In with elbow, Out with hand: a Working Hypothesis

by Jean-Philippe Ranger on 2012/01/31

In the last few months, I have been trying to follow the rule of “elbow in, hand out”. In my practice and teaching I have observed something that needs further verification, but that seems correct. “Elbow in” means: initiate the arm rotation from the upper arm. Conversely, “hand out” means: initiate the arm rotation from the hand. There are a few possible explanations for this. First, I will say a few words about the “elbow in” part of rule, then I will suggest an explanation of the “hand out” part of the rule.

In the “elbow in” part of the rule, it seems to me that at least two things are concurrently happening:

  1. Since the contact with the opponent is somewhere along the lower arm, the upper arm and shoulder area are free, namely free to rotate.
  2. Not only is the arm free to rotate, it generates more torque in the contact point with the opponent than if the rotation was initiated from the wrist or hand. By initiating the rotation from the upper part, our arm acts like a long-handled screwdriver. If a screw is tight, it is easier to use a long-handled screwdriver to unscrew it than to use a short-handled screwdriver. The arm rotation performs the same function.

As far as I can tell right now, when the “elbow in” is done properly, the opponent’s grip is compromised while I am in a stronger position than before. This makes the idea of having the hand drive the rotation and the outward push seems possible. By initiating the rotation with the hand, I force the opponent to “stick to me” and the rotation ends up compromising his whole structure. Then, as the hand is drives out, the opponent is pushed out.

The evidence is there to confirm the validity of the rule. What I have tried to do here is explain why they work and more importantly, how to follow them.


{ 12 comments… read them below or add one }

pingwei January 31, 2012 at 9:20 am

I remember clearly that Master Chen demonstrated in a video about “elbow in” so that the “peng” never get lost in the hand. To do that, you need to find a spot on the forearm as a fixed point. In push hand, the fixed point is the contact point by the other person.


cshum00 January 31, 2012 at 9:32 am

It is also explained by of Master Chen’s positive circle videos. Here is a good picture explaining how the elbow in is a good lever; how you don’t mind my horrible drawings.

There are more physics, dynamics and mechanics into it. Like how in normal actions you only involve the mass of the forearm while the elbow-in action involves the entire arm or more depending how good your connections is. The fact that it is a curved motion in a 3D plane instead of a linear motion, etc.


JohnnyU January 31, 2012 at 8:17 pm

Drawing is good! It clearly shows the “longer lever” with the fulcrum point on the upper arm (point of contact)…


Calvin Chow January 31, 2012 at 9:07 pm

I like your sketch. It is very clear. However, I think the shoulder should not be lower down vertically, the shoulder joint should open up and stretch down to keep the same head suspension and it creates a better torque.


cshum00 January 31, 2012 at 9:37 pm

I think that the shoulder should go down vertically slightly.

First, it establishes a stronger structure between the arm and the upper body. Let’s say that you need more force; so you start to use your legs. If you don’t push down your shoulder downward, your shoulder easily pops up and loose connection between the legs and shoulder. This is specially true when your opponent is pulling your arm.

Although horribly drawn on the sketch, the shoulder going downward does not mean that the body does not go downward with it. The shoulder stretches downward. This is specially true when twisting the elbow towards the center. Sorry i am not good at drawing, otherwise i could have illustrated those better.


cshum00 January 31, 2012 at 9:59 pm

Double negation typo. I meant, that the shoulder going down does not mean that the body rest of the body should go down.


cshum00 January 31, 2012 at 10:33 pm

lol. Another typo. It would be nice if there were a “Edit” button. I mean that the shoulder goes downward vertically only due to the stretch. The rest of the body stays intact.


cshum00 January 31, 2012 at 11:38 pm

Here is the video i was referring to. Also notice by comparing both shoulders, the shoulder does go down slightly.


Calvin Chow February 1, 2012 at 1:20 am

Well noted with thanks. Keep the head up and rotate the shoulder down.


cshum00 February 1, 2012 at 12:03 pm

Pleased to be of some help.


edliaw January 6, 2015 at 12:16 am

I gave this drawing some thought. I wonder if the following makes sense: do we prefer the fulcrum to be not on a joint? The reasoning I have is that if the fulcrum is on a joint, only the muscle connected to that joint can exert force. If it is a point on the forearm like in the picture, though, the whole body from the elbow onwards can exert force.


bruce.schaub January 31, 2012 at 11:57 am

i recently watched master chen’s video on “shoulder movements in a positive circle” and found it to be incredibly informative in regard to this topic…. what i think i’m beginning to understand is that in the second third (of the three part sequence) we are supposed to expand the space between the back of the arm (tricep) and the center of the (lat muscle)…proportionally! this does two things… it gets the opponents power off of our center and aligns it (the forearm segment) with the rear calf segment (as is very clearly demonstrated in the germany private 3 video) if perfectly aligned it is as if there is no space between the two bone segments (or as if a stick magically appears between the knee and elbow….i think this is related to a physics concept called linear displacement)

now we dont have to think about the elbow anymore… on the way out we expand the space between the bicep and center of pectoral…proportionally… meaning the right chest has to cave in an equal amount to the forward movement of the bicep (as if they are expanding away from a point in space between them….this is critical) also the shoulder pushes down toward the kua as the arm extends…

it seems as if what is functionally happening on the way out is that the line that was established from the forearm though the calf to the ground must be maintained and we use rotation to squeeze more body parts towards that line proportionally (without breaking our rotational axis)which drives the hand out in a spiral….i have two new painful strips of muscle (in my shoulder)that my brain was previously unaware of trying two accomplish this…

i am new to practical method so my interpretation of what’s going on should be guarded against somewhat… i hope i am trying to accomplish the right things….


Leave a Comment
Leave a comment on the content only. For admin issues, please click the "contact" button on the top left.

Previous post:

Next post: