Gi vs No-Gi

BJJ STATISTICS

Which rule set gets a higher submission rate in Brazilian jiu jistu, gi or no-gi? There’s two ways to answer this: what the stats show us, and what actually matters.

What do the stats say?

I pulled the data for 65,100 adult (18+) BJJ matches held by NAGA and Grappling Industries in 2025 from Smoothcomp.

Let’s just pause for a moment on that: over 65k BJJ matches occurred last year in just those two promotions. Wow.

Also, I’d say that this would make any analysis pretty accurate…there’s not a lot of room for outliers to skew the results one way or the other.

When we look at all of those matches (24,876 gi, 40,224 no-gi), no-gi leads the way with 50.54% of them ending with a submission. Gi pulls in a respectable 48.74%.

Now I’m not a statistics wonk, in that I haven’t gone near anything resembling in-depth analytics since that one stat class I had to take when I was going for my Masters degree, so I wanted to find out if these numbers were significant.

I learned about and ran something called a “chi square test”. This is a statistical tool that can show whether the number of occurrences of an event (in this case, jiu jitsu matches won by submission) is influenced by different factors (in this case, the rule set of gi vs. no-gi).

(I won’t bore you with the details of showing you my work…if you’re interested in seeing that, let me know, and I’ll gladly share it. I’m sure there’s some overlap in the Venn Diagram of jiu jitsu practitioners and statistics nerds.)

Anyways, I ran this test, and lo and behold, it appears that the rule set (gi vs no-gi), from a statistical standpoint, actually does influence the result. In this case, the no-gi rules influence the submission rate in some way that results in a higher instance of submissions.

So from a purely statistical point of view, no-gi wins.

What actually matters

So no-gi is better, right?

Not necessarily. Let’s pump the brakes.

When we look at this in real, practical terms, this means that for every 100 matches of each, 2 more no-gi matches end in submission than gi matches.

Looked at from that perspective, the difference is almost negligible. It means that in reality, you aren’t likely to see that difference play out in a meaningful way at any of your competitions.

Why do no-gi matches have a higher submission rate?

Who knows?

For starters, we simply know there’s a statistical relationship here, not the mechanism that affects the outcome. We’ve established correlation (these two things are happening at the same time), not causation (thing A causes Thing B to happen).

The chi square test shows that there is something about the different rulesets that is influencing the outcome, but not exactly what that thing is. There are likely a lot of factors at play, but without more data or a deeper study, it would be difficult to nail down a specific reason or reasons that the stats work out the way they do.

But just for fun, let’s speculate.

Possible Explanation #1: there are more legal submissions allowed at more levels of no-gi matches than gi matches.

While in gi there are more controls (sleeves, lapels, pants) and some additional choke options not available in no-gi (namely lapel chokes), aside from straight ankle locks, most attacks below the waist are illegal in BJJ competitions, or at least limited to higher ranks.

In no-gi, however, just about everything below the waist is legal: heel hooks, calf crushes, straight ankle locks, toe holds, knee bars, chuckleberry handshakes…I’m sure there’s more.

Is this the reason, though? Tough to say. There would need to be data from a decent sampling size that looked at the prevalence of different submissions used in competition.

Possible Explanation #2: there are significantly more no-gi matches than gi matches.

That there are over 1.6 times as many no-gi matches (40,224) as there are gi matches (24,876) in this data set…that’s a significant difference. That could also explain why there are more submission endings; there are simply way more matches, and therefore way more opportunities for a match to end in a submission.

I tested this by looking at a random selection of no-gi matches, but matched the gi number. The submission rate actually went up to 50.69%. So this might mean that more matches actually reduces the number of submission endings.

But! When I looked at the NAGA and Grappling Industries matches from January and February of 2026, the numbers flipped: gi (3,084 matches) had a 53.44% submission finish rate, and no-gi (3,056 matches) had a 49.90% rate.

Weird.

At this point, I’m going to tentatively say that the number of matches doesn’t have influence over the results. It has to be something else with the rule set.

What this means for training for BJJ competition

In all honesty…probably not much. This is a case of the difference meaning so little in a meaningful way that you shouldn’t fundamentally alter your training.

If there were a much larger gap (say, a 10-15% difference), then I might say there would be reason to look at how you prepare for a competition. If gi were, hypothetically, ending in submission 40% of the time, that might tell me that either a) submission offense in that division isn’t great, or b) submission defense is better than average. I might tell the competitors I’m coaching that we need to focus more on scoring points in gi.

If, on the other hand, no-gi matches got submissions 60% of the time, that might tell me that there are a lot of submission hunters in no-gi, and I’d say that we really need to work on submission defense.

But in this case, with not only a very, very small practical difference between gi and no-gi submission rates, and with those rates sitting right in the middle of the pack (you essentially have a 50/50 chance of a match ending in a submission), it’s tough to prescribe a change in tactics.

And remember, this is just an average. We can never discount the value of individual performance, strengths, weaknesses, experience, talent, and so on. If you are a submission artist, nailing armbars from everywhere, then keep playing your game. If you tend to lose position every time you go for a submission, then focus on your point scoring game. (Or, I don’t know, get better at not losing position.)