As Max points out, the effect of misdirection is lesser on tape than in living 3D. But the larger problem here is - if you are going to test the effect of something, it might be a good idea if that something actually is present.
Teller's handling is is a cool juxtapose of old and new. An ancient trick with rock solid traditional technique, performed with flimsy modern plastic cups.... But notice, the technique used is both pre-Slydini and pre-Ramsay. It relies on using patterns to induce inattention blindness more than anything else. (For example, each ball is tossed between the hands back and forth before a false transfer is made, instead of using the precision the Ramsay approach would provide).
As an example, let's look at this clip:
https://dfzljdn9uc3pi.cloudfront.net/20 ... e%20S6.mov
First of all, the eye is attracted to movements. Slydini pointed this out. So what are the movements here?
I took the video clip above and got the difference from frame to frame - everything that doesn't move or change is black, only the difference from moment to moment is visible:
As you notice, head movements is not a big part of the routine. Without any extra techniques, in the natural state, the hands are the main attractor. There is nothing that contradict Ramsay or Slydini here.
All right, let's look at Ramsay's "If you want someone to look at something, look at it yourself."
Where is Teller's gaze during the piece? It is steadily fixed onto his own hands. I.e his gaze only reinforces what the audience would look at anyway. Not really as redundant as it might seem, because had his gaze been uncontrolled all over the place, it would detract and confuse. But his gaze doesn't really add
anything in his choreography, as it is pattern based misdirection. So blocking out his head would not provide any new insights or knowledge.