Judo grip fighting: read an opponent's kumi-kata and plan the next contest
In this article
After a contest, write down the first grip your opponent took and what followed it. Which hand went where, sleeve or collar? Did they attack straight away, or settle in with two hands and wait? Much of what you can learn about an opponent shows in the exchange before the throw, and that exchange is easy to forget once the next contest starts.
Keep those notes on an opponent page so you can check them the next time the draw puts you together. Meeting the same opponent more than once is common on regional and national circuits, and it is worth giving your regular randori partner a page as well. When a rematch is coming, your coach can help you choose one or two changes from the page to test. Below are our suggestions for what to look for and write down.
What should you watch in the grip exchange?
Start with stance. Does the opponent fight right- or left-handed, and which way do you fight? Right against right or left against left (ai-yotsu) is a different grip fight from right against left (kenka-yotsu), because the pairing changes which grips come naturally and which ones you have to fight for. The pairing itself changes if either of you switches stance, so record it for each contest rather than as a permanent label.
Then watch the first grip. Which hand reaches first, and where does it go?
- to your sleeve
- to your lapel
- high on the collar
- over the back
- across your body in a cross grip
- to the belt
Some judoka attack off one grip; others wait until both hands are in place. Note how long they keep their grips before they attack, and what they do when you break a grip — go straight back for the same one, switch to something else, or step away and reset.
Watch what they do to your grips as well, for example whether they strip your sleeve grip as soon as you take it. Keep that separate from your own mistakes. If you kept reaching for the lapel first and lost the sleeve each time, that is a note about you.
Which attacks follow the grip?
Record the throws you actually saw. An opponent who only attacks forward to their left needs different preparation from one who mixes a forward throw with a backward one to their right. For each throw, note:
- Direction: forward or backward, and to which side
- Setup: a feint, a pull, a step, a combination
- Attack side: whether the attack came on the same side as the grip
The research further down shows why that last point is worth recording.
Keep watching after the attack. Did a failed throw end with the opponent in turtle, waiting for the referee? Or did they stand up and re-grip at once, or look for a turnover, a hold-down, a strangle or an armlock?
Finally, look at penalties and pace. Note any shido given to either of you and the reason as you understood it, and what changed late in the contest or in golden score if it went there. There is no need to look up the rule — rules change, and your notes only need what happened.
What do judo studies show about gripping and attacking?
Javier Courel-Ibáñez of the University of Granada, with colleagues at Granada, the University of São Paulo and the University of Niš, analysed 242 contests from 12 World Ranking tournaments in the International Journal of Performance Analysis in Sport (2014). Attacking on the same side as one’s own grip was associated with a higher chance of scoring and of winning, independent of sex and weight category, and same-side attacks in kenka-yotsu were the most effective configuration, especially in the lightest categories.
This was systematic observation of elite judoka, not an experiment. For your notes, the useful questions are:
- which side the opponent grips on
- which side they attack to
- whether the two match
Daniel Brandão Kashiwagura and Emerson Franchini of the University of São Paulo mapped 41 studies of kumi-kata in a scoping review in Revista de Artes Marciales Asiáticas (2022). Across that literature, collar-and-sleeve was the most used grip and kenka-yotsu the most common pairing. More experienced athletes held the judogi for less time before attacking; men, especially in the heavier categories, held longer and used more defensive collar-and-collar grips.
Those are baselines from mostly elite studies. The note worth keeping is how a particular opponent departs from them.
An older study by Franchini and colleagues in Perceptual and Motor Skills (2008) compared 17 super-elite medallists, the best in each weight category, with 16 other medallists from World Championships and Olympic Games between 1995 and 2001. The super-elite group scored with more different throws, applied in more directions, and that variety correlated with wins. It is an observational study of the very top of the sport, from before several rule changes, but it is a reason to record direction and not only the name of a throw.
A systematic review by Suzan Schoof, Froukje Sliedrecht and Marije Elferink-Gemser at the University of Groningen, published in the International Journal of Sports Science & Coaching (2024), found strong evidence across 23 studies that grip-fighting characteristics separate judoka of different performance levels, alongside the ability to throw and to vary technique. It was written for talent identification, draws on observational studies and says nothing about which grip to take. It is a reason to watch grip fighting closely, not a recipe for it.
Which notes are worth keeping?
After each contest, we recommend that the note answers three questions: what worked, what did not work and what you will try next time. In our view nothing else you could write down matters as much, in judo or any other combat or racket sport. Context and the opponent come first because the answers depend on them. Whether keeping notes improves judo results was not something the studies above set out to test.
- Context. Event and round, or randori; the opponent; category; the stance pairing.
- The opponent. Which grip they took first and with which hand; how they answered your grips; whether they attacked off one hand or waited for two, and how long they held their grips before attacking; throws and their direction, setups and whether the attack side matched the grip side; what they did after a failed attack and what they looked for on the ground.
- What worked. Grips, attacks or defences that worked, even if only once.
- What didn’t work. Where you lost the grip exchange, your penalties and the moments you were passive.
- What to do next time. One or two adjustments, each with a sign that would tell you to abandon it.
No rule sets the number, but one or two changes are usually enough for a single contest, and anything beyond that tends to overcomplicate the plan before you even step onto the mat. Anything you have forgotten can stay blank. Any mark on a mat sketch should say what happened at that spot.
How might one contest’s notes shape the next plan?
The note below is made up for teaching; nothing in it comes from a real judoka’s record or from research data:
Context: Regional open, my second contest of the day, against Kasper. We have met once before. He fights left-handed, I fight right-handed, so kenka-yotsu.
The opponent: Right hand to my right sleeve first, every time. Then left hand high on my collar, over the top. He attacked soon after the second hand landed. When I broke the collar grip he went straight back for it. Mostly left uchi-mata, forward; I did not count. Ko-uchi-gari when I bent forward to defend. Both to his left, the same side as his grip. After the uchi-mata failed he dropped to turtle and waited; the referee stood us up.
What worked: My o-uchi-gari off a single lapel grip nearly scored once, so my attacks were not hopeless.
What didn’t work: I kept going for the lapel first and lost the sleeve every time. Shido for passivity, I think, while I was trying to re-grip.
Next time: Get his right sleeve before he gets mine, then attack early off that first grip. Drop it if he starts taking the collar first, or if my attacks still come too late to matter. When he drops to turtle, try the turnover we have been drilling. If he stands straight up or the referee stops it quickly, save the energy for the grip fight.
The useful detail is the sequence: sleeve first, collar second, attack soon after, both attacks to his grip side. Exchanges recalled from one contest are not enough to say how often that will happen again. Keep the o-uchi-gari that nearly scored in the note; leaving it out would make the contest look more one-sided than it was.
What do you do with the notes before the next contest?
Take the changes you picked to your coach and turn each into a short cue. When you meet again, the first grip exchanges will show whether your opponent still fights that way. Since your last meeting they may have changed stance, category or first grip — and your own judo will have moved on too.
After the contest, note what went well and what failed, and where you are still unsure. Get it down soon, while you can still piece the exchanges together, and include a number only if you counted it. Our article on match memory looks at how far you can trust recall, and it sets no deadline by which a contest is forgotten.
Compare pages across opponents. If you keep losing the sleeve to left-handers, that is a question for randori and uchi-komi rather than for the next draw — ask a partner to take that grip and work from there.
For the evidence on reviewing your own contests, see the research on reflection and sporting performance. It cannot tell you whether putting things on paper is what causes improvement. If your ne-waza notes are growing, the BJJ training-notes article covers recording sequences on the ground.
In OpponentBook’s free journal, judo has a mat drawing where you can place pins, and named positions for standing grips, kuzushi, the main hold-downs and turtle defence, so a note can sit on the position it belongs to.
Synced records are saved in the storage you connect; OpponentBook’s account database does not hold a copy of your journal. When you request AI processing, the necessary content passes through an OpponentBook server to the AI provider. Optional end-to-end encryption protects encrypted records from the storage provider, while photos, videos and provider-visible file metadata remain outside that protection. The privacy guide explains these distinctions.
Pick the last contest or hard randori you can remember well. Write down the first grip the other judoka took, which hand it was and what came next. Finish with your three answers: what worked, what didn’t, and one or two things to try before your next contest against them.
