Interesting, however the animation with the overhead view doesn’t quite synchronize with that analysis. It appears to me that it shows Trump’s right hand all the way up near his ear at the time of the second bullet.
I now have the animation. The right hand flinches just before the second shot; not just after the first shot as I tried to recall earlier. A few in the crowd begin to duck shortly after the second shot.
First Bullet | | Right Hand Begins to Rise | | Second Bullet | | Third Bullet |
I would say the bullet reports arrived at the camera's location about the same time Trump would have heard it; the camera and Trump are roughly a similar distance from the muzzle. The span between each shot differs; since the shooter was using a semi-automatic, he had to manually fire each shot. So between each shot, he was experiencing recoil, meaning his trajectory roamed a bit. I don't think he had a gun rest.
The gunman fired off another five rounds after the initial three-shot sequence. Going by the Times' gunman view, Corey Comperatore seems a bit away from the Trump line-of-attack.
_____
The overhead animation was made just a few days after the shooting. I suspect they didn't have a model of the overall area, didn't know exactly where the shooter was, didn't see where the first three bullets struck. Maybe they synced it to an audio; maybe they didn't.
For an animation done fast (maybe with AI), it was pretty good. The bullet would have struck Trump's skull if he hadn't moved. That's the point they were trying to make. To do all eight trajectories right and with precision, it will take the FBI months. There's the metal fragments to account for; maybe some bullets disintegrated. They might have to do ballistics tests. In a lessor case, they would just account for the Comperatore and wounding shots and leave it at that.