How to coach yourself in judo with video, slow motion and measurable criteria instead of gut feel — from rotation angles to grip-strength benchmarks.
Straight talk: beyond a certain level it's not more training that moves you forward but better feedback — and you deliver that to yourself most honestly by video. The point isn't the filming but what you measure against. Good judo technique has measurable criteria, not vague feelings. In the seoi-nage, for instance, a shoulder rotation of around 247° (elite) vs. 208° (amateur) separates the levels — an angle you can read frame by frame. Likewise the pulling force: 4.36 N/kg for the elite vs. 3.25 N/kg in the midfield.
The principle reaches beyond a single throw. Efficient movement follows a kinetic chain — force runs in order from the large, proximal segments to the small, distal ones, and this sequence is measurable. Even your grip strength is a number: around 43–61 kg per hand depending on weight class for competition judoka. Whoever measures against such anchors instead of "felt good" coaches like a pro. The best part: you need no lab for it, just a phone and a fixed viewing angle. At this level, video is your most honest trainer.
Before you film, decide what "good" means. In the seoi-nage, for example: shoulder rotation toward 247°, center of gravity forward-up first, then down, full pull. Write the criteria down on paper first — what isn't defined you can't assess. Without defined criteria you only film pretty videos without insight.
Set the camera up fixed — from the side for rotation and entry, frontal for balance. Always the same angle, or you compare apples to oranges. Slow motion on if your phone can do it.
Now the actual coaching step: go through the throw one frame at a time. Check the sequence (kinetic chain: proximal before distal), the rotation angles and the moment of kuzushi. Compare with your reference clip.
From the analysis almost always one dominant fault emerges. Take only that — say "shoulder rotation too small" — and train specifically against it until the next clip shows the progress. One criterion per cycle.
After two or three weeks: same angle, same technique, new clip. Measure the same number and compare. Log the number in a table, or the comparison fizzles out. This re-check is the core — progress you don't measure is progress you're only telling yourself.
Video analysis is risk-free in itself — the trap lies elsewhere: those who optimize for numbers tend to "force" technique under fatigue and with too much speed to get a better clip. That's exactly where injuries happen. So analyze on fresh, controlled repetitions, not at the end of an exhausting session. And never optimize an angle at the expense of your uke's safety: a harder throw for the camera that your partner can't fall cleanly isn't progress but a risk. Measurability doesn't replace the basic law — partner and lower back first.
Build a personal metrics table with three or four hard values and track them over months: shoulder rotation of your seoi (target toward 247°), dead-hang time as a grip proxy, randori rounds sustained. This table is worth more than any gut feeling, because it shows you in black and white whether your training works or you're just busy. The best athletes at their core do nothing else — they replace opinion with measurement. What you measure you can improve; what you only feel stays chance.
Film yourself from a fixed angle in slow motion and check the throw frame by frame against pre-defined criteria — such as shoulder rotation (elite around 247°) or the sequence of the kinetic chain. Take on one issue, train it and do a re-check after weeks with an identical setup.
More than you think: rotation angles of shoulder and pelvis in the throw, the proximal-to-distal sequence of the movement, the pulling force (elite around 4.36 N/kg) and grip strength (around 43–61 kg per hand for pros). Such anchors turn "felt good" into a checkable number.
Because only then can you compare clips. If you change the angle between recordings, the measured quantities seem to change even though your technique stayed the same — and vice versa. A tripod in a fixed position is what makes your numbers reliable and your re-check meaningful.
An analysis re-check cycle every two to three weeks is a good rhythm: enough time for a deliberately trained issue to change, but short enough to stay on it. Repetition is what matters — a single video is a snapshot, the series shows real progress.
Judo: Getting Started and Overview