Height · 4,043 measured bouts · 2015–2026
Rank every fighter by height inside his own division and the shortest thirty per cent shoot forty-two per cent more takedowns than the tallest. It holds in all eight men’s divisions. What it does not do is make them better at it — or, in the end, win them more fights.
The question
The received wisdom
A low centre of gravity wins the takedown battle. The short fighter gets underneath you, changes levels faster and finishes from angles a taller man cannot reach — so of course he is the one shooting.
TLDR; Quick Facts
3.02 — 2.12
takedown attempts per fight · shortest 30% — tallest 30%
Rank every man by height inside his own division. The shortest third shoot 42% more often than the tallest — a difference of 0.89 attempts a fight, on an interval from 0.54 to 1.25.
8 of 8
men’s divisions pointing the same way
From bantamweight, where it is 3.95 against 2.34, to heavyweight at 1.94 against 0.95. Not one division reverses it, so this is a property of the sport rather than of one weight class.
36.57% — 35.93%
takedown success rate · shortest — tallest
And here is what does not change. They are no better at it. The low centre of gravity does not convert — it just gets used more.
0.85 — 0.63
how far height separates them — how far reach does
It is not the arms either. Ranking by reach instead of height gives a weaker gradient. Whatever is happening is about hips, not limbs.
48.44% — 52.53%
win rate · shortest 30% — tallest 30%
The uncomfortable part. All that extra wrestling does not buy a better record. The shortest third win slightly less often than the tallest.
In short
Being short for your division changes how you fight far more than it changes whether you win. The shortest third of every division shoot far more takedowns, land more of them in raw terms, and convert at exactly the same rate as the tallest third — then finish the night with a marginally worse record. The low centre of gravity is not an advantage being exploited. It is a style being forced.
Every man ranked inside his own division
Take every fighter since 2015 and rank him by height against the division he competes in, so that the same man can be short at welterweight and unremarkable at flyweight. Then ask what the shortest third do differently from the tallest third — whoever they happen to be facing.
They shoot. The shortest 30% attempt 3.02 takedowns a fight against the tallest third’s 2.12 — a difference of 0.89 attempts [0.54, 1.25], or 42% more. That is one row per fighter rather than per fight, on 456 short men and 256 tall ones with at least three bouts in the division, because a busy wrestler with fourteen fights would otherwise vote fourteen times — and wrestlers are exactly the population in question.
| Height band | Fights | TD attempted | TD landed | Success | Win rate |
|---|---|---|---|---|---|
| shortest 30% | 3,565 | 2.89 | 1.06 | 36.57% | 48.44% |
| 30-50% | 1,625 | 2.56 | 0.93 | 36.46% | 49.54% |
| 50-70% | 2,006 | 2.51 | 0.9 | 35.89% | 50.6% |
| tallest 30% | 1,980 | 2.04 | 0.73 | 35.93% | 52.53% |
Four bands, and the attempt rate falls at every step. What does not fall is the success rate: 36.57% for the shortest against 35.93% for the tallest. The extra takedowns are extra attempts, converted at the same rate as everybody else’s.
The obvious objection
Heavy divisions are tall and wrestle less, so a gradient across the whole roster could be nothing but weight class wearing a disguise. Ranking inside each division should already prevent that, but the check is cheap and the failure would be total, so here is every division on its own.
| Division | Shortest 30% | Tallest 30% | Gap |
|---|---|---|---|
| Bantamweight | 3.95 | 2.34 | +1.61 |
| Featherweight | 3.16 | 2.1 | +1.06 |
| Flyweight | 3.4 | 2.56 | +0.84 |
| Heavyweight | 1.94 | 0.95 | +0.99 |
| Light Heavyweight | 2.08 | 1.63 | +0.45 |
| Lightweight | 2.7 | 2.12 | +0.58 |
| Middleweight | 2.7 | 1.97 | +0.73 |
| Welterweight | 2.76 | 2.14 | +0.62 |
Eight of eight. The size varies — bantamweight is the loudest and light heavyweight the quietest — but not one division reverses it, and the two extremes of the sport behave the same way as the middle of it.
And inside a single fight
Yes. And the bigger the height gap, the more lopsided it gets. Comparing the two men inside the same fight — same weight class, same night, same opponent, so nothing else can explain it — the shorter fighter wins the takedown count more often than the taller one at every gap worth measuring.
The white line is 50% — the height at which height tells you nothing. The narrowest gap sits on it: between men within four centimetres of each other, the interval runs 49.1 to 55.7 and cannot establish anything. From five centimetres upward it clears, and past ten it clears by a distance.
Only bouts where the two men actually landed different numbers of takedowns can answer the question, so those are the bouts counted. Fights where neither man landed one — a third of the sample — are silent on it, and including them would have dragged every figure below 50% by construction.
The mechanism
Here is where the gym wisdom breaks. If low hips and a low centre of gravity made the takedown easier to finish, the shorter man's success rate would be higher. It is not. At every height gap, the two men convert their attempts at almost exactly the same rate.
That pattern holds all the way up and down the range. At 5–9 centimetres the shorter man converts 38.4% against 35.2%; at fifteen and beyond, 40.2% against 43.3% — the taller man ahead, if anything. Nothing in these numbers says a short fighter finishes takedowns better. What they say is that he tries, and keeps trying.
Which makes sense the moment you stop thinking about leverage and start thinking about the alternative. The short fighter's other option is to stand in front of a longer man and trade. The takedown is not his talent. It is his escape.
The emblem
One hundred and sixty-eight centimetres, and in nine of his fights since 2015 he has been the shorter man — by six centimetres on average. In those nine he landed 68 takedowns. Nobody in the database has more from underneath.
He is the thesis with a face on it. Not a man who solved the geometry, a man who decided the geometry was irrelevant and shot anyway, for twenty-five minutes, against people built to stop it.
The same test, in the women’s divisions
Run the identical thing on the women’s divisions and it comes back the same. The shortest 30% attempt 2.94 takedowns a fight against the tallest third’s 2.02 — a gap of 0.93 [0.3, 1.55], which clears zero.
That is 46% more against the men’s 42%, on 92 short fighters and 62 tall ones rather than 456 and 256. The interval is roughly twice as wide for exactly that reason, and the two point estimates are well inside each other’s range. Two games that share very little else answer this question identically.
One loosening to declare. The women’s divisions are smaller — four of them against eight — so a division has to carry twelve fighters to be ranked here rather than the twenty used on the men’s side. That is a real relaxation of the standard, and the wider interval is where it shows up.
Extra takedown attempts per fight · shortest 30% against tallest 30%
Men’s divisions +0.89
Clears zero
Women’s divisions +0.93
Clears zero
← no differenceshort fighters shoot more →
One quantity: extra takedown attempts per fight. Both clear the line and the two intervals overlap almost entirely, so the honest reading is that this is the same effect in both games rather than a bigger one in either. The women’s bar is wider because it rests on 92 fighters against 456, not because the effect is less certain in kind.
What this means for your training
The shortest third of every division shoot 42% more takedowns and convert them at the same rate as everybody else. The extra work is going in. It is not coming back out as a better success rate, and it is not coming back out as wins.
Your finish rate against a taller man will be about 36% whatever you do, which is roughly what his is. The edge in this data comes entirely from attempt count, so the session that pays is the one that raises how many entries you can make in five minutes without emptying the tank — not the one that drills a cleaner double.
You are not facing someone who is better at taking you down. You are facing someone who will try roughly twice as often, and the danger is cumulative rather than technical: the third and fourth entries of a round are where the count gets away from you. Cage position and getting back to open space are worth more preparation time than sprawl mechanics.
The interval at 1–4 centimetres runs from 49.1 to 55.7 — it contains the no-effect line. A tale of the tape showing three centimetres tells you nothing about who wrestles, and any minute of game-planning spent on it is a minute taken from something that does.
And the flat line under all of it. The shortest third win 48.44% of their fights and the tallest third 52.53%, so whatever the low centre of gravity is worth, it is not visible in the result. That is the most useful thing on this page for anyone planning a camp: the style is being forced by the matchup, not chosen because it pays.
