How to measure progress in Brazilian Jiu Jitsu

The reliable signals are negative: things that stop happening. You get caught by the same thing less often, you survive longer in bad positions, you are less tired for the same round. Belts are too slow to steer by and submissions are too noisy — and both are what most people use.

Brazilian Jiu Jitsu is unusually bad at telling you how you are doing. There is no weight on the bar and no time on the clock, and the two obvious measures both fail.

Why the obvious measures fail

Belts are too slow and too coarse. A blue belt might take one year or three. As feedback, a signal that arrives every eighteen months cannot tell you whether the thing you changed last month helped. Belts also measure something broader than skill — consistency, conduct, time in the room — which is reasonable for what they are and useless as a training instrument.

Submissions are too noisy. Tapping people depends heavily on who you rolled with, how tired they were, whether they were experimenting, and how much bigger than them you are. A night of catching people can mean you improved or can mean the room was full of new white belts. Counting taps mostly measures your training partners.

Feeling better is unreliable. Confidence tracks sleep, mood and the last five minutes of the session. Most people’s sense of whether they are improving is dominated by how their final round went, which is close to noise.

What actually tracks improvement

Six signals, in roughly the order they appear.

You are less tired for the same round

The first thing that changes, often within two or three months. It is efficiency rather than fitness — you have stopped spending effort on things that did not need it. See why beginners gas out.

Easy to miss, because it arrives as an absence.

You survive longer in bad positions

Before you can escape something reliably, you become hard to finish from it. This precedes visible improvement by months and is the most useful leading indicator there is. If you used to get submitted from under mount in thirty seconds and now it takes two minutes, that is real, and it will turn into escapes later.

You know what happened

Early on, rounds are a blur. Later you can describe the sequence: where it started, what you tried, why it failed, what they did about it. Being able to narrate a round is a large cognitive step, and it reliably happens before the results change.

You get caught by the same thing less often

The clearest signal of all. If you were getting armbarred from the same position every week in March and it is now rare, that is improvement, independent of who you were rolling with. This is the one worth tracking deliberately, because it is specific enough to act on.

You have preferences

At some point you stop attempting everything and start having a game — positions you steer toward, techniques you trust, situations you know what to do in. That is the beginning of having a game at all, and it usually appears somewhere in the second year.

New people find you difficult in a boring way

Not spectacular. Just hard to move, hard to pass, hard to submit. This is what competence actually looks like from the outside, and it is a better sign than a highlight-reel submission.

The measurement problem

All six are real, and all six are hard to see, because they happen slowly and your memory of training is dominated by the most recent and most emotional moments.

That gap is where most of the useful information goes. Somebody who has been swept from the same position eleven times this month will not recall it as a pattern; they will recall the one time it happened in front of the whole room. Ask most people which position they end up in most often and they will name the one they like.

This is the entire argument for writing down what actually happened — not as record-keeping for its own sake, but because these patterns only become visible once they are outside your head. It does not need to be elaborate. Two lines after a session, naming where you kept ending up and what kept catching you, is enough to see a month later what you could not see at the time.

A reasonable cadence

Do not assess weekly. Weekly variation is almost entirely noise, and checking that often is a reliable way to feel like you are going backwards.

Every two or three months is about right. Three questions:

1. What was getting me caught three months ago, and does it still?
2. What can I now do that I could not then?
3. What have I been avoiding?

The third one usually produces the most useful answer, and it is the one people skip.

About plateaus

Almost everybody hits a stretch where nothing seems to improve — most often somewhere in the first year, and again around blue belt, where the novelty has gone and the gap to the next level is large.

Some plateaus are real and some are measurement failures. The common measurement failure: you improved, and the people around you improved at the same rate, so nothing visible changed. Rolling with somebody you have not seen for six months usually settles that question in about a minute.

Real plateaus normally resolve by changing what you do rather than how much of it: working the positions you avoid, or doing more positional sparring instead of only free rolling. Adding a fourth session to a stalled schedule tends to produce a tired plateau rather than progress.

What not to do

• Do not count taps. It measures the room.
• Do not measure against the best person in the academy. They have years on you and that gap is not the signal.
• Do not treat one bad session as data. Sleep, work and the flu all show up on the mat.
• Do not wait for a belt to tell you. By the time it arrives, you improved a year ago.

The uncomfortable truth is that progress in this sport is genuinely hard to perceive from inside it. Most people who are convinced they have stopped improving have not — they have simply run out of ways to notice.