Our model went into the main event of UFC Fight Night: Strickland vs. Hernandez on Feb. 21 with a HIGH-confidence pick on Anthony Hernandez, and the result went against it.
The night in brief
Sean Strickland beat Anthony Hernandez by knockout in the third round of the headliner in Houston. We had picked Hernandez at 76.0 percent, and we had the fight going to a decision.
The undercard supplied the early violence. Three fights ended in the first round by knockout: Uros Medic, Melquizael Costa and Jacobe Smith each finished the job before the opening five minutes were up. With Strickland's finish, four fights on the card ended by knockout.
Jordan Leavitt beat Yadier del Valle by decision over three rounds. That was our other flagged miss.
The model autopsy
We grade our own predictions in public, and this card was a middling night that two flagged misses dragged down.
The model went 9-for-14 on the full card, 64 percent. That is below our season line and below where we want to be, and the miss that matters most sits at the top of the card.
On the flagged picks, the calls we tag LOCK or HIGH because the model is most confident, the card went 4-for-6. Both misses were HIGH picks, and one of them was the main event.
The model picked Anthony Hernandez to beat Sean Strickland. It was wrong. Worse, it was a HIGH-confidence flag: the model had Hernandez at 76.0 percent. That is not a coin flip we lost. At 76.0 percent the model expected Strickland to win about one time in four, and he did, by knockout in the third round. One result cannot show whether that was bad luck or a misread of the matchup, so we log it as a miss and leave the explanation open.
The other flagged miss was Yadier del Valle at 72.5 percent, who lost a three-round decision to Leavitt.
We log the Hernandez pick as a wrong HIGH pick on the headliner, the most expensive kind of miss we track, because it is the fight most readers actually watched. Owning it is the entire point of doing this in the open.
The other side of the 4-for-6 flagged record is that the model still cleared two-thirds of its confident calls. Alden Coria, Carli Judice, Melquizael Costa and Joselyne Edwards all won as picked. Costa and Edwards won by finish where we had called a decision, but the winner was the right one.
One honest takeaway
A 64 percent card with a busted HIGH pick on top is a reminder of what our headline numbers do and do not promise.
As of June 27, the model sat at 68.8 percent on all live 2026 picks and 82.7 percent on 75 LOCK and HIGH flagged calls. The leak-free backtest lands near 70 percent overall and 84.3 percent on LOCK. Those are the canon numbers, and this card sits under both of them.
That is how a confidence band works. It is a rate, not a guarantee. A flagged record of roughly 83 percent means roughly one in six confident calls misses, and on Feb. 21 one of those misses was the main event. The Feb. 21 result does not move those season numbers on its own, and we are not blending it into a rosier figure to make the night look better than it was.
If the flagged hit rate is going to mean anything, the misses have to be reported with the same ink as the hits. This one was a 76.0 percent HIGH pick on the headliner that Strickland ended in the third round. We called it, we got it wrong, and the grade stands.