He is pushing any two or?

[derdo]

I posted this here because I think this is relevant to the basic assumption we make about hand ranges while using ICM in bubble time.
For example we assume that a pusher's range is 99+,AQ+. This means we assume that pushing probability for these hands is 1 and 0 for all the other holdings.
However, since our assumption is never perfect, the prob of seeing AA is higher than 99. And the probability of this guy pushing with 66 is not 0.
You might think one offsets the other but with extreme hand assumptions like pushing with any two (or pushing with 80%) it doesn't.
We use a discrete proabbaility function. If he has a hand in this range he is pushing 100% of the time, if not he is folding 100%.
But in reality the probability function is a continous decreasing one.
I am not sure if this has a practical value for ICM calculations but I feel like it might, particularly with extreme range assumptions about your opponents...