2023, had its 39th loss.
Our score prediction was WRONG, and it went wrong in the exact way we said it would not. We wrote down Texas A&M 33, Kentucky 17 and said A&M would win with its pass rush and its passing game. The pass rush reached the quarterback twice all afternoon. The passing game gained less than five yards a throw and gave the ball away twice. We also printed a history that said teams favored by this much almost never lose. This was one of the few times they do.
Minchey's passing game against Aggie coverage, Texas A&M big: the worst chip on the page. A quarterback we described as unable to stretch anyone went 14 of 18 for 252, 14.0 an attempt, two touchdowns and no interceptions, with DJ Miller's 53-yarder, Kenny Darby's 42-yard score and Henry Boyer's 32-yarder. We had the Aggie big-play problem on the board as a slight Kentucky edge and did not let it touch the coverage chip next to it.
Our biggest miss was Kentucky's passing. We said A&M's pass defense would shut it down. Instead Kentucky's quarterback completed 14 of 18 throws for 252 yards, including passes of 53, 42 and 32 yards, and two touchdowns. We knew A&M gives up long passes, we said so on the same page, and we still called this one for A&M.
Reed against the Wildcat secondary, Texas A&M big: wrong the other way. Kentucky broke up seven passes and intercepted two. Reed completed 10 of his first 15, which our tell said would prove the edge; then he went 16 of 34 the rest of the way and nothing longer than 21 yards. Bateman's coverage took away everything past the sticks, which was the exact fear in our "Bateman's book" line, and the board did not price it.
We were also wrong about A&M's passing. We said Reed would pick apart Kentucky's pass defense. Kentucky knocked down seven of his passes and caught two of them. Reed started well, then completed fewer than half his throws the rest of the way, and no A&M catch went longer than 21 yards. Kentucky's defensive coach used to run A&M's defense and knew what was coming; we wrote that down and still picked A&M.
Three more chips went to Kentucky. Wildcat run game against the Aggie front, A&M slight: CJ Baxter ran 19 times for 87 and two touchdowns, Jason Patterson 6 for 58, 4.3 a carry for the team. Aggie deep game against Wildcat coverage, Even: A&M's longest completion was 21 yards. The rebuilt Aggie line against Kentucky's front, Kentucky slight: we followed Sunday's rule and gave the opponent the edge until proven, and the line proved it, 187 yards at 5.3, Owens 4.1 a carry, Reed 80 on 10. That chip was wrong in the Aggies' favor and it is still wrong. Coaching & Luck, Even: the turnovers went 2-0 Kentucky, the fourth-down decision at the Aggie 32 handed Kentucky a 33-yard touchdown drive, and Elko's own verdict was "they beat our ass."
Four more of our matchup calls were WRONG. We said A&M would slow Kentucky's running game; Kentucky's CJ Baxter ran for 87 yards and two touchdowns. We said the two sides were even on A&M's long passes; A&M did not complete a pass longer than 21 yards. We gave Kentucky the edge when A&M ran the ball; A&M ran for 187 yards, so that one was wrong in A&M's favor. And we said coaching and luck were even; A&M gave the ball away twice, Kentucky never did, and A&M's coach said afterward that Kentucky "beat our ass."
The first tell misled us. Ten of Reed's first 15 throws were complete, above the nine we said would prove the passing edge was real. It was not real. A completion rate over 15 short throws measured the wrong thing; yards an attempt would have said 6.0 over the same window and 4.8 by the end. We grade the read, and the read was WRONG.
Our first early sign pointed the wrong way. We said that if Reed completed nine of his first 15 passes, A&M's passing would win the day. He completed ten, and it did not. Counting completions on short passes told us nothing about how far the ball was going.
Texas A&M's first drive went 14 plays and 76 yards from its own 13 and ended on downs at the Kentucky 11; its second reached the Kentucky 29 and ended with Sam Greene's interception. Kentucky's first two possessions produced a punt and a 39-yard miss by Adam Zouagui. Carsyn Baker's 2-yard run capped a 14-play, 79-yard drive for 7–0 with 5:18 left in the half, and Minchey answered in six plays, the 53-yarder to DJ Miller setting up Boyer's 3-yard touchdown with 1:59 left. The third quarter was the game: Kentucky 75 yards in seven plays, Darby's 42-yard touchdown; A&M three-and-out; Kentucky 83 yards in nine, Baxter from the 3; A&M fourth-and-2 from its own 32, incomplete to Baker; Kentucky 33 yards in four, Baxter from the 2 with 16 seconds left in the quarter, 28–7. The Aggies answered with a 13-play, 75-yard drive and Reed's 5-yard run, Zouagui hit from 52 for 31–14, and Reed's 40-yard run set up first-and-goal before Jordan Castell intercepted him at the 7 with 6:41 left. Reed's 5-yard pass to Micah Riley made it 31–21 with 3:45 to play, and the last Aggie drive ended on downs at its own 27.
| Texas A&M | Kentucky | |
|---|---|---|
| Total yards | 423 | 408 |
| Passing / rushing | 236 / 187 | 252 / 156 |
| Yards an attempt (pass) | 4.8 | 14.0 |
| Third downs · fourth downs | 9 of 17 · 3 of 6 | 3 of 9 · 1 of 1 |
| Turnovers | 2 | 0 |
| Penalties | 2 for 10 | 7 for 59 |
| The board's widest gap | Kentucky's protection against the Aggie pass rush: two sacks, one hurry, and a quarterback who was never rushed into a mistake | |
Next week is LSU, on the road. Texas A&M at LSU, Saturday, Sept 26, Tiger Stadium, kickoff not yet set. The scouting report and a frozen call publish Monday morning. Saturday told us the run game is real and the coverage is not; Monday's board starts from there, not from the preseason numbers.
Ten of 15, completed by 9:32 of the second quarter. The tell said the edge was real; the game said 4.8 an attempt, two interceptions, seven passes broken up and nothing longer than 21. The number we chose to watch was answered and the answer was wrong, so the read is WRONG. Timing goes in the ledger; the lesson goes in the template: a completion count on short throws does not measure a passing edge. Yards an attempt does.
WRONG. Reed completed 10 of his first 15 passes, more than the nine we said would prove A&M's passing was the better side. It was not the better side: 4.8 yards a throw, two passes caught by Kentucky, seven knocked down. We watched the wrong number.
Two sacks (DJ Hicks in the first quarter, T.J. Searcy early in the second) and one hurry all afternoon: two or three pressures across the first 20 dropbacks, at the "lives on paper" end of the tell. The widest gap on our board did live on paper, and Minchey threw for 252 on 18 attempts. The read was RIGHT, which is cold comfort: it was right about our own chip being wrong.
RIGHT. We said that if A&M reached Kentucky's quarterback five times early, our biggest edge was real, and that two or fewer meant it existed only on paper. A&M got to him twice. The edge was on paper, and Kentucky's quarterback had all day.
Seventy-one yards on the first 15 carries, 4.7 a rush, past the 60 we set, and the day finished at 187 on 35 (5.3): Owens 18 for 74, Reed 10 for 80 with a 40-yarder, Edwards 5 for 28. The line took the step. RIGHT, and the one part of Monday's page the game confirmed outright.
RIGHT. We said 60 yards on A&M's first 15 runs would show the new blockers had improved. They gained 71, and A&M finished with 187 rushing yards at 5.3 a carry. The running game was the one thing that went the way we said.
All twelve pregame pairings, the edge we printed Monday, and what the game said. Chips are quoted verbatim from the scouting report.
| Pairing | Our call | What happened | Grade |
|---|---|---|---|
| Reed's Passing Game vs. the Wildcat Secondary | Edge: Texas A&M, big | 26 of 49 for 236 (4.8), 1 TD, 2 INT; 7 passes broken up; longest completion 21 | Wrong |
| The Rebuilt Aggie Line vs. Kentucky's Front (Run Blocking) | Edge: Kentucky, slight | 187 on 35 (5.3); 2 Kentucky TFL; the line won the front | Wrong |
| Owens and the Committee vs. the Wildcat Run Defense | Edge: Even | Owens 18-74 (4.1), Edwards 5-28, Baker a TD; the backs were fine, not dominant | Right |
| Aggie Pass Protection vs. the Wildcat Pass Rush | Edge: Even | 0 sacks, 5 hurries on 49 attempts; Reed had time, and Even undersold it | Partly |
| Aggie Deep Game vs. Wildcat Coverage | Edge: Even | Longest A&M pass 21 yards; nothing over the top all day | Wrong |
| Minchey's Passing Game vs. Aggie Coverage | Edge: Texas A&M, big | 14 of 18 for 252 (14.0), 2 TD, 0 INT; completions of 53, 42, 32 | Wrong |
| Wildcat Run Game vs. the Aggie Front | Edge: Texas A&M, slight | 156 on 36 (4.3); Baxter 19-87-2, Patterson 6-58 | Wrong |
| Wildcat OL vs. the Aggie Front (Run Blocking) | Edge: Texas A&M | 4.3 a carry allowed, 3 Aggie TFL; Kentucky 3 of 9 on third down but 1 of 1 on fourth | Partly |
| Wildcat Pass Protection vs. the Aggie Pass Rush | Edge: Texas A&M, big | 2 sacks, 1 hurry on 25 dropbacks; "big" promised a ruined afternoon and Minchey had a clean one | Partly |
| Wildcat Deep Game vs. the Aggie Big-Play Problem | Edge: Kentucky, slight | 53 to Miller, 42-yard TD to Darby, 32 to Boyer; the problem is the profile | Right |
| Special Teams & Field Position | Edge: Texas A&M, big | Zouagui missed from 39 and hit from 52; White 45.5 on two punts; Olano never attempted a kick; the short field Kentucky got came from a fourth down, not a kick | Partly |
| Coaching & Luck | Edge: Even | Turnovers 2-0 Kentucky; the fourth-and-2 at the Aggie 32 became a 33-yard touchdown drive; Stein: "the more physical team won"; Elko: "they beat our ass" | Wrong |
Reading the grades honestly: two of twelve edges were right, and both were the ones that leaned Kentucky or Even. Every "big" chip pointed at Texas A&M and every one of them was wrong or partly wrong. The board was built from 2025 percentiles and two games of 2026 watch, and it did not weight the one 2026 fact that decided the game, that this coverage leaks long throws and Kentucky had a quarterback who could make them. That goes in the template: a "big" chip needs a 2026 number behind it, not a 2025 one.
Our model said Texas A&M by 21 (36.6–15.9); the market said Texas A&M by 16.5. Actual: Kentucky by 10. Model error 30.7, market error 26.5. The market takes the week by being less wrong, which makes the season line market 3, model 2 on margin; straight up the market is 4-1 and the model 3-2. Neither had a case for Kentucky. The published call stays market-anchored; the rule was set by a backtest of hundreds of games, and one upset is the kind of thing the backtest already counted.
The base rate we printed Monday: favorites of 14.5 or more were 712-38 straight up since 2023, with a median winning margin of 28. Saturday made it 712-39. The median has now beaten the line twice and missed once; the loss is the 5% the number always carried. Raw rows: predictions.csv.
The film behind the grades. Every link below was verified live before publish; anything we could not verify was dropped rather than guessed at.
Kentucky Wildcats vs. No. 9 Texas A&M Aggies | Game Highlights (SEC, YouTube). Miller's 53-yarder, Darby's 42-yard score, the fourth-and-2, both Reed interceptions. The condensed game is on the same channel.
No. 9 Football Falls to Kentucky in SEC Opener (Texas A&M Athletics). The scoring sequence, the 105,186 attendance, Craver's career-high 11 catches, and Elko on "eye violation issues" in coverage.
Kentucky 31, Texas A&M 21 (AP via ESPN). Both coaches postgame, the box, the drive chart. KBTX's live file has the game as it happened.
Postgame press conference video: none passed link verification by press time; when Texas A&M Athletics posts the clip, the Monday edition carries it.
How the game is landing across the sanctioned doors, as of Sunday morning. These are conversations, not our claims; our claims are graded above.