See how the right shoulder joint spins rapidly in this maneuver, despite the upper arm remaining mostly stationary. Contrast with the left arm remaining mostly fixed.
This is because they are playing too close to the polar singularity of the shoulder, or as a biomechanist might recognize as the root of Codman's Paradox.
The elbow is unnecessarily high in this move. Given the elbow redundancy, there is no excuse to have it so high, and is poor motion planning.
In addition, I have pointed this out before, the easy way to avoid this problem is to mount the shoulder at 15 to 30 degrees above horizontal, so you can play in this region without entering the singularity event horizon.
Why is this a problem? Joint overspeed errors, and torque limitations, excess power consumption, and wear.
Or you can all design a better shoulder joint. I have ideas.