I don't think that holds water. You're going to get different range measurements whether the shooter is the same distance directly ahead of you, to the side, or directly behind you (the observer, who I'm assuming is not standing right next to the victim -- maybe the videos are taken from close enough that my point is moot).
I might buy the argument that if you observed large lag differences that could indicate multiple shooters firing from different (but not calculable) distances. But I'd still be skeptical about that since I don't know the path the sound takes to get to the observer (which will probably differ with a guy firing into a crowd).
Of course, if someone could show that the difference between the time of impact and observed report was so small that it simply could not come from the known shooters location, I'd say that's pretty strong for multiple shooters.
Okay, let's try this with just one round (actually the last one in the mag, which is what he claimed to have used). Sound travels at a given speed for given temperature, pressure, and humidity. That bullet does, too. The bullet is faster than sound. The bullet from small arms, in this case a rifle, travels along a line of sight, dropping with distance from that trajectory being flat, as it slows and the effects of gravitational acceleration downward take over. The sound, incidentally, does, too (at least that which does not reverberate off of or echo off of other objects). The first you hear will have traveled most directly. Knowing the speed of the bullet, its time to arrive at any given location can be calculated. Ditto for the sound of the bullet being fired from the gun. One (sound, in this case) is slower than the other. The difference for a .223 55 grain bullet in this case was found to be a little over half a second. To get that spread, the bullet would have had to travel (while remaining supersonic) a specific distance. That distance is the range. It doesn't matter which way anyone is facing, except that the bullet is being fired toward the target. It's all simple math. The rate the bullet slows can be calculated, therefore the time lag between the arrival of the bullet and the sound of the rifle can be used to determine distance.
The fellow in the video claims that by analyzing the sound track from videos, he can filter out other sounds and measure the time lag between the last bullet impact from a magazine and the last rifle report from that weapon, and that the differences were significant, with one set falling in the 0.5+ (and change) second range, the others around .275 seconds (iirc), and those two clusters of data indicate two shooting positions with one about 200 yards closer than the Mandalay Bay position. That shorter radius from the center of the concert venue takes in a couple of rooftops and another feature which was present on google earth (but may be gone) which would have been good, elevated, firing positions, but could include anywhere on that circle, give or take a few feet.
We use the same principle counting seconds between the faster than sound flash of a lighting bolt and the thunderclap to get the distance to the lightning strike. The shorter the time lag between the flash and the boom, the closer the lightning hit.
In this instance, the faster object is a bullet instead of light, so the time lag will be smaller (bullets being so much slower than light), but ideally measurable, and because the ballistics of the bullet and sound are predictable, the distance traveled by the bullet will coincide with a specific time lag between the arrival of the bullet and the sound of the gun being fired, just as I'd see the splashes of rounds hitting the Potomac, fired from the fixed mounts at Dahlgren miles away, and hear the sound of the guns firing a few seconds later.