Brendan Allen has built his UFC run on wearing opponents down, finding the scramble, and finishing from the back. Christian Leroy Duncan has built his on standing in the pocket and cracking people. When those two profiles meet on October 10, the fight comes down to one question: does Allen get the fight to the mat before Duncan's hands and feet do the damage they've done to everyone else who tried to grapple with him?
Brendan Allen

Allen's game plan rarely changes. He closes distance, grinds in the clinch, and looks for the scramble that lets him climb to a dominant position. Once he's on a back, his rear naked choke has been the finishing touch more than once, and nothing on tape suggests an opponent has made him pay for hunting it. The problem is getting there. Every loss that matters here came against a fighter who kept him on the end of strikes instead of letting him into the clinch. Sean Strickland stopped him with punches, Chris Curtis hurt him with knees, and Nassourdine Imavov beat him clearly on the feet while stuffing nine of his ten takedown attempts. The Curtis rematch tape is the clearest tell: when Allen backs straight up along the fence, he gets caught by body hooks, and he doesn't have an answer for a mobile striker who refuses to let him set the terms.
Christian Leroy Duncan

Duncan is a switch-stance kickboxer with real pop in both stances, a four-inch reach advantage, and the kind of flying knees and clinch elbows that end fights early. From range, he's been excellent: he outstruck Roman Dolidze 54 to 10 and Jared Cannonier 68 to 15 in wins where he simply had the better toolkit on the feet. The hole is just as real. Grapplers have owned him once they get him down. Michal Oleksiejczuk, Rodrigues, Dolidze, and Cannonier have all logged heavy control time on him, up to 500 seconds in a loss to Rodrigues, with 427 and 372 seconds in his last two fights despite winning both. Once he's there, he goes passive on his back instead of working to return to his feet. He has also never had to fight past the third round in the UFC, so there's no tape of him finishing a fifth round against sustained pressure.
How the fight plays out
This fight is a straight collision of Allen's path against Duncan's path, and the tape says Duncan's path opens first. Allen needs takedowns or scrambles to get where he wants to go, and the last time he faced a rangy, live-handed striker who didn't want to clinch, he couldn't get there: Imavov stuffed nine of ten attempts and beat him up on the outside the rest of the way. Duncan doesn't need to change anything that's working. He can sit behind his reach, pick his spots with kicks and the occasional flying knee, and make Allen walk through real power to find the clinch, the same shape that has hurt Allen against every durable striker he's faced.
Duncan's defensive hole is the honest counter. If Allen does get a hand on him and drags this into deep water, the control numbers against Rodrigues, Dolidze, and Cannonier say Duncan will go passive and could get stuck there for a long stretch, maybe long enough for Allen to hunt the choke he's used to close out fights. But getting there is the hard part, and it's the part Allen has failed to solve against exactly this archetype. The fight is won or lost on whether Allen can close distance against a fighter who has made a living punishing people who try. Until he shows a new answer for that problem, the pick goes to Duncan.
The verdict
The pick is Christian Leroy Duncan. Brendan Allen's path runs through takedowns and scrambles that have repeatedly failed him against rangy strikers with real power, and Duncan is the sharpest version of that problem he's faced yet. Duncan's own weakness, going passive once a grappler gets on top of him, is real, but it only matters if Allen can first solve the distance problem that has beaten him before.
- Event · UFC Fight Night: Allen vs. Duncan · 2026-10-10 · Saturday, October 10, 2026 · 5:00 PM ET
- Market · Christian Leroy Duncan +104 — recorded when the pick was posted
- Model · Bones, Botcappers' AI UFC / MMA model — how it is built and what it does not know
- Record · every pick is timestamped before the event and graded in public, win or lose
- Statistics · the model's own match/game database, current to the posting time above. Figures a reader cannot open here are internal to that database, not published third-party data.

