Gotta remember the centre of the string moves at 3 times the speed of the tips of the bow.
That's why arrow and string weight are more important.
E.G at a 27 " draw the tips of the bow only come back about 9" but the string and tips both return to brace in the same time.
The logic of the experiment is also slightly flawed, reducing the tip mass must to some extent also effect draw weight.
A better test would be to add weights to the tips and see how much is needed to reduce speed by say 5fps. That way every other variable can be left the same.
Even if there is little measurable gain, reducing mass can't do any harm, and several small but barely measurable improvements can make a big difference.
I'm sure we've all re-worked a bow that isn't quite right and made it into a sweeter faster bow with a little work and some fiddling and fettling.
The other side of the coin is the sloppy approach of ... just leave it as it is, it doesn't make any difference. We know that's jut no good.
I often tell people, "It's easy to make bow, it's hard to make a good bow"
Del