Texas Tech 28, Houston 26: A Two-Point Try Decided Everything

No. 13 Texas Tech beat Houston 28-26 on Friday, September 18, 2026, when Conner Weigman's two-point conversion came up short at the end. J'Koby Williams and Cameron Dickey ran for second-half touchdowns for the defending Big 12 champions. Our live card returned 28.9 units on the night: $39,782 wagered returned $42,674.77. The whole game settled on one coaching decision worth studying.

Friday nights in college football are small slates with big prices, and this one ended the way live bettors dream about and dread in equal measure.

What Happened

Texas Tech built its lead on the ground in the second half. J'Koby Williams and Cameron Dickey both found the end zone rushing, and the defending Big 12 champion turned a competitive game into a two-score one.

Houston came back. Late, they scored to make it 28-26, and instead of kicking the extra point to tie, they went for two. Conner Weigman came up just short.

Game over. Two-point margin. Every market that hinged on that play settled in an instant.

Why the Two-Point Decision Matters More Than It Looks

This is the part worth your time.

A live model prices the endgame from a probability tree: Houston scores, Houston converts or does not, overtime or not. It assigns a probability to the conversion, typically somewhere near a coin flip, and prices everything downstream from it.

What the model prices poorly is the decision. Going for two rather than kicking is a coaching choice driven by factors no in-game model reads well: how tired the defense is, whether the coach trusts overtime, what the kicking situation looks like, what the week's narrative has been.

By the time Houston lined up, the live market had already collapsed to a near-binary. Either the conversion lands and this is a different game, or it does not and everything settles instantly. There is almost no edge left at that point, because the price has become the probability of one play.

The edge was earlier. It was in the third quarter, while Texas Tech was running the ball and the live number had not yet accepted that this had become a ground-control game rather than the shootout the pregame total implied. That is where the money was, and that is where our night was made.

Why Coaches Go For Two More Often Now

The decision Houston made is more common than it was five years ago, and understanding why helps you price these endings before they arrive.

Analytics departments have pushed coaches toward the aggressive option in specific spots. Down two late with a defense that has been on the field a long time, the win probability of a successful conversion frequently beats the win probability of reaching overtime, particularly for the road team. Overtime is close to a coin flip at best, and often worse than that for the visitor.

So the play was defensible. It just did not work.

What matters for a bettor is that this shift has changed the distribution of final margins in college football. Two-point attempts create more one-point and two-point finals and fewer games reaching overtime. Any total or margin market built on older data is working from a distribution that has quietly moved.

The practical version: when a game is heading for a late score that sets up a two-point decision, the market is usually still pricing overtime as more likely than it now is. That is a narrow, repeatable spot, and it is the kind of structural edge worth more than a read on any single game. It shows up across college football picks all season, not just on Friday nights.

The Documented Day

Note the handle. Friday was a small slate, so the card was smaller than a Sunday. A quiet night with a clean result is the shape of a good Friday, not a heroic one. Every day is on the results page in this same format.

What This Game Teaches About Football Handicapping

Game script changes are the most valuable live signal there is. A team that decides to run the ball changes everything downstream: the clock, the number of possessions left, the variance of the remaining game. Live totals adjust for the score but adjust slowly for pace. When a game converts from a passing game to a rushing game, there is usually a window before the number reflects it. That window is where live college football picks against the spread actually get made.

Endgame markets are the worst place to look for value. Everyone watches the last two minutes, so everyone prices them. The two-point conversion market in this game was efficient by the time it existed, because thousands of people were staring at exactly that number. Markets are efficient in proportion to the attention they receive, which is the single most useful idea in this whole business.

Friday night college football is underrated by bettors and overrated by nobody. The slate is small, the games are often mid-major or Big 12, and the sharp money that floods Saturday is not there yet. Fewer eyes means slower corrections. Anyone selling real football betting picks should be working nights like this harder than Saturday afternoons, and most are not, because Saturday is where the audience is.

That last point is the practical one for evaluating sports handicappers: look at whether their card matches where the edges actually are or where the attention is. A service that only posts marquee games is selling entertainment.

The Big 12 Context

Texas Tech entered as the defending Big 12 champion, and that label does specific work inside a pricing model.

A conference title from last season raises a team's rating going into this one, and in September that rating is carrying most of the weight because current-year data barely exists. Houston hanging within a two-point conversion of the defending champion is the kind of result that quietly tells you the gap between those programs is smaller than the number implied.

For live betting the practical consequence is straightforward: when a highly-rated team is struggling to separate from a lesser-rated one early in the season, the in-game price keeps expecting the separation to arrive because the pregame rating says it should. That expectation is often the mispricing. It is the same mechanism that produced the following Saturday's upsets and the Sunday after that.

What This Doesn't Prove

A 28.9 unit Friday is a good night, not evidence of a system.

Nothing about this result says the read was better than the alternative. A single Houston conversion and this article reads very differently while the underlying decisions were identical. That is the nature of games decided by one play, and it cuts both ways across a season.

The night before this one, we lost 23.9 units. That day is published too, in the same place and the same format. Past results do not guarantee future performance.

Where the Week Went

Friday was the second day of a five-day stretch that ran Thursday to Monday and finished at +66.4 units, with two losing days inside it. Saturday was the worst day of the season. Sunday was the best.

The full card is on best bets today and the season ledger is on the results page.

Frequently Asked Questions

What was the final score of Texas Tech vs Houston on September 18, 2026?

Texas Tech won 28-26 on Friday night. Houston scored late and attempted a two-point conversion to tie rather than kicking the extra point, and Conner Weigman came up short.

Why did Houston go for two instead of kicking the extra point?

The decision was Houston's coaching staff choosing to win or lose in regulation rather than play overtime. Those calls turn on factors like defensive fatigue and confidence in the kicking game, which is exactly why automated models price them poorly.

Who scored for Texas Tech?

J'Koby Williams and Cameron Dickey both ran for touchdowns in the second half as the defending Big 12 champions built their lead on the ground.

Where was the betting value in this game?

In the third quarter, when the game had become a rushing contest but the live total still reflected the pregame expectation of a higher-scoring game. Pace changes move slower through a live number than score changes do.

Are late-game betting markets good value?

Usually not. The final two minutes attract the most attention of any part of the game, and attention is what makes a market efficient. The better spots come earlier, when fewer people are watching a number that has stopped being right.

Is Friday night college football worth betting?

It can be better than Saturday for exactly the reason most people ignore it. Smaller slates draw less sharp money, so numbers correct more slowly. The trade-off is fewer games to choose from.

How do I get alerts when these spots appear?

Alerts go out by Email and Discord the moment a mispriced live number appears, with the side, the price at dispatch and the unit size. Packages start at $39.95 a month on the packages page, or reserve a free live pick to see one first.