To clarify, I'm not sure that it matters curve or not. I would think it does matter, but then maybe his measurement is within the margin of error. I'll have to look more closely at it sometime.
For now, after having taken a little closer look at the trig calculator used, and having heard the description given, I now think my initial suspicion is more likely to be correct, that the curve can be there without affecting the trig, and that just because trig uses straight lines, doesn't mean it proves the earth is flat.
He said the mountain is 2226 higher than the observation point. That suggests to me, that they are starting from the same reference ground level, which allows a bulge to exist between them without affecting the math. Basically, on a globe model this experiment centers the bulge evenly between the 2 observation points, so the mountain's measured height is not down behind the curve, because the observer is the same amount behind the curve. (see the curved ground drawing in my first post)
At some point this experiment might work to prove the globe or FE if the distance is far enough, because at some point an object will be hidden behind the curve if the curve exists, the problem being that due to perspective, hills and trees would hide a very long distance target even on a FE. Also, on opposite sides of the globe, there would be a contradiction to solve when plugging the numbers into the calculator of how high above the observer is the mountain, since both the observer and mountain are above ground level, yet at the same height according to a system of measurement based on space itself and an artificial horizon line reference rather than ground level being the reference.