Allen vs Duncan breakdown: How Brendan Allen won in round five

The version of Saturday's main event that will stick is the rescue. Christian Leroy Duncan out-landed Brendan Allen 106 to 51 in significant strikes, won that count in four of the five rounds, and Allen found a submission at 2:46 of the fifth with the night running away from him. I told it roughly that way myself in Sunday's recap, where I wrote that without the finish two judges were probably looking at a fourth round that was Duncan's by a mile.

The fourth round was Duncan's by a mile. The fight was not. Chris Lee and Sal D'Amato, the two judges whose cards we hold, both had Allen ahead 38-37 when the fifth round started. The submission did not save a losing night. It closed out a winning one early.

So let's take a look at how a man who lost the striking by more than two to one was still in front, and why the one round he gave away clean was the one where he stopped doing his job.

Round by round, the cards included

Here is the sheet from our stats for the UFC Fight Night: Allen vs Duncan main event, Allen first in every column.

Two rounds each, and Allen led because one of his two was a 10-8. That is the whole story of the scorecards, and it lives in a round Duncan won on significant strikes.

Round two: the 10-8 nobody saw on the strike count

Duncan landed 14 significant strikes in the second round and Allen landed 10. If the significant-strike number is the one you trust, and it is the one a broadcast puts on screen, Duncan edged the round.

Look one column to the right. Allen landed 33 total strikes on 47 attempts in that round, which means 23 of them did not count as significant: the short shots from top position that the stat sheet files under "landed" and then ignores. He took Duncan down on one of two tries, held him for 2:49 of the five minutes, and attempted a submission. Duncan had no control time at all in the round and no submission attempt.

Lee and D'Amato both went 10-8 for Allen. The criteria the commissions work from for a 10-8 talk about impact, dominance and duration, and duration is the one that shows up cleanly on a stat sheet: Allen spent more than half the round on top of a man who never got on top of him. You can argue about whether that is a 10-8 round. You cannot argue that it was Duncan's round, which is roughly what the strike count implies.

This is the myth I want to bury. Significant strikes are a count of what is easy to see. Rounds are scored on more than that, and this fight was decided in the gap.

Round three: five minutes on the mat, both ways

The third round is the strangest line on the sheet. Each man landed 6 significant strikes on 6 attempts, a perfect night of accuracy that tells you nobody was throwing from range. Allen held 2:57 of control, Duncan 1:57. Add them up and somebody was on top for 4:54 of the 5:00.

Allen got his takedown (one of three) and each man reversed the other once. That reversal matters more than its single digit. In his two previous fights, Roman Dolidze and Jared Cannonier took Duncan down eight times between them, and he won both of those fights by getting back up and out-striking them. On Saturday he did something different: he scrapped his way on top. Duncan banked 3:51 of control across the fight without landing a single takedown, all of it in the first and third rounds.

So Duncan's ground game was not a weakness Allen simply walked through. Allen won the third round 10-9 on both cards by finishing it with more of the top time, 22 total strikes to 13. It was close, and it was the round that put him ahead.

Round four: the round without a shot

Then Allen stopped wrestling. Zero takedown attempts in the fourth, the only round of the fight without one, and Duncan took the round apart: 46 significant strikes landed on 89 attempts against Allen's 17 of 43. Both judges gave it to Duncan, 10-9, which on that output is the most generous score Allen could have hoped for.

This is not how Allen usually handles a fourth round. He had been past fifteen minutes twice before in the UFC. In the Chris Curtis rematch in April 2024 he landed two of three takedowns in the fourth. Against Reinier de Ridder in October 2025 he landed his only attempt in the fourth and held 4:56 of control in a five-minute round, after which the fight was stopped. Three of four takedowns in fourth rounds, and then none attempted on Saturday.

I can't tell you why from a stat sheet. By the end of the third he had shot nine times and landed three, and six failed shots against a 6'2" man with a 79-inch reach is a lot of energy spent for very little. Maybe it was a plan, maybe it was his legs, maybe Duncan's distance simply would not let him in. What the numbers do show is the cost: the one round Allen spent standing was the one round he lost badly.

Duncan did not fade

Here is the second thing worth correcting. None of Duncan's ten previous UFC fights had gone past the third round, and the easy story about a first five-rounder is that the man without the miles runs out of them.

He did not. Duncan's best striking round was the fourth, 46 significant strikes, and in the 2:46 of the fifth he out-landed Allen 16 to 3. Across the two championship rounds he landed 62 significant strikes to Allen's 20. Over the whole fight he landed 56 percent of what he threw, and at distance he out-landed Allen 87 to 32. The striking was everything the matchup said it would be, for all 22 minutes and 46 seconds.

What ended the fight was not Duncan's gas tank. It was the next takedown.

Round five: one shot, 82 seconds

Allen went back to it. Two attempts in the fifth, one landed, and from there 1:22 of control and his second submission attempt of the night. Marc Goddard stopped it at 2:46. Of the 166 seconds the round lasted, Allen spent 82 of them in control, almost exactly half, and needed nothing else.

It is his first submission win since Paul Craig in November 2023, and the latest finish of his UFC career, later than the de Ridder stoppage after four. It also extends a run worth knowing about: Allen has now headlined five UFC cards and won all five, two of them by third-round submission (Andre Muniz and Craig), one on a five-round split decision, one after the fourth and now this one in the fifth. Every one of those main-event finishes came at the third round or later. He is a slow-cooking main-event fighter, and Saturday was the slowest yet.

What the pre-fight numbers missed

Our model gave Allen 37.1 percent going in, and the pre-fight breakdown walked through why: every factor group it uses pointed at Duncan, led by striking. The sportsbook consensus closed with Allen at -115 and Duncan at -105, about 51 percent for Allen once the margin is out.

Grade it honestly and the model read the striking perfectly. Duncan was the better striker on the night by every measure we keep. What it has no column for is how much of a fight one man spends on top of the other, and Allen spent 8:39 of the 22:46 there. That is 38 percent of the fight, in a sport that scores rounds on effective grappling as well as striking.

The full numbers are on the bout page. Allen is 28-7 and still No. 4 at middleweight on the last board before the fight; Duncan, No. 10 going in, is 15-3.

The principle is an old one in grappling and it held up again: position is the score, strikes are the commentary. Duncan won the count everyone could see. Allen won the rounds, and then he made sure nobody needed to read the cards.