Natalia Silva vs Wang Cong: Neither Fades, So Who Owns Round 4?

The standard line on a first five-round fight goes like this: somebody hits a wall in the fourth, and the fight belongs to whoever has the better gas tank. I have heard it about every new headliner for as long as I have watched this sport, and it is usually said with a straight face and no evidence.

On Saturday at UFC 332 in Salt Lake City, Natalia Silva and Wang Cong fight for the vacant women's flyweight title over 25 minutes. Neither has gone past 15 in the UFC. So I pulled every round of every UFC decision either of them has fought and lined them up by round number, looking for the wall. I did not find one. Both women get busier as fights go on. They just do it in completely different ways, and that difference is the fight.

The curve

Silva has six UFC decisions. Wang has four, and one of them (more on this in a minute) was a grappling fight that tells you nothing about her striking curve, so the Wang columns below use her three stand-up decisions: Bruna Brasil, Ariane Lipski da Silva and Tracy Cortez. All numbers are significant strikes per round, averaged.

Read across the rows and both landed columns go up every round. Read down the absorbed columns and the two fighters split apart. That split is what I want to talk about.

Silva: the margin is a third-round thing

Silva's output climbs from 20.8 to 28.5, which is a 37 percent jump from the first round to the third. Her accuracy climbs with it: 39.9 percent of her significant strikes land in first rounds, 44.3 percent in seconds, 47.4 percent in thirds. She is throwing more and missing less at the same time, which is not what a tired fighter does.

The part that gets less attention is the other column. Her opponents land 13.7 in the first round, 12.0 in the second and 11.2 in the third. Silva's per-round margin goes +7.2, +11.5, +17.3. The third round is her best round of the fight in four of her six decisions, and against Jessica Andrade in September 2024 it was 47 of 65, 72 percent, the most she has landed in any UFC round.

My read of that, and it is a read, is a fighter who spends the first five minutes collecting information. She lands less early while she finds out what comes back, and by the third she has mostly stopped getting hit by it. That is a reading pattern, and reading patterns do not run out of gas. If anything, they want more time.

Wang: a shootout that grows on both ends

Wang starts where Silva finishes and keeps going. In those three stand-up decisions she lands 39.0 in the first round, 43.0 in the second and 45.7 in the third. Against Lipski da Silva in June 2025 her third round was 55 of 93. Those three fights contain seven rounds of 38 or more landed. Silva has two such rounds in her entire UFC career, 41 against Jasmine Jasudavicius and that 47 against Andrade.

But her accuracy runs the other way: 66.1 percent in first rounds, 61.7 in seconds, 59.1 in thirds. She is throwing more, from 59.0 attempts a round to 77.3, and a bit more of it misses as the fight goes on. And her opponents get busier too. They land 16.0 on her in the first round and 21.7 in the third. Lipski da Silva landed 27 of 76 in that third round, and Cortez landed 27 in the second.

So Wang's margin does not grow, it holds: +23.0, +25.3, +24.0 a round. Her fights get louder on both sides as they go. That is a pressure fighter's curve, and it is a good one, because the margin is huge and it never shrinks. It also means a fourth and fifth round are two more rounds of trading, and the trading has been getting more even as fights age, not less.

Under a grappler

Photo: Andy Grahn / CC BY-SA 4.0

Here is the part I did not expect. Each woman's quietest round in her recent UFC decisions came on her back, under a grappler, and each answered it the same way.

Silva's was against Rose Namajunas at UFC 324 in January. Round two: Namajunas landed a takedown, held her for 2:09, and Silva landed 8 of 27, tied for the fewest she has landed in a round of any UFC decision. Round three: Silva landed 28 of 62, took Namajunas down on her only attempt and kept her there for 47 seconds. Namajunas landed 8. The round it ties is the first against Viviane Araujo in February 2024, when Araujo held her for 3:31; Silva went 12 and 18 in the two rounds after it.

Wang's was against Eduarda Moura in February, the decision I left out of the table. Moura took her down in the second round and held her for 4:31 of the five minutes. Wang landed 3 significant strikes in the round. In the third, Moura shot six times and landed none of them, and Wang controlled her for 2:11 of her own. Across their UFC decisions, opponents have landed 4 of 26 takedown attempts on Wang and 2 of 23 on Silva.

The lesson cuts both ways. Neither woman has been worn down by being held, and both came back from a lost round with their own control, which is about as good a sign for rounds four and five as a three-round sample can give. But it is also the only way either has been slowed, and neither one is the wrestler in this matchup. Silva has landed three takedowns in eight UFC fights, Wang two in six. If Saturday stays standing, the bad-second-round scenario mostly stays in the archive.

There is one more data point on Wang, and it is the uncomfortable one. Her only loss, to Gabriella Fernandes in November 2024, came by submission at 3:49 of the second round. She was ahead 46 to 18 in significant strikes when it ended. Output was never her problem in that fight. Position was.

What rounds four and five ask

Put the two curves next to each other and you get a clean question. Wang lands more in every round, by a lot: her lowest stand-up round average is higher than Silva's highest. Silva gets hit less in every round, and gets hit less the longer the fight goes, while Wang gets hit more.

So rounds four and five are a test of which curve is the real one. If Wang's pressure keeps climbing the way it does from round one to round three, Silva has to answer a volume she has never seen across a full fight, from a woman who outlands her opponents by more than 20 a round. If Silva's reading keeps working the way it does, Wang's late rounds become what they were against Lipski da Silva and Cortez, louder on both ends, except this time against a fighter whose opponents land less every round, in exactly the rounds where Silva has been at her best.

I do not know which bends first, and nothing in either woman's file says it. Our model sits right on the line at 50.3 percent for Silva. The betting consensus has her at -210 across 18 sportsbooks, about 65 percent. The UFC 332 pick'em is 27 to 26 for Silva. The case for the market and the case for the model are laid out in why the model calls this a coin flip, and I will not rerun them here.

What I will say is this. The fourth-round wall everyone will mention on the broadcast does not show up anywhere in the numbers. These two have landed more in the third round than the first in eight of their ten UFC decisions. Whoever loses on Saturday probably will not lose because she got tired. She will lose because the other woman's third round turned out to be the one that kept going.