The bigger problem is the thread pitch, which as I mentioned puts all of the stress on just a couple of short lengths of the thread spiral along it's 1 meter of approximate length.

Then, under heavier torque, the stressed short portions of the threading yield, thus spreading the loading over much greater lengths of the 1 meter or so of total thread length.

At some level of torque there will be a moving point of failure along the thread spiral, such that the most-deformed sections of threading will yield entirely. This yielding is accelerated by any diameter mis-match which would force the stress out toward the narrow peaks of the thread spiral, thus further reducing the torque capacity when it is an English or Italian-threaded freewheel being tightened onto a French-threaded hub.
However, when making actual diameter measurements, surprise number one, you may be hard-pressed to discern such diameter differences even when using a good digital caliper. Go figure. So you can't reliably tell the threading standard by measuring the diameter.

One should carefully inspect hub threading visually for thread distortions before concluding that there is a thread match with any particular freewheel threading, since often freewheels have been forced onto the wrong hub, thus altering the hub's thread pitch.
And in the case of an English or Italian-threaded freewheel installed onto a French-threaded hub, this forced installation can require relatively little force, often occurring within just the last 2-4 turns.