OK, I drew a few straight lines to show what I mean.
The image to the left is his wrist position at address, the one to the right is at the end of the backstroke. Compare the angles taken from the outside of the 2nd metacarpal bone in the hand to the back of the ulna. You can't cheat this angle by opening up the fingers because the bone is positioned inside the hand itself. Yes, he is wearing a shirt, but I did my best to draw the lines where the bones appear to lie under his shirt. If anything, I tried to err on the side that demonstrates less movement rather that more.
I also went back and looked at the video frame-by-frame. There are about 20 clicks of the keyboard arrow keys to one second of video clip. That means that each frame represents 1/20th of a second (I thought I'd find 30 FPS as most videos are shot at, but...) I also made a scale on a Post-It using a snooker ball as a gauge for estimating inches and affixed this to my computer screen along the line of his cue.
His forward stroke begins almost imperceptibly slowly at about 3:01:17. Six frames later at 3:02:03 his cue has moved about 1". In the next two frames his hand has only moved another 1.5", but it is obviously increasing in speed.
The next frame shows his hand in a motion blur, but his knucle appears to have suddenly moved forward about 6". By the end of the following frame he has moved his hand another 5" and his tip has already contacted the white (this is where one of those 300 FPS cameras would be nice to have). At this point, the angle between his 2nd metacarpal and his ulna has closed sharply compared to the angle at the end of the backstroke, showing significant wrist motion took place just prior to contact.
In the following frame his hand only moves about 2" due to the deceleration of the cue caused by the collision with the white ball. Then the stroke ends with the tip of the cue... a mere 5" from the back of where the white sat before the stroke. This demonstrates that there is no need for an extended follow through to finish up a superior stroke.
I also noted that his elbow rises about 1" at the same time the cue lowers about and inch, indicating that his wrist has "broken over" to a straight position, extending the distance from the elbow to the cue.
BTW at no point have I stated that this is a "power snap", or that there is any tension in his stroke. But his wrist does move through a fairly substantial arc, and it all happens during the portion of the stroke that appears to have the quickest acceleration - just prior to contact with the ball.
I see absolutely no reason to have to incorporate wrist motion into your stroke, but many pros do it and it seems to work for them. If it creates tension, then that tension must be resolved or the motion must be eliminated because tensing the wrist or hand before contact with the ball will certainly throw the cue off course. Done correctly and with the proper timing, a lot of speed can be added to the stroke with just a few degrees of wrist adduction.
Anyway, this took up way too much of my time, so I am out of here (again). I am not showing this to stubbornly try to prove a point, but only to show how I try to analyze videos of athletic motions. Take it however you want, I don't really care, but this is how I see it. I started doing this years ago in order to analyze students flycasting strokes since I din't have a high-speed video camera. It has its flaws, but it can pick up little glitches that go unnoticed to the unaided eye.
