Seven of the 12 fights on the UFC's Allen vs. Duncan card ended by knockout or technical knockout Saturday. A week earlier, UFC 332 in Salt Lake City produced 11 in 14, the most on any UFC card in our records back to 1994.
Neither night was a fluke in isolation. Since July 1, the UFC's knockout rate has run well above any full calendar year since 2008, and the jump belongs almost entirely to knockouts. Submissions have not moved. The fights that used to go to the judges are ending on the feet instead.
The obvious explanations do not cover it. Matchmaking has become more lopsided, but not by enough to account for the change. Fighters have been paid more for finishes since January, yet the knockout rate held steady for five months after the money arrived. What changed in July is still an open question.
Eight years of 31%
From 2018 through 2025, the UFC staged 4,025 fights. About 31% ended in a knockout or technical knockout, counting doctor's stoppages, and the yearly figure never left a narrow band: 28.4% at the low end in 2024, 33.6% at the high end in 2021. Submissions took about 17%. Half the fights went to a decision.
The first half of 2026 looked the same. Across 262 fights from January through June, the knockout rate was 33.6%, with submissions at 17.2% and decisions at 48.9%.
Then it moved. From July 1 through UFC 332 on Oct. 3, there were 167 fights across 13 cards. Eighty ended in knockouts, or 47.9%. Submissions stayed at 17.4%. Decisions fell to 34.1%. Two of every three fights now end inside the distance. Add Saturday's card and the run stands at 87 knockouts in 179 fights.
| Period | Fights | KO/TKO | Submission | Decision | |---|---|---|---|---| | 2018-2025 | 4,025 | 31.3% | 17.4% | 50.0% | | Jan.-June 2026 | 262 | 33.6% | 17.2% | 48.9% | | July 1-Oct. 3, 2026 | 167 | 47.9% | 17.4% | 34.1% |
The rise is not confined to the heavy divisions. In non-title bantamweight bouts, the knockout rate went from five of 31 in the first half of the year to nine of 17 from July through UFC 332. Non-title women's strawweight bouts went from two of 13 to five of 10. Not every division followed: men's flyweight produced three knockouts in 11 fights, close to its long-run rate. Those division samples are small, and no single one of them proves anything. The pattern across most of them is the point.
The money arrived first
The UFC changed its bonus structure in January, at the start of its broadcast deal with Paramount+. MMA Junkie reported on Jan. 24 that, starting with UFC 324, the four standard fight-night bonuses would double from $50,000 to $100,000, and any fighter who won by knockout or submission without one of those awards would receive $25,000. MMA Junkie credited Sports Business Journal with first reporting the change.
That is a direct cash incentive to finish. It also did nothing to the knockout rate for five months. Two weeks into the new system, MMA Junkie counted 15 stoppages in the first three events of 2026, "about a 40 percent clip," and noted that the promotion "hasn't had to stress too much about those finish bonuses."
UFC CEO Dana White had predicted as much. After UFC 304 in 2024, MMA Junkie reported, White said he would never be pressured into raising bonuses again: "Upping them doesn't change anything" and "it doesn't make them fight any harder." The first half of 2026 supports him. The second half does not obviously contradict him either, because nothing about the bonus structure changed in July.
The mismatch argument
The loudest explanation after UFC 332 was matchmaking. In MMA Fighting's post-event staff roundtable, one panelist called the event "UFC Squash Lake City." Another wrote, "please don't confuse mismatches playing out as expected with actual card quality." A Sherdog opinion column on Oct. 9 went further: "It didn't look like a happy accident; it looked like the product of deliberate matchmaking." The same column described the card's 11 knockouts as a UFC record and noted seven consecutive first-round finishes. Our results data confirms the 11 knockouts and that nine of them came in the first round.
The critics have a point. We measured each fight by the pre-fight chance the betting market gave the favorite, with the bookmaker's margin removed. The average favorite since July has been priced at about 70%, against 65% from 2018 through 2025. Fights with a favorite at 75% or better made up 37% of priced bouts since July, against 17% from 2018 through 2025. Across all priced fights since July, favorites have won 73%, up from 67%. UFC 332 was among the most lopsided cards we have measured: more one-sided than 95% of the 294 cards from 2018 through 2025 with enough prices to compare.
But mismatches explain only a small slice of the knockout jump. Among priced fights since July, 51% ended in knockouts. If those fights had been matched the way cards were matched from 2018 through 2025, and each kind of fight kept its new knockout rate, the figure would still be about 49%. The more even fights are ending in knockouts more often too.
UFC 332 shows it. Five of its 11 knockouts came in fights where the favorite was priced at 67% or less, including Johnny Walker over Mick Parkin at about 53% and Imanol Rodriguez over Alden Coria at about 57%. In one of the five, Roman Kopylov, the underdog, stopped Ateba Gautier.
Card by card, the link barely exists. The two most lopsided cards since July, on Aug. 8 and Aug. 29, produced four knockouts in 12 fights and eight in 13. Saturday's card was the least lopsided of the priced cards in the run and still produced seven in 12. The Sept. 12 card, close to the period's average for mismatch, had two knockouts in 13 fights.
So the mismatch reading of UFC 332 is half right. The card was unusually lopsided. That does not explain why it, or the four months around it, produced so many knockouts.
What else could it be
Several explanations remain, and the numbers cannot separate them.
A lagged response to the money. Fighters sign bouts months in advance, and camps take time to change. A finish bonus introduced in January could plausibly show up in behavior by midsummer. The timing fits. Nothing in our data can show that fighters are actually taking more risks for the $25,000.
Refereeing and stoppage timing. A knockout rate can rise if referees step in sooner. There has been at least one public complaint. King Green said on Instagram, as MMA Fighting reported, that he wanted referee Herb Dean to let him continue at UFC 332: "I was still fine. I wanted to keep fighting." We have no reliable way to measure stoppage timing across hundreds of fights, so we cannot say whether this has changed.
Roster turnover. The share of fights involving a debutant rose from 20% in 2018 through 2025 to 25% since July, and heavyweight and light heavyweight bouts, where knockouts are always more common, rose from 15% of fights to 19%. Both shifts are small, and the knockout rate outside those two divisions also rose, from 29% to 45%. Accounting for debuts, division and five-round fights barely changes the size of the jump.
Chance. If the old 31% rate still held, a run of 80 knockouts in 167 fights would be very unlikely to happen by luck alone. But a streak picked out after the fact always looks more remarkable than it is, and four months is a small sample in a sport where single cards swing wildly. Runs like this often partly fade.
What to watch
The simplest test is time. If the knockout rate stays near 45% through the end of the year, something has changed in how UFC fights end. If it drifts back toward the low-to-mid 30s, the summer was mostly a streak.
Two other signs are worth tracking. Watch the close fights: if bouts priced near even keep ending in knockouts at the summer's rate, matchmaking is not the story. Watch the decisions as well. They have fallen from about half of all fights to about a third, and that is the number most likely to tell us whether fighters are fighting differently, or are only being matched differently.
*How we did this: Fight results from 2018 through Oct. 10, 2026, come from the UFC's official statistics site. Knockouts include technical knockouts and doctor's stoppages. Pre-fight favorite chances come from sportsbook prices taken before each card, with the bookmaker's margin removed, and were available for 147 of the 179 fights since July 1.*