Then why is the effect both predictable and
consistent?

When the results of an experiment conducted a million times are identical in
all regards, at what point can one draw a conclusion without being guilty of
voo doo science?

You could make a case that, under unique conditions, a vessel in displacement
mode
will attain speeds (in knots) greater than (sq rt of waterline expressed in
feet) X 1.3. You can make an accurate case that the multiplier for specific
hulls may vary between 1.2 and 1.4.

What cannot be argued is that beyond the calculated hull speed a vessel meets
the resistance of the bow wave, and must start climbing that wave (leave
displacement mode) in order to increase speed.

Exceptions prove the rule. If I observed that it is possible to have a rain
squall pass 20 yards away and stay perfectly dry, that would neither establish
that such a lucky break is a common occurence or that water isn't wet. :-)