The Teaching Methodologies
By OSJJ
Open Source Jiu-Jitsu
THE CURRICULUM SERIES · PART SIX
The Teaching Methodologies
There is more than one way to teach Jiu-Jitsu, and most coaches only use one. A tour of the methods you can actually put to work on the mat, what each is, who it reaches, and where it helps.
Justin Hall / Co-Founder, Open Source Jiu-Jitsu
Part Five asked a single question, what is the constraints-led approach, and answered it by running a test. CLA turned out to be a teaching methodology, one point on a third axis we called delivery, the how of coaching rather than the what or the why. This piece opens that axis all the way up. If CLA is one method, what are the others, and what is each one actually for?
This is a tour, not an argument. The goal here is only to introduce the methods you can genuinely use in a Jiu-Jitsu room, name them plainly, and show what each one looks like on the mat. Some you already use without a label for them. A couple may be new, and those are worth sitting with, because a method you have never tried is a door you have not walked through. We are not ranking them, and we are not yet talking about how to combine them. First you have to know your tools.
One spectrum, from telling to discovering
Before the list, one idea that organizes all of it. Teaching methods sit on a spectrum. At one end the coach supplies the answer, this is the technique, do it this way. At the other end the student finds the answer, through games, problems, and constraints, while the coach shapes the conditions. Everything below is a point somewhere on that line, from most teacher-directed to most learner-directed.
This is not a new idea. In physical education it is the heart of Mosston’s Spectrum of Teaching Styles, a fifty-year-old framework built on exactly this range, from command at one end to discovery at the other. And here is the part worth borrowing: Mosston’s central claim was that there is no single best way to teach, only different approaches suited to different circumstances. That is the whole spirit of this piece. None of the methods below is the right one. Each is a tool with a job.
The methods on one spectrum, from teacher-directed to learner-directed. Left to right is not worst to best, only how much the coach supplies versus how much the student discovers.
The methods you can use
A note on the order. These follow the spectrum above, left to right, from the coach supplying the answer to the student discovering it. That happens to run from familiar to unfamiliar too: the ones near the top you already see in almost every academy, and the ones near the bottom are rarer, which is exactly why they are worth a look. They are not novelties to collect. Used well, a less common method can reach a student the usual approach never quite did, and it can spark a level of curiosity and engagement that a room running on the same default every day rarely produces. If something below is unfamiliar, treat that as a door worth opening, not a footnote.
The advance organizer
What it is. The short framing at the top of a class, what today is about, what we are trying to accomplish, and why it matters, delivered before anyone touches the mat. Almost every coach does some version of this without knowing it has a name and a research base. It does. In education it is called an advance organizer, from David Ausubel’s work in the 1960s, and it is one of the most consistently supported ideas in teaching: give the learner a conceptual frame before the detail, and the new material has somewhere to land. It is direct instruction in its conceptual form, aimed at ideas rather than movements.
On the mat. Two or three minutes at the start. Today is a transition day. If we are in side mount, where would we rather be, and why? Then name the one road you are working. Done well it is a conversation, not a speech, you ask, they generate the map, and the lesson gets a spine before a single rep.
Who it reaches. Everyone, which is why it is first on this list. Learners who need to see the big picture before the parts cannot function without it, and even the ones who want to skip straight to rolling retain more when the day has been framed.
Strength and limit. Strength: cheap, fast, and it measurably improves retention of everything that follows. Limit: it is easy to overdo. Three minutes is a frame, fifteen minutes is a lecture, and a lecture is where you lose the room before you have started.
Direct instruction
What it is. The standard model, and the most teacher-directed. The coach demonstrates, breaks the move into parts, and the class drills it, then applies it live. In the research this is command and practice style, or explicit instruction.
On the mat. Show the armbar, break it into steps, everyone drills it in pairs, then work it from a positional round. This is what most academies do most of the time.
Who it reaches. Learners who want a clear model to copy and a defined thing to practice.
Strength and limit. Strength: fast, clear transmission, and it builds precise technique. Limit: used alone it can produce clean drilling that falls apart under live resistance, because the student practiced the move but not the decision of when to use it.
One clarification, because it is exactly the confusion this series exists to clear up: direct instruction is a single methodology, but it can be run many ways. You can teach a move whole or break it into parts, work forward from the entry or backward from the finish, drill one thing repeatedly or mix reps up. Those are variations in how you deliver direct instruction, not separate methods, and choosing between them is its own craft, one worth an article of its own. We will give it that, in the piece on practice design. For now the point is only that the method almost everyone already uses has far more control available inside it than most coaches ever touch.
Reciprocal teaching
What it is. Students coach each other. One performs, the other observes and gives feedback against a clear standard, then they switch. The coach’s job shifts from teaching every rep to setting the standard and circulating. In the research this is the reciprocal style.
On the mat. You teach the armbar, hand the room a short checklist of what a good one looks like, and pair everyone up. One drills, the partner watches for the checklist points and corrects, then they trade. The room does the correcting, not just you.
Who it reaches. Learners who understand a thing more deeply once they have to explain and diagnose it in someone else. Teaching is one of the strongest ways to learn.
Strength and limit. Strength: it multiplies coaching, everyone gets feedback every rep, and the corrector learns as much as the performer. Limit: it is only as good as the standard you give and the eye of the partner, a beginner correcting a beginner can groove a mistake.
A second use, developing your instructors. Reciprocal teaching is also one of the most common and effective ways to build assistant instructors, and most coaches never connect the two. Correcting a partner against a standard is the instructor’s core skill in miniature: see the error, name it, fix it. Have an assistant observe and correct students against a defined standard and you are growing that eye under supervision, with the standard doing the guardrail work so they are spotting deviations from a known model rather than inventing corrections. It is how you develop someone into a teacher before you ever hand them a class, and it is one of the practical mechanisms behind a system that outlives the person who built it.
Self-check
What it is. The student assesses their own performance against a model or a standard, and adjusts. In the research this is the self-check style.
The catch. Pure self-check does not really exist. You cannot fully see your own performance from the inside, there is no way to truly check yourself by yourself. In practice you always need an external reference, a partner, a coach, or film. The dummy-drilling argument exists, but nobody actually develops real skill that way. So self-check almost never stands alone, it rides along with reciprocal, or with video.
On the mat. You film your own rep and compare it to the instructor’s, or you check your position against a set of cues while a partner tells you what they see. The self-assessment is real, but it leans on something outside you to work.
Who it reaches. Independent learners who improve fast once they can measure themselves against a clear model.
Strength and limit. Strength: it builds a student who can coach themselves over a career, the most valuable trait a learner can have. Limit: without an accurate external reference it becomes guessing, you cannot correct an error you cannot perceive.
How the two combine in Jiu-Jitsu. Reciprocal and self-check are the engine behind the online, distance-learning model of Jiu-Jitsu. You watch the standard on video, drill it with a partner while the two of you correct each other (reciprocal), then film yourselves and check the result against the model (self-check), often submitting it for outside review. Gracie University is the most prominent example of a program built on this combination. It is a real and legitimate answer to a real problem, how do you learn when there is no qualified coach in the room, and it is worth understanding on its own terms as a learning strategy.
Guided discovery
What it is. The coach does not give the answer. The coach designs a problem or asks a sequence of questions, and the student arrives at the solution themselves. In the research this is guided discovery, and its one rule is never tell the answer, guide toward it.
On the mat. Instead of teaching the counter to a common escape, you put a student in the spot, let the escape happen, and ask, what did you feel, where did the space open, what could close it. They discover the counter, and because they found it, it sticks.
Who it reaches. Problem-solvers, and anyone who tunes out being told but leans in when handed a puzzle.
Strength and limit. Strength: deep retention and ownership, the student understands why, not just what. Limit: slower than telling, and it demands a skilled coach who can design the right questions, done badly it is just a confusing class with no answers.
Game-based teaching
What it is. Also called Teaching Games for Understanding, or Game Sense. Instead of teaching the technique first, you start with a game or a live context, and the technique emerges because the student needs it to win the game.
On the mat. Positional sparring is game-based teaching, and almost everyone already does it. Start in a scramble, or in a specific position, and let people problem-solve live. The pass, the escape, the finish show up because the game demands them.
Who it reaches. Learners who learn by doing and competing, who glaze over during long instruction but come alive the moment it is live.
Strength and limit. Strength: high engagement, and skill that already works under resistance because it was built there. Limit: skips precision if it is never paired with instruction, students find something that works but not necessarily the best or cleanest version.
Versus CLA. This is the close cousin of the constraints-led approach, and the two blur in practice, so it is worth drawing the line. Game-based teaching manipulates the game to build tactical understanding, learning to read the situation and decide what to do. CLA manipulates constraints to build the movement solution itself, letting the body discover how to do it. One is aimed mostly at decisions, the other mostly at mechanics, and a well-designed constraint game is quietly doing both at once.
Constraints-led approach
What it is. The subject of Part Five. You design constraints, rules, limits, tasks, and the skill emerges as the student solves the position under those constraints. It is the implicit, learner-directed end of the spectrum, closely tied to nonlinear pedagogy and ecological dynamics.
On the mat. Two players start on the feet in a hand-fighting tie-up, and the only way to score is to win inside control, the underhook and the inside position, nothing else counts, no takedown, no submission. That single constraint forces the student to solve one narrow mechanical problem, the inside-position battle, and the solution self-organizes through dozens of live reps. That is CLA, positional drilling zoomed all the way in to one specific fight.
Who it reaches. Learners who thrive on exploration and self-organization, often the same ones who love game-based work.
Strength and limit. Strength: adaptable, live skill, and deep understanding of a specific problem. Limit: empty without content underneath it, and easy to overuse, it is a how, not a what, and cannot carry a program alone.
Versus game-based. See the note under game-based teaching above. The short version: game-based work shapes the game to teach tactics, CLA shapes constraints to grow the movement itself. Same family, different target, and they overlap constantly.
Differential learning
What it is. A newer, research-backed idea that runs against intuition. Instead of grooving one clean repetition, you deliberately add variability, slightly different reps every time, even imperfect ones, so the learner’s system self-organizes a more adaptable solution.
On the mat. Rather than drilling the exact same armbar entry fifty times, you vary the grip, the angle, the timing, the setup on purpose, so the student learns the principle of the armbar rather than one memorized path to it.
Who it reaches. Learners who plateau on rote repetition, and advanced students who need adaptability more than they need more reps of what they already own.
Strength and limit. Strength: builds skill that transfers and holds up when conditions change. Limit: unfamiliar to most coaches and easy to do badly, without a guiding principle it can look like sloppy, aimless practice.
Not to be confused with variable practice. There is a common misread worth clearing up, because a lot of good coaches think they already do this and do not. Teaching a scissor sweep from three or four different angles, ninety degrees, forty-five, across the chest, is variable practice, and it is excellent, far better than drilling one angle, because it prepares the student for the fact that the sweep never shows up in exactly one configuration live. But it is still direct instruction. The coach prescribes every version, and every version is real and meant to be kept. Differential learning goes one step further. The variations are deliberately imperfect, exaggerated, off-angle, never repeated, and they are not meant to be kept at all. They exist to disturb the movement so the body is forced to learn the principle underneath rather than any specific form. Think of it as three points on a line: one version, several good versions, and deliberate perturbation. Most coaches sit at the first without knowing the other two exist. Moving to the second, teaching more angles, is an upgrade you can make tomorrow. The third is a different tool.
Implicit teaching tools
What it is. Not a full method so much as a set of techniques for building skill without loading the student with verbal rules to consciously manage. That is what unites them, and it is why they all produce skill that survives pressure: a student running on a rule list chokes the moment the rules get overwhelmed, while a student running on an image, an accumulated correct feel, or an outcome to chase does not. There are three main tools, and each deserves its own explanation, because they are distinct and every one of them is immediately usable.
Analogy learning. You teach a whole complex movement with a single image instead of a list of steps. The classic research example taught a table-tennis stroke by saying move the bat as if you are drawing a right-angled triangle, one picture that carries the entire mechanic. On the mat, instead of ten cues for the hip-bump sweep, you say sit up like you are reaching to shake someone’s hand. The analogy loads the whole motion implicitly. It works because the student never has to consciously process rules, so there is nothing to forget when a roll gets stressful. You almost certainly teach a few things this way already without having named it.
Errorless learning. You design the early practice so the mistake is nearly impossible to make, then remove the scaffolding gradually. The logic is that when a beginner errs, they start consciously analyzing and correcting, which builds an explicit, fragile, rule-based skill. Prevent the errors early and the skill develops implicitly and stays robust. On the mat, teach a guard pass with the bottom player offering almost no resistance at first, so the pass cannot fail, then dial resistance up slowly. Or start a submission from a spot where it is already mostly locked in and back the starting point out over time. One caution: this is not going easy on them. It is a deliberate progression from cannot-fail to full resistance, and the progression is the whole technique.
External-focus cues. You point the student’s attention at the effect of the movement on the opponent or the environment, never at their own body. Internal cues, rotate your hips, extend your arm, make people worse. External cues, drive him toward the far wall, push the mat away, make his shoulder touch the floor, make people better and more automatic. This is one of the most replicated findings in all of motor learning, across dozens of studies. It works because attention on your own mechanics disrupts the automatic movement system, while attention on the outcome lets the body organize the mechanics on its own. It is also the easiest of the three to start using, you just change how you phrase the cue.
Who it reaches. Overthinkers, and anyone who performs worse the moment they are given a checklist of steps to remember mid-roll.
Strength and limit. Strength: skill that holds under pressure, because it was never dependent on remembering a list. Limit: less precise for the fine details, and finding the right analogy or cue is its own skill.
The seed, and where this goes
That is the toolkit, the methods you can actually put to work in a Jiu-Jitsu room. Notice what this piece did not do: it did not tell you which one is best, because none of them is. Each reaches some students and misses others, and each is strong at something the next is weak at. The direct-instruction coach and the games-only coach are both right and both incomplete.
Which raises the real question, the one the rest of this section is about. If no single method reaches everyone, and you cannot sort a class by how each student learns, what do you do? You already have the seed of the answer, you use more than one. But before we get to combining methods, there is a layer underneath that most coaches never look at: how you run the reps inside any one of them. Whether you teach a move whole or in parts, forward from the entry or backward from the finish, in identical repetitions or with variety built in. Those choices shape whether a technique sticks as much as the method does, and they are the subject of the next piece, on practice design. After that we get to the bigger question, why blending these methods reaches more of the room than any single one can. For now it is enough to know the tools by name, and to notice that a few of them may be doors you have not yet walked through.
Open Minds. Open Mats. Open Source.