But don't human heads vary, sometimes greatly, in size from one to another?
As noted the older samples are outside of the range of current measurements (certainly with respect to overall body sizes).
But for fun let's say that you're right. There is and has always been a range of features, cranium volumes and more that overlap with the current human range. So everyone of these specimens can be fitted in that normal range. Just pretending here, right?
Now then, explain a couple of things:
How come only those that are on a very extreme end of this hypothetical range are the ones perserved?
and
How come when they are laid out in time various features are closer to the current norms when they are nearer in time and further from those norms when they are further away in time? (With some variance since I don't see that any feature would constantly and smoothly approach the current state)?
When you are hypothosizing you must include
all the facts.