The no-gi judo grip set

The Principle

The judo throw system is often described as if it depended on the gi — as if the collar grip, the sleeve grip, and the lapel were the technique. They are not. The gi grips are the gi-context delivery mechanism for a more fundamental requirement: the connection that allows the thrower to load the opponent’s weight onto a single leg, then remove that leg. That connection requirement is invariant. The grips that satisfy it can be substituted as long as they deliver the same mechanical effect.

The no-gi judo grip set is the catalogue of substitutions. Each gi grip has a no-gi equivalent that achieves the same connection. A collar tie replaces the collar grip. A wrist control or two-on-one replaces the sleeve grip. A front headlock replaces the cross-collar. A body lock replaces the belt grip. The throws themselves do not change — uchi mata is uchi mata, osoto gari is osoto gari — but the entry geometry shifts to use the substituted grips.

Invariants Expressed

The Grip Mapping

Each Grip in Detail

Collar tie — the collar substitute

The collar grip in gi judo controls the opponent’s posture by attaching to the cloth near the neck. The same control is achieved in no-gi by the collar tie — palm cupped behind the opponent’s neck or behind the trapezius, elbow heavy on their shoulder. The collar tie does what the collar grip does: it controls the head and shoulder line, it provides a pulling vector for kuzushi, and it gives the thrower a fixed reference point against which to rotate. The double collar tie intensifies the same control, equivalent to a strong two-handed collar grip in gi context. Both versions preserve the throws that depend on head-and-posture control: osoto, ouchi, harai goshi, and the broader hip-throw family.

Wrist control / two-on-one — the sleeve substitute

The sleeve grip in gi judo controls the opponent’s arm by attaching to the cloth at the forearm. The no-gi equivalent is wrist control or, more strongly, the two-on-one (Russian tie). Both attach to the same anatomy that the sleeve grip attaches to via cloth — the forearm and wrist — and both deliver the same mechanical effect: the controlled arm cannot post, frame, or underhook. Uchi mata’s classic sleeve-pull-and-rotate entry adapts directly: the no-gi thrower pulls the wrist in the same vector, across the body, to load the opponent forward. The arm drag is the no-gi grip set’s most distinctive entry — there is no clean gi equivalent — and it serves as a back-take entry that gi judo achieves through different mechanical paths.

Front headlock — the cross-collar substitute

The cross-collar grip in gi judo reaches over the opponent’s shoulder to attach to the far lapel, providing a long lever for posture-breaking and rotation. The no-gi equivalent is the front headlock, achieved after a snap down or directly from a collar tie. The front headlock attaches to the head and the shoulder rather than to the cloth, but it produces the same effect — the opponent’s posture is broken forward and downward, and the thrower has the long lever they need for the seoi family of throws and for the snap-down-to-front-headlock chain. The no-gi seoi nage and the drop seoi (seoi otoshi) both use the front headlock as the entry grip rather than a cross-collar.

Body lock — the belt substitute

The belt grip in gi judo, particularly the rear belt grip used for suplex family throws, provides hip-to-hip connection and the rotational platform for lifting throws. The no-gi equivalent is the body lock — front or rear — locked around the opponent’s hips with the thrower’s hands clasped at the lumbar. The body lock provides the same hip control that the belt grip provides via cloth, and it preserves the throws that depend on hip-to-hip connection: the suplex, the lateral drop, and the broader lifting-throw family. The body lock also enables sacrifice throws like tani otoshi that depend on controlling the opponent’s hip line as the thrower drops.

Practical Application

The mapping is descriptive, not prescriptive. A no-gi competitor who has trained judo with a gi grip set does not need to relearn the throws — they need to learn the substituted grips and re-drill the throws with those grips as the entry. The kuzushi vector, the rotation axis, and the finishing geometry are unchanged. Only the grip is different.

The most common error in no-gi judo training is treating the no-gi grip as mechanically inferior to the gi grip and therefore expecting the throws to “almost work.” This is connection precedes control misapplied. The no-gi grip is not inferior — it is different. The collar tie does not produce the same friction that the lapel grip produces, but it produces equivalent connection at the same anatomical point. A throw that depends on lapel friction can be redesigned to depend on collar-tie tension, and the throw still works. A throw that “almost works” without the substitution is usually a throw being attempted without the prerequisite no-gi grip — the fix is not to add force but to add the grip.

For training purposes, the no-gi judo grip set should be drilled as a pummel sequence that reaches each grip in turn — collar tie inside hand fight, wrist control to two-on-one upgrade, snap down to front headlock, body-lock entry from over-under. This pummel produces the grip set as a chain rather than as four separate skills. Each grip flows into the next, and each grip enables a throw family. The pummel is the no-gi judoka’s primary drill in the same way that the gi pummel and grip-fight is the gi judoka’s primary drill.

Related Pages