What if the numbers looked like this:
Aircraft "usage"
P-40B 48.46%
Spit XVI 5.44%
N1K2 4.11%
F6F-5 2.99%
SeaFire 2.53%
F4U-1D 2.42%
La-7 1.79%
A6M5b 1.46%
Typh IB 1.44%
P-38L 1.12%
Would we call the P-40B "unbalancing" and perk it?
Granted your original example was with the P-51D in that spot, but I would expect an aircraft to be perked long before it reached the 48% mark. Maybe 25% like the famously quoted percentage for the F4U-1C? With the number of competitive aircraft we have in the game (today), an even lower percentage like 12-15% should raise eyebrows.
Yes, the P-40B example is an intentional exaggeration: these numbers don't say as much about aircraft performance as we might think. Case in point is the P-51, which isn't the best arena plane in the game (we all agree on that), but it is often the most popular. If the P-40B had "unbalancing" usage numbers, we all agree that it should remain unperked nonetheless.
Another variable is pilot quality. Suppose that all of the people who fly the Spit16 suddenly have the abilities of someone like drex. Even if the usage percentages remained the same, the Spit16 flown to the best of its abilities would change the face of the arena, whereas if the same thing happened to P-40B pilots (or FM-2, A6M, etc) everyone would be able to adapt. Even if every P-51 were flown by skilled pilots, it would not be a game changer.
These are my concerns about the usage definition of "unbalancing." I think it's a definition that sort of works, like the 3rd grader who finds a long division shortcut that yields the right answer 75% of the time. Moreover, the usage definition doesn't really capture our intuitive notion of what it is for an aircraft to be unbalancing: to be used very frequently is not the same as use that makes an average pilot deadly.