In part 1 we saw the overall stats, but now we will take a closer look at all the stats for takedowns, subs, passes, etc in this article. The start of the year was a bit slow, but once things started we got a lot great matches. We saw some new events as well, most notably the ADCC Open events and the AIGA promotion. We also saw the long awaited return of Quintet. 2023 was definitely one of the best years in grappling, so let's take a look at the various stats and try to draw some conclusions from them.
So as you can see the RNC came out on top as always. This seems to be the #1 submission in all rulesets and is the only sub I'd recommend everyone to learn. Following this is the inside heelhook, the armbar and kimura respectively. I was a bit surprised about the kimura, but this year it seems to have been a good tool for finishing matches.
Now we are coming to sweep and here the sub sweep is on top. You might be asking, "what is a sub sweep?" This category are sweeps that are due to someone attempting a submission and it ends with them on top. This could be an omoplata that someone is forced to rollout of or a guillotine that forces a roll. From my memory however these were mostly due to leglocks. Anyway, these were followed by single and double legs from wrestle ups, as well as half guard sweeps.
Finally we have passes. Half guard passes came in as number one here, followed by bodylock passes, kneeslice and NS(North South) passes. I should also add that I combined the 50/50 arm pass and the duck under pass, as I feel they are just variations of the same pass. I think this will make things easier going forward.
The back was the best position to finish from and after it came the mount, 50/50,the saddle and finally side control. Knee on belly and the crucifix didn't have any finishes that I recorded this year.
Some of these were somewhat predictable, but some things this year I was a bit surprised about. Let's get into some of the more interesting ideas.
Let's take a quick look at the top submissions from last year and see if there are any changes compared to 2022.
The outside heelhook passed the armbar this year to achieve its 3rd place spot. This is a bit surprising seeing as about 3 years ago people were declaring it dead. Now it is coming back and might be something people need to take a second look at if they aren't using it. The ankle lock moved up as well. We have been seeing this more and gi BJJ, but now its popularity seems to be increasing in nogi as well. This seems to have caused everything else to move down compared to last year, but the kimura completely dropped off the list. It was replaced by the darce and toehold, which tied for the 10th spot.
As I said in the previous post, leglocks are becoming more prominent instead of less. 5 out of the top 10 subs are leglocks. It is also worth noting that it is no longer just heelhooks, but also other leg submissions such as anklelocks, kneebars and toeholds that made it into the top 10. It seems leglocks have finally transitioned from something only specialists do, to something that is practiced and used by almost everyone at the pro level.
2023 was a great year for grappling and hopefully with these new stats we can learn even more about what really goes on in the matches. I'm hoping 2024 continues this momentum and we get more good events. I'm looking forward to ADCC 2024 and I'll be covering it all next year, so stay tuned for that.
Grossest Match of the Year
Stats
Highest Percentage Subs(With over 5 finishes)
- RNC(41.8%)
- Armbar(31.8)
- Kimura(28.8%)
- Inside Heelhook(28.7)
- Outside Heelhook(15.7%)
There were 775 attempts and 138 submissions. That means submissions had an overall success rate of 17.8%.
Moving on to takedowns single leg came out on top again. The double leg was right behind it and then the mat return. Surprisingly the deashi barai has become quite popular and was used more than many other more popular takedowns.
Highest Percentage Takedowns(With over 5 finishes)
- Mat Return(70.8%)
- Bodylock(56%)
- Throw-by(46.6%)
- Kneetap(40%)
- Anklepick(35.7%)
There were 654 attempts and 207 takedowns. That means takedowns had an overall success rate of 31.6%.
Highest Percentage Sweeps(With over 5 finishes)
- Sub Sweep(69.8%)
- Double leg(53.8%)
- 50/50(43.7%)
- Berimbolo(42.8%)
- X-Guard(42.1%)
There were 470 attempts and 169 sweeps. That means sweeps had an overall success rate of 35.9%.
Highest Percentage Passes(With over 5 finishes)
- High Step(47%)
- 3/4 Mount(46.1%)
- Smash(41%)
- Float(37.5%)
- Bodylock(32.6%)
There were 772 attempts and 206 passes. That means passes had an overall success rate of 26.6%.
- Back(41.5%)
- Mount(11.2%)
- Side Control(5.7%)
- Saddle(17.6%)
- 50/50(14%)
- Knee on Belly(0%)
- Crucifix(0%)
Stat Analysis & Conjectures
Some of these were somewhat predictable, but some things this year I was a bit surprised about. Let's get into some of the more interesting ideas.
The back position is the strongest position in nogi BJJ by far. The fact that 41% of the time people got there they were able to finish their opponent, is amazing. On the other hand the mount was actually rather weak. It was achieved almost as often as the back, but the number of finishes in quite small comparatively. Both the saddle and 50/50 had higher finishing positions than the mount. Although the mount is seeing a renewed amount of popularity in nogi, it seems people haven't figured out how to convert it into submissions.
The best way to sweep someone is to attack a submission. Outside of omoplatas and occasionally inside heelhooks, this isn't really mentioned much, but it was by far the most effective way to sweep people in 2023. I think this should be a new area of study as there isn't much info on it. Chaining submissions into wrestle ups, forcing people to fall to ground to defend, or forcing scrambles could be a key to victory for guard players in the future.
The top vs bottom debate isn't going to be resolved this year. Sweeps were a higher percentage attack than both takedowns and passes in 2023. That means it is better to play from the bottom and sweep then right? Well maybe not. There were many more passes and takedowns than sweeps this year, so although sweeping seemed to be a higher percentage option, it didn't happen as much. This could be due to a lack to trying, in which case people should just attempt them more. However it could also be because the top players weren't giving them a chance to work. It looks like 2023 won't be the year we end this debate and for now both options seem pretty solid.
Above you can see the total submission stats for the year. You might be thinking, "why are there more subs here than above". There were a few events that I recorded the submission for, but not attempts, takedowns, passes, etc. This was mostly due to the high number of matches, and thus as you can see more many more submissions here. Due to the increased number of matches covered you might see slight differences in the most common subs, but the RNC always comes out on top.
Year to Date Submissions
- RNC
- Inside Heelhook
- Outside Heelhook
- Armbar
- Ankle lock
- Guillotine
- Triangle
- Kneebar
- Arm triangle
- Darce/Toehold
Submissions through the year
I'm a bit surprised the triangle is there, but it is a BJJ staple and if you look at this list there are a lot of traditional techniques here. The armbar, triangle, kimura, and arm triangle still all work at the highest levels, consistently. I think this proves it is important to not set these aside for other newer techniques, because they still work really well.
The ankle lock has made it into the top 5 this year, which is surprising considering how people felt about it even 3 years ago. Until about 2-3 years ago, anklelocks were thought of as moves that didn't work at the highest level, wouldn't really cause a serious break and people could just grit their teeth through without tapping. That seems to be false now, as it garnered more wins than many traditionally sound techniques such as the triangle. It now seems to be a proven technique that works from whitebelt all the way up to the highest levels of competition.
There were some things that might be slightly off in the stats such as the choi bar and the aoki lock. With the choi bar there might have been times I counted it as an armbar by accident. Due to the fact that they have the same finish, I'm not sure if I made a mistake and misattributed it to the armbar category instead. With the aoki lock, it is sometimes difficult to see what happened, so some of the ankle lock might have actually been aoki lock and I just couldn't see due to the camera angle.
There were a few surprised here for me as well. I thought there would be more Z-locks this year, but it didn't really happen. There were a lot of people talking about it going into 2023, but that didn't really result in finishes(or attempts). The same can be said of the choi bar. On the other hand smoothers, texas clover leafs and Americanas seem very high. I'm guessing the smoother was popularized by John Danaher and Gordon Ryan and so more people have been trying it. I don't know why there were so many Texas Cloverleaves. This is a sub that is rarely talked about or taught, but it was hit 7 times this year. The biggest surprise it the Americana. This is known as a move that only works on people who are terrible and isn't really a viable high level sub. There were 15 Americanas this year, which beats Estima locks, Anaconda chokes, North-South chokes, inverted armbars, omoplatas, aoki locks, calf slicers, etc. Now normally you could dismiss this saying, "Well it must have been used on low level atheletes", but we saw one being used to submit a North American ADCC trials winner(Big Dan). Perhaps the collective wisdom of the BJJ community needs to give a second look to the Americana.
Overall
Also check out part 1 is you haven't already.
What does the Shoulder crunch count as?
ReplyDeleteWhat does the Shoulder crunch count as? (in terms of a sweep)
DeleteIt will usually be a hook sweep.
DeleteWicked analysis. thank you for creating this!
ReplyDeleteThanks
DeleteHey, just an idea, but I think it would be super useful to know what moves people are hitting from common positions. For example, tracking what single-leg finishes and defences are hit from the single-leg position, or what attacks and escapes are hit from mount. Might be too much extra stuff to keep track of, but I think it would be super cool.
ReplyDeleteDon't suppose you know how much of those guillotine we're arm in?
ReplyDeleteSick analysis, thank you very much!
ReplyDeleteOne question, why do you seperate the Choi Bar from the Armbar at all? It's the same finish, just a different entry, right?
Thank you for reading. I did it initially to see how often it occurs. However I might get rid of it and fold it into the armbar category next year. I'm still working the kinks out.
Delete