# The Price of Control Source: https://fightstats.online/insights/the-price-of-control Published: 2026-08-09 > The least popular thing in mixed martial arts turns out to have a price list, and almost everything anyone argues about sits below the line where it starts costing anything. --- Ground control · 15,444 fighter-performances · men’s divisions The least popular thing in mixed martial arts turns out to have a price list, and almost everything anyone argues about sits below the line where it starts costing anything. - 49.8% one to three minutes of control - 82.7% six minutes or more Is holding someone on the mat actually worth anything? Below three minutes, no — a fighter with one to three minutes of control wins 49.8% of the time, and the interval around it runs 48.1 to 51.9. Above three minutes it becomes one of the strongest signals in the sport. The crowd is booing something that is either decisive or worthless, and almost never in between. The ladder ### Nothing, nothing, nothing, then everything Control time behaves like the takedown does: flat across the range people argue about, then steep once it passes a threshold nobody talks about. The two middle bands are the story. Under a minute of control wins 50.4%; one to three minutes wins 49.8%. Both intervals contain fifty, which means neither can establish any advantage at all. A fighter who spends two and a half minutes on top has, as far as this data can tell, done nothing to the result. Men’s divisions only. Dot: measured win rate. Bar: the 95% interval. Scale runs 30% to 90%; the line is at 50%. Grey bars either cannot establish an advantage or establish a disadvantage. The top band is the one that stops arguments. A fighter with six minutes or more of control wins 82.7% of the time, on 1,627 performances, with an interval running 79.7 to 83.9. Very little else in this sport moves a result that far. And the bottom band is worth its own sentence. Having no control at all is not neutral: it wins 24.7%, emphatically the wrong side of even. Control is one of those measurements where the absence tells you as much as the presence. The objection, answered ### It wins fights nobody scored The standard complaint about control is that it is a scoring artefact — that lying on someone persuades judges without actually winning a fight. That is a testable claim, and it is wrong. Half a point apart, on intervals that overlap almost entirely. Men’s divisions only. Two independent samples, one of which contains no judging at all, land on 70.0% and 69.5%. The intervals are 68.2–71.7 and 66.9–72.1, so the true values need not be identical — but the point survives the coincidence entirely. Control predicts finishes as strongly as it predicts decisions. Whatever holding someone down does, it is not a trick played on the officials. It is the same thing in a fight nobody scored. The route ### There is one road to three minutes If three minutes is the threshold that matters, the practical question is how anyone reaches it. The answer is not subtle. - No takedowns landed 5.7% reach three minutes mean control 39s · 8,692 - One takedown 25.0% reach three minutes mean control 125s · 3,017 - Two or three 58.2% reach three minutes mean control 248s · 2,408 - Four or more 92.5% reach three minutes mean control 450s · 1,327 Men’s divisions only. Share of fighters reaching three minutes of control, by takedowns landed in the fight. Without a takedown, reaching the threshold is close to impossible — 5.7%, on 8,692 performances. One takedown gets a quarter of fighters there. Four or more gets nine in ten, at a mean of 445 seconds, which is most of a round and a half. This is why the two findings belong to each other. A single takedown, on its own, is worth almost nothing to a result. What it is worth is a 25.0% chance of reaching the only control threshold that pays — and the second and third takedown are what turn that into a coin flip and then a near-certainty. Every division, the same line Win rate with three or more minutes of control, by weight class. Every one of the eight clears fifty comfortably, and the spread between the highest and the lowest is just over seven points — on samples where that gap is well inside the noise. This is not a heavyweight trick or a lightweight one. Men’s divisions only. Bars scaled between 60% and 75%, not from zero — the differences here are small and the scale says so. The emblem ### Seven minutes and twenty-two seconds, every time No fighter with at least five UFC appearances averages more control than Georges St-Pierre: 442 seconds a fight, with a single-fight high of 1,231. That is not a style, it is a business model. Behind him it is a dead heat: Colby Covington on 441, Khabib Nurmagomedov on 437, Sean Sherk on 436, Chael Sonnen on 424. Four men inside six seconds a fight, and Pat Sabatini, who used to top this list, now eighth on 408. Every one of them sits above the six-minute band where the win rate is 82.7% — which is to say they have each found the one thing on this page that reliably works, and organised a career around it. Men and women ### The threshold holds in both Control is a positional fact rather than a physical one, so it ought to travel between two games that differ almost everywhere else. It does. #### Win rate past three minutes 3,878 performances — the sample everything above rests on. #### Just under three points behind 622 performances. The difference is 2.9 points with a range of −1.1 to 6.9 — consistent, and on a sample large enough to have found a real gap. Against men of comparable weight the two are closer still: men’s flyweight and bantamweight come out at 67.2%, a shade above the women’s 67.0%, a difference of 0.2 points with a range of −5.0 to 5.4. So there is no sex effect here and no weight effect either — unusually for this site, the same number turns up wherever you look. The charts above are men’s divisions because that is the deeper sample, not because the women’s game does anything different. What this changes ### Control is a threshold, not a currency You are not accumulating something valuable a second at a time. You are either past a line or you are not, and most of the effort spent near that line buys nothing. #### Holding position, not getting there The gap between two minutes of control and four is the whole effect: 49.8% against 60.6%, and 82.7% by six. If a session ends when the takedown lands, it ends before the part that matters. Drill the ride-out, the re-take after the scramble, and the third minute specifically. #### Takedowns are the only road Without a takedown, 5.7% of fighters reach three minutes of control. With four or more, 92.5% do. There is no separate path — a control-based plan is a takedown-volume plan wearing a different name, and should be trained as one. #### The two-minute grind One to three minutes of control wins exactly half the time. If a game plan produces a couple of minutes on top and no more, it has spent a round of energy to arrive precisely where it started. Either commit to the position or leave it. One caution before any of this becomes a drill. Control time is a judgement call — a statistician decides when it starts and stops — and it is only recorded from 1999 onward. More importantly, a fighter who is winning tends to accumulate control because they are winning, so some of this effect runs backwards through the number. The scorecard test rules out one explanation, not all of them: it shows control is not merely persuading judges, but it cannot show that control is what caused the win rather than a symptom of one. --- ## Data Structured claims: https://fightstats.online/insights/the-price-of-control/data/findings.json Underlying query results: - https://fightstats.online/insights/the-price-of-control/data/control-by-weight-class.json - https://fightstats.online/insights/the-price-of-control/data/control-is-not-a-scorecard-artefact.json - https://fightstats.online/insights/the-price-of-control/data/control-ladder.json - https://fightstats.online/insights/the-price-of-control/data/emblem-the-controllers.json - https://fightstats.online/insights/the-price-of-control/data/how-control-is-obtained.json - https://fightstats.online/insights/the-price-of-control/data/men-vs-women.json --- Analysis and prose licensed CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Underlying fight records sourced from ufcstats.com. Method: /data.