I’ve always thought it was fairly obvious that since the elbow is a hinge, any change in trajectory had to come from some combination of moving the scapula around and movement of the humerus in the shoulder joint. I think the rotation of the humerus is very obvious in many players anyway. What’s exercised me (in both senses) since I do fair bit of elbow, is whether moving all that machinery around is practical for the purposes of high speed TWPS. I mean…‘down up rotate’ for example - rotate here entails shifting your whole upper arm assembly about. Not easy at high bpm, surely? Not to mention long term damage? It needs to be very minimal and accurate I’d have thought because of the mass and changes of direction…inertia!
Can Vinnie still do what he did in that old video? Would be great to see that on Troycam.
Hi @Troy, thanks for the concise explanation! Have you tried something like the Volcano pattern with this mechanic? I wounder if it’s possible to mix smoothly alternate + sweeps with the elbow.
The shoulder is probably the joint with the most unconcious movements in the body, so my guess is shoulder too.
It might be relatively easy to check that, if it’s shoulder there should be a rotation in the upper arm.
@Troy: can you post another clip where your mighty biceps is visible?
Hmm, I think we all agree, that the elbow can just produce a movement in one plane, or more precisely one arc. But what keeps this plane from being tilted in relation to the string plane? Nothing I guess.
What I mean is, DWPS or maybe even UWPS is not impossible using just elbow movement, it just needs the correct posture. TWPS would involve rotating the guitar around the neck axis 
Vinnie is a uwps player so he shouldn’t need any shoulder movement for two-way pickslanting. He can get his uwps pickstroke with pronated elbow, and he can get his fully escaped pickstroke with forearm + pronated elbow.
In other words, forearm still works when you’re using elbow - i.e. you can still get an escaped upstroke by turning your arm. The mystery here is only if we assume I’m not actually doing that in this clip. But who knows - I might be!
We’ll try some more test shots when we get a moment, to try and determine what is actually moving.
I’m not with you. In the Pepsi Lick, the “rotate” of the “up-down-rotate” on the B string escapes the B string, but it doesn’t provide a downward trajectory for the downstroke back to the E string. Maintaining the same elbow position even after the forearm rotates to a less pronated orientation surely will still result in missing the E string?
If you’re referring to string tracking, sure. But that’s not really a “2wps” issue specifically. In fact that would come into play even in a uwps phrase where you have to relocate the picking motion from string to string. And there is indeed often an upper arm or shoulder component to that. But it’s not a thing that’s happening on every picked note, which is I thought what you were asking about.
I’m not sure I follow what you’re saying with the Pepsi lick.
In uwps, after the 3 notes on the high age you would hit the B string with an upstroke first, followed by down, rotate. That last rotate puts you perfectly in line for the downstroke back to the high E.
So only one rotational movement per 6 notes in uwps.
Well, it’s the “rotate” phase where the “slant” (trajectory!) is being reversed that seems to me requires significant upper arm re-adjustment. It’s true, that even with the UWPS on the E string, there’s a small change of trajectory needed between the escape trajectory of the final downstroke (this is sounding more and more like rocket science :)) to the upstroke to the B string, which entails some relocation of the elbow, but not as much as the “rotate” on the B string which is a complete reversal of slant (trajectory).
No, this is my point. The rotation on the B string aims the pick at the E string again, but that’s not enough. “Slant” isn’t that, it’s the trajectory that matters and that’s still upward until you change something further up because the elbow is only a hinge.
Totally. The forearm just turns, the pick turns with it to play the upstroke, and escapes as it does it. The forearm rotates back to its original position, so does the pick. That’s your ‘trapping’ movement. You can do this on its own, or while elbow flexion/extension is happening. In fact you can do it in the air right now with no guitar. Unless I’m missing a subtlety here, that’s all I think is happening as far as the pickstrokes themselves are concerned.
The tracking motion is a different story. If I’m understanding you correctly, you are asking how do you get from string to string. If the elbow/guitar planes are angled to each other enough to permit uwps in the first place, then it should not technically be possible to do a pure elbow string track and still reach all the strings, because a downstroke movement just pushes away from the guitar. This is true for any elbow picking movement that also tracks across the strings - it really has nothing specific to do with 2wps, as far as I can see.
You can do a simple test with a kitchen cutting board. Hold it up to your midsection, like you would a guitar, and angle its plane different ways to simulate different pickslanting paths. Then place your hand flat on the board and run it from top to bottom. Look in a mirror while you do that. What movement do you see? Forearm is clearly happening when I do this - it is rotating to maintain hand orientation against the board. At more extreme angles, upper arm rotation is also obvious. At angles closer to the elbow’s plane of operation, it’s hard to perceive any upper arm rotation.
It’s tricky!
Ok, let me throw this at you guys.
I’m an uwps but I use wrist deviation and not the elbow. For all intents and purposes the deviation movement is exactly the same as the elbow motion. It moves in one plane.
In order for me to pick something like the Pepsi lick in uwps, I need to add another motion to the mix. For me it’s a forearm rotational movement. I’m not necessarily moving to a dwps, but I need to at least get back to my original uwps starting position.
Without that additional movement the lick can’t be played…the same has to happen with an elbow mechanic, no?
I’m surmising that even though we may not always see the additional movement (Zakk Wylde), it HAS to happen.
Am I missing anything here or are you guys seeing what I’m driving at too?
No hurry.
It’s just kindof funny to
a) see the master himself not knowing how one of his own motions is executed
and
b) being able to participate solving the riddle
You wrote before that there’s just one direction for the elbow, I totallly agree with that (I think read somewhere it’s the most fixed joint we have), so this one should be relatively easy to spot. If it comes from the shoulder it must be visible, in the shoulder or in a rotation of the upper arm (the shoulder itself can be hard to spot).
On second thought I’m pretty sure you know all that joint stuff way better than me (and probably anybody else here).
So summary: No hurry 
I’m not sure we’re completely understanding each other here. And it is tricky, so I’m going to think more about this.
But regarding the KCB (kitchen cutting board :)) test. I think that on moving the KCB through realistic pickslant planes from UWPS to DWPS it does require visible amounts of upper arm rotation when tracking my palm across it. But the amount depends on the shoulder position. If I hold my should back, then it requires quite a lot of humerus rotation with the downslanted KCB. With the shoulder forward, the way say Al Di Meola hangs it right over the guitar, then a lot less.
Edit: And just to add, my concern is that if a relocation of some part of the upper arm is to happen so as to allow the slant to be reversed and then a period of successive pick strokes at the new slant without an accompanied upper arm rotation on every stroke, that relocation is going to be big and sudden.
My spacial visualisation centre has temporarily overheated…
Not a problem I am frequently dense.
The way Vinnie does the “Pepsi Lick” there really isn’t a period of successive pickstrokes in dwps - there’s just a moment at the sixth note where the upstroke escape happens, and the arm rotates to facilitate that.
Have you been able to make the movement work hands-on or is this just a theoretical concern? Because I can tell you there’s nothing disjointed about the feeling of doing it. Obviously I want to understand these things better, but it’s from the perspective of being able to give someone clear steps for making these movements work for them, without wasted time and guesswork. From that perspective, I think we have a decent practical handle on how the base uwps elbow movement works, as well as the 2wps version of it. There may be some details about what precisely happens at the moment of the string change, and I do think it’s worth knowing. Especially if it is something that a learner might find problematic. But I think getting the base movement happening, and then learning the forearm piece, will get most people in the ballpark for the time being.
If someone flat-out can’t do the uwps elbow movement at all, that’s a problem. And that’s where we need to understand better what is going on. We’ve definitely had people on here say they can’t do dwps elbow. That makes a little more sense if we’re not entirely sure what is going on with the movement itself.
I’m not the master, I’m just the student with the biggest YouTube channel!
Edit: Actually that might be Ben Eller. So I’m not even that!
No, I didn’t meant that specifically. Just, say, with an UPWS setup, and then switching to DWPS for say successive DWPS pentatonics, you might not want to be still operating with an UWPS biased position and having to rotate the upper arm a lot when with a better DWPS setup, it’d be almost all be elbow flexion-extension.
Personally, I find elbow UWPS, comfortable. For elbow DWPS, I feel the need to keep connection with the body of the guitar and that that tends to produce a flexed wrist. A somewhat ‘gypsy’ setup seems to want to happen naturally, albeit that they’re using a lot of forearm rotation. But even there, you often see quite a bit of upper arm rotation. The trouble with a flexed wrist, is that rotating the forearm to change the slant creates a huge arc of the pick. Which is why I think the 2WPS I’ve seen generally have a straight and even slightly extended wrist - when they’re doing it - because it allows the pick to change slant within a small space and not bump into strings.
Maybe I just need to get into a more neutral/midway position and then the switch between UWPS and DWPS would be less traumatic.