Texas by a point and a half is a coin flip with a home crowd attached, and the board says the flip lands on one question: whether Ohio State's rebuilt defense is still the unit that held Texas to 7 and 14 points in the last two meetings. The best offense on the field is Ohio State's, by every 2025 percentile and by Julian Sayin's 21-of-25 opener, and the widest computed gaps on this page all point at the Buckeyes when Texas has the ball. But the defense that produced those gaps lost eight starters and three first-round picks, and Texas has added the receivers, backs and linebacker it did not have in Columbus. The teams in one line each: Ohio State is the same offense with a defense it has not yet had to trust; Texas is a deeper roster that has scored 21 points in eight quarters against these people and went 0 for 2 in the red zone in Columbus.
Texas is favored by less than a field goal, so this is close to a coin flip with a home crowd behind it. Ohio State has the better offense. The big question is Ohio State's defense: it held Texas to 7 and 14 points in the last two meetings, but eight of its starters left for the NFL. Texas has more good players, the best pass rusher on the field, and a bad habit of stalling near the end zone against these guys.
The 26–24 call is market-anchored (−1.5, O/U 49.5, ~52% implied win probability; DraftKings opened and sits at −1.5, Bovada opened at −2.5 and has drifted to −2). Our own SP+ model disagrees with the market for the first time this season: it says Ohio State 28–26, leaning the visitor by 2.4 and sitting 3.9 points from the line. It is reference only, because the backtest says the market wins, but it is printed so it can be graded.
Base rates from 226 comparable games (favorites of 1 to 3 points, 2023–25): the favorite went 111-115 straight up, and home favorites in that band went 60-66. There is no edge in the number; a line this short is the market saying it does not know either. Pace check: these offenses averaged about 67 (Texas) and 64 (Ohio State) snaps a game in 2025, and the two meetings since January 2025 produced 21 and 42 points, which is why a 49.5 total for two teams that just scored 59 and 56 is not low.
Finish in the red zone. Zero for two inside the 20 in Columbus last year; a 62.5% scoring rate in losses since 2023 against 84.4% in wins. Touchdowns, not threes, are the whole difference in a two-point game.
Own the middle of the field. Sayin completed 90.6% of his throws between the hashes inside 9 yards last year. Cryer and Biles have to take that away and make him throw outside and late.
Win the turnover battle. Texas is 6-12 under Sarkisian when it loses it and has forced one Buckeye turnover in the last eight quarters against them.
How Ohio State wins it
Keep Sayin clean from Simmons. Phillip Daniels was benched on the opening drive of the opener; whoever plays tackle has the SEC's sack leader across from him all night.
Get Jeremiah Smith the ball early. Forty-six yards in two games against Texas; if the third looks like the first two, the Buckeye offense is playing without its best player again.
Make Texas run into the interior. Texas managed 2.6 a carry on A- and B-gap runs against Texas State; Houston and Walker inside can turn the Longhorns one-dimensional and hand the game to a rebuilt secondary.
Final call: Texas 26, Ohio State 24. The visitors have the better quarterback stat line and the better receiver, the hosts have the deeper roster, the pass rusher, and a stadium that will be loud at 6:30. The two things that decided the last two meetings, Texas's red zone and Ohio State's ability to make Smith disappear and win anyway, are exactly the two things each side changed this offseason. Watch the first Texas trip inside the 20 and the first Smith target; the loser of those two moments is chasing the game.
Our call: Texas 26, Ohio State 24. Ohio State has the better quarterback numbers and the best receiver. Texas has more depth, Colin Simmons, and a loud stadium. Watch two things early: whether Texas scores a touchdown the first time it gets close, and how much the ball finds Jeremiah Smith. Whoever loses those two moments is playing catch-up.
Game-Week Intel
Texas Storylines
Biles was a problem on day one. The transfer linebacker's Texas debut (five tackles, a forced fumble, an interception) made him co-SEC Defensive Player of the Week and gives Muschamp a second-level player who can run with Ohio State's tight ends and backs. The defense forced three turnovers, including Graceson Littleton's 59-yard scoop-and-score. On Texas Football
Sarkisian is not throwing all fastballs. His stated plan for Sayin: "When you have a plan against somebody like him, you can't just throw all fastballs. As good as a fastball as you have, if that's all you've got, sooner or later they're going to hit it." And on Smith: "We'd better know where No. 4 is on every snap." Expect disguised coverage over the middle, where Sayin lived last year. On Texas Football
Cooper is out, in a boot. The freshman running back turned an ankle in the second quarter against Texas State; Sarkisian said he "will be for the foreseeable future." He was third on the depth chart behind Smothers and Brown, so the backfield rotation shortens rather than changes. Heavy, via Yahoo
Ohio State Storylines
The line is under evaluation. Right tackle Phillip Daniels was benched after a missed assignment on the opening drive, the front produced 35 rushing yards on 17 first-half carries, and Day said he will "continue to evaluate that throughout the week to see if we have the right mix out there." Four of five 2025 starters are back; the question is which five play. Buckeyes Wire, via Yahoo, Yahoo Sports
Eight defensive starters are gone, and the numbers say it matters more than the coverage suggests. Arvell Reese went fifth, Sonny Styles seventh, Caleb Downs eleventh, Kayden McDonald in the second round, Caden Curry and his 11 sacks in the sixth. Kenyatta Jackson Jr., Jaylen McClain and Jermaine Mathews Jr. are the returning starters; Payton Pierce steps in at middle linebacker. Every 2025 Buckeye defensive percentile on this page belongs to a unit that no longer exists. Yahoo Sports, Eleven Warriors
Smith has not beaten Texas yet. Six catches for 43 yards in Columbus, one catch for 3 yards in the Cotton Bowl, seven for 46 across the two games, and Ohio State won both anyway. Day says Smith is "always at the forefront" of the game plan; Arthur Smith, the new coordinator, called seven 20-yard plays in his debut. On Texas Football, Eleven Warriors
Chatter check: The week's loudest threads are motivational, not material: Day's "You come to Ohio State to win games like this" (Eleven Warriors), the 30 recruits Sarkisian has coming to the game and Jaxon Smith-Njigba's supposed grudge (Hook'em Headlines), and Texas fans pleased with No. 4 as a launching pad. Nothing in it moves a number on this page. Sentiment on the Texas side reads confident and a little haunted by the red zone; on the Ohio State side, curious about its own offensive line.
Availability & Injury Report
Derrek Cooper · Texas RB · ankle · Out: in a boot "for the foreseeable future" per Sarkisian (Sept 7). Third on the depth chart; well below the 30% unit-share downgrade threshold, so no chip moves. Trevor Goosby · Texas LT · Healthy: the August bone bruise is behind him; he started and played the opener. Cole Hutson · Texas OL · Ineligible pending the NCAA: not a projected starter; Sarkisian says the program is working on it. Qua Russaw · Ohio State DE · Expected to play: dressed but did not play in Week 1; Day says he had "a full week" of practice. Phillip Daniels · Ohio State RT · Benched, not injured: starting five to be settled this week. K. Parker · Ohio State WR · Questionable (undisclosed, per Covers). This is the game-week baseline; Friday's availability pass confirms it against the final depth charts.
Two vintages: the full 2025 season (garbage time excluded) and 2026 through one game each, printed as a watch, not a verdict. Texas's season-level baseline, the 2025 review, roster ledger and unit outlooks, lives in the 2026 Preseason Report; this section carries only what decides THIS game. New to a metric? Every acronym, calculation, and selection rationale lives in the Data Dictionary.
Schedule-adjusted view (SP+)
Metric
Texas
Ohio State
What it means
SP+ overall, 2025 final
+16.2 (No. 17)
+30.1 (No. 2)
Points better than an average team, adjusted for who you played.
SP+ overall, 2026 in-season
+25.4 (No. 4)
+30.0 (No. 1)
Same rating after one game; still mostly the preseason prior.
SP+ offense, 2026
+39.2
+40.0
Points the offense is worth against an average defense. A dead heat.
SP+ defense, 2026
+14.2
+10.1
Lower is better; the model still trusts the Buckeye defense more than the roster turnover suggests.
Returning production, 2026
72.1% (passing 85.4%)
68.3% (passing 94.0%)
Share of last year's value that came back. Both high; Ohio State's is almost all offense.
So whatFour and a half SP+ points to Ohio State, minus 2.5 for the home field, is a two-point Buckeye lean, and that is exactly what our model prints. The market gives Texas a point and a half instead, which means the market is pricing something the ratings cannot see yet: eight new defensive starters, or a Manning who was not this good in August 2025.Our rating system says Ohio State is about two points better even after giving Texas credit for home field, and that is what our model prints. The betting line leans Texas instead. So the market is counting on something the ratings cannot see yet: eight new Ohio State defenders, or a better Arch Manning than last August's.
Scoring value added per snap. Returns 94% of its passing value.
Offensive success rate
51.7% (99th)
Stayed on schedule more than almost anyone; this offense does not need big plays.
Offensive explosiveness
1.16 (9th)
And it did not get them. The Ball State tape (seven 20-yard plays) says Arthur Smith wants that to change.
Rushing PPA/play
0.078 (35th)
Bo Jackson's 1,090 yards came at 6.1 a carry, but the run game as a whole was ordinary.
Havoc allowed
11.2% (95th)
Sayin was almost never touched. Four of five blockers return; the fifth got benched.
Defensive PPA/play allowed
−0.023 (98th)
The best defense in the country by this measure. Eight of its eleven starters are in the NFL.
Defensive explosiveness allowed
1.07 (99th)
Nobody got behind it. That unit's coverage core is the part that left.
Field goals
16 of 20 (80%)
Fine, not a weapon; Texas kicked 83%.
So whatRead the offensive rows at face value and the defensive rows as history. The offense that beat Texas twice is intact and has added a play-caller who wants explosives; the defense that beat Texas twice is a memory with three new starters at linebacker and safety. The Buckeye bars on the board below carry that split.Trust the offense numbers; treat the defense numbers as history. The Ohio State offense that beat Texas twice is back with a new play-caller who likes long plays. The defense that beat Texas twice mostly left for the NFL.
Turnover watch (in-season)
After one game: Texas +2 (three takeaways, one giveaway), Ohio State 0. The context worth knowing: Texas was the luckiest team on this page in 2025, +14 in turnover margin with +9 on interceptions alone, and that number snaps back harder than almost any in football; Ohio State's +3 was all fumble margin, the luckiest stat there is, so neither number is a habit. Against these Buckeyes specifically, Texas has forced one turnover in eight quarters. A Texas takeaway on Saturday is a swing, not a habit.
Ohio State's 2025, Game by Game
One row per game: the bar is the final margin, growing rightward in crimson for wins and leftward in ink for losses, with the score printed on every row. (Texas's season log lives in the preseason report; only the opponent's form travels with the weekly preview.)
Ohio State. 14 games, 12–2
Twelve straight wins by an average of 29, shaded, then two games that scored 24 points combined: the Big Ten title game against Indiana and the Cotton Bowl quarterfinal against Miami.
Win (margin right of zero)Loss (margin left)The 12–0 start
The frame for all of it: Ohio State's 2025 came against the Big Ten and ended with two playoff-caliber losses in which its offense scored 10 and 14; Texas's came against the SEC, including this game's 7-point first act. Both Week 1 lines came against overmatched opponents (Texas State, Ball State), which is why 2026 percentiles appear here only as a watch. The SP+ table is the fair comparison; the eight departed starters are the fair discount on the Buckeye defense.
The Matchup Board
Every unit, reviewed inside the collision it actually plays. The bars are computed: each side's national percentile on the pairing's headline metric (2025 data, garbage time excluded, unadjusted for schedule), bar length = the gap. The chips are our call after adjusting for competition and roster change; where a chip disagrees with its bar, the note says why. Every Ohio State defensive bar carries the eight-starters-gone discount.
TexasOhio StatePercentiles at the ends (higher = always better) · bar = computed gap · chip = our adjusted edge
The Quarterbacks: Manning vs. Sayin
Edge: Ohio State, slight
QB Arch Manning · Texas
Year two as QB1. Rough September last year (17 of 30 for 170 in Columbus), elite finish, and a Week 1 that looked like the finish: 20 of 27, four touchdowns. Day, unprompted: "He's really improved, and I give him credit for putting the work in."
QB Julian Sayin · Ohio State
The most efficient passer in the country last year, 12-2 as a starter, now in Arthur Smith's offense. Lives between the hashes: 96 of 106 on middle throws inside 9 yards in 2025, per PFF charting. His first road start of the season.
Manning · 2025 pass yds
3,163
Sayin · 2025 pass yds
3,610
Manning · 2025 comp %
61.4
Sayin · 2025 comp %
77.0
Manning · 2025 pass TD / INT
26 / 7
Sayin · 2025 pass TD / INT
32 / 8
Manning · Week 1 yds/att
11.3
Sayin · Week 1 yds/att
12.8
So whatSayin's 2025 is the better season by every column, earned in the same tier of football, so the chip goes to Ohio State even though Manning's arrow points up. Watch Sayin's first third-and-long against a disguised look: if he holds it, Sarkisian's "not all fastballs" plan is working; if the ball is out to the middle on time, it is 2025 again.Sayin had the better season last year in the same level of football, so Ohio State gets the edge here even though Manning is improving. Watch Sayin's first long third down against a disguised coverage: if he holds the ball, the Texas plan is working; if he throws over the middle on time, it looks like last year.
When Texas has the ball
Coleman and Wingo vs. Sanchez and Mathews
Edge: Even
+33 Ohio Statecomputed gap. TX passing PPA 63rd pctl vs OSU pass PPA allowed 96th; the bar is last year's secondary, the chip is this year's
Cam Coleman, Ryan Wingo, Emmett Mosley V · Texas
A receiver room Texas did not have in Columbus: Coleman by transfer (two touchdowns in his debut), Wingo established (7 for 121), Mosley healthy. Texas produced 16 explosive plays in the two meetings and turned five of them into touchdowns.
Devin Sanchez, Jermaine Mathews Jr. · Ohio State
A five-star recruit and Jermaine Mathews Jr., in front of a safety group that lost Caleb Downs to the eleventh pick. The 2025 unit allowed a 99th-percentile 1.25 on pass explosiveness; this one has played Ball State.
Texas · pass explosiveness, 2025
1.75
OSU · pass explosiveness allowed, 2025
1.25
Wingo · Week 1 rec yds
121
Coleman · Week 1 rec yds
70
So whatThe bar says Ohio State by a lot and the roster says the bar is out of date: 92nd-percentile explosive receivers against a secondary that lost its best player to the eleventh pick. Called Even until Sanchez and Mathews hold up on the first two deep shots; the explosives are coming either way, the question is whether they end in the end zone this time.The numbers say Ohio State's pass defense was much better last year, but that group lost its best player to the draft, and Texas added a big-play receiver. We call it even until the two Ohio State corners survive the first two deep throws.
Texas OL vs. the Buckeye Interior (Run Blocking)Edge: Ohio State
36%
93%
The board's biggest gap (−57) and the one Week 1 confirmed: 36th-percentile line yards in 2025, then 14 gap runs for 37 yards against Texas State. Houston and Walker inside for Ohio State are the reason the chip is not "big": McDonald, the 2025 anchor, was a second-round pick, and the front is younger than its bar.
Smothers and Brown vs. the Buckeye Run DefenseEdge: Ohio State
39%
97%
Two portal backs (Smothers 7.0 a carry in the opener, Brown 3.1 with two scores) against a run defense that graded 97th but lost Reese, Styles and McDonald. The 2025 Texas rushing value was 87% Manning's legs; if Smothers has 60 yards by halftime, this chip was wrong in the right direction.
Texas Pass Protection vs. the Buckeye Pass RushEdge: Ohio State, slight
42%
87%
Goosby is healthy and the line allowed one sack in the opener; Ohio State's rush lost Curry (11 sacks) and Reese but keeps Kenyatta Jackson Jr. (6.5) and gets Russaw back. Slight because the pressure that wrecked Texas in Columbus is mostly in the NFL now.
Texas Deep Game vs. Buckeye CoverageEdge: Even
92%
99%
Strength on strength on paper (−7), and both sides are different now: Texas's 92nd-percentile explosiveness added Coleman, Ohio State's 99th-percentile coverage lost Downs. See the Coleman-and-Wingo card above; this is the same collision from 30 yards further out.
When Ohio State has the ball
Jeremiah Smith vs. the Texas Secondary
Edge: Ohio State
+37 Ohio Statecomputed gap. OSU passing PPA 98th pctl vs TX pass PPA allowed 61st: the widest gap when Ohio State has the ball
Jeremiah Smith, Devin McCuin, Brandon Inniss · Ohio State
Smith is the best player on the field (87 catches, 1,243 yards, 12 touchdowns; 8 for 151 in one half last week). McCuin, the UTSA transfer, gets the Chris Olave comp from his own coach. Against Texas, though: 46 yards in two games.
Mascoe and Kade Phillips · Texas
A 61st-percentile pass defense by 2025 PPA that held Smith twice and forced three turnovers in the opener. Muschamp's answer is where the safeties sit; Sarkisian's stated priority is knowing "where No. 4 is on every snap."
Smith · 2025 rec yds
1,243
Smith · yds in two games vs. Texas
46
OSU · passing success rate, 2025
58.4%
Texas · pass success allowed, 2025
39.1%
So whatThe data says Ohio State, clearly, and two games of evidence say Texas can make Smith a decoy and still lose to the rest of the offense. The chip stays with the Buckeyes because a 99th-percentile success-rate offense does not need its star; watch McCuin and the tight ends over the middle in the first quarter, because that is where the yards went last year.The numbers clearly favor Ohio State's passing game, and yet Texas has held Jeremiah Smith quiet twice and still lost. Ohio State keeps the edge because its offense wins without its star. Watch the passes to McCuin and the tight ends over the middle early; that is where the yards came from last year.
Colin Simmons vs. the Buckeye Tackles
Edge: Texas, slight
+13 Ohio Statecomputed gap. OSU havoc allowed 95th pctl vs TX front-seven havoc 82nd; the bar belongs to a line whose right tackle just got benched
Colin Simmons, Lance Jackson, Zina Umeozulu · Texas
The SEC's 2025 sack leader in his junior season, projected top-10 pick, with Jackson and Umeozulu rotating opposite. The front produced two sacks and three takeaways in the opener. Simmons in June: "There's business to be taken care of."
Phillip Daniels, Ian Moore, Austin Siereveld · Ohio State
Four of five 2025 starters back from a line that allowed 16 sacks and 95th-percentile havoc. Daniels was pulled after a missed assignment on the opening drive; Day is still deciding the five.
OSU · havoc allowed, 2025
11.2%
Texas · front-seven havoc, 2025
11.2%
OSU · sacks allowed, 2025
16
Texas · sacks made, 2025
39
So whatThe bar leans Ohio State because Sayin was almost never touched last year, and the chip leans Texas because the tackle who let him get touched is the one Day benched. Two Simmons pressures in the first quarter means Arthur Smith is keeping two tight ends in to block, which is the version of this offense Texas can live with.The numbers lean Ohio State because Sayin was barely touched last year, but the tackle who let him get touched last week is the one his coach benched, so we lean Texas. If Simmons is in the backfield twice in the first quarter, Ohio State will keep extra blockers in, and that is the version of this offense Texas can handle.
Bo Jackson vs. the Texas FrontEdge: Texas
35%
86%
A 1,090-yard back on a 35th-percentile rushing offense (the two are not a contradiction: he was the good part) against an 86th-percentile run defense that added Biles. The Buckeyes managed 35 first-half yards on 17 carries against Ball State before the 65- and 49-yard breakaways; Texas's job is the breakaways.
Buckeye OL vs. the Texas Front (Run Blocking)Edge: Texas, slight
55%
78%
Middling line yards (2.99) against a front that allowed 2.69 line yards a carry (78th percentile) and stuffed runs at a 76th-percentile rate in 2025. Slight because four returning starters usually improve, and because a benched tackle can unbench.
Sayin's Deep Game vs. Texas CoverageEdge: Texas, slight
14%
71%
The most surprising bar on the page (−57 for Ohio State): the country's most efficient passing game was also one of its least explosive (1.42, 14th percentile), and Texas allowed 1.48 (71st). Arthur Smith's opener had seven 20-yard plays, so the chip is only slight; if Smith gets behind Mascoe once, treat the bar as expired.
The margins
Special Teams & Field PositionEdge: Texas
85%
46%
Texas started drives at the 85th-percentile spot last year, Ohio State at the 46th, and the Buckeyes' kick returner twice failed to take a knee in the opener. In a two-point game the hidden yards are not hidden; the kickers are a wash (83% vs. 80%).
Coaching & LuckEdge: Even
No computed bar, no honest metric for it. Day is 2-0 against Sarkisian in the last 20 months; Arthur Smith is calling his first big game for Ohio State against a Muschamp who, in Day's words, "really hasn't called the game in a while." The luck ledger (turnover watch above): Texas's +14 is due to shrink, Ohio State's +3 is not.
Pairing percentiles are computed by our metrics pipeline over the committed snapshots. Chips are this report's judgment; the bars can only be contextualized, not argued with.