[Darryl_P]

If my opponents played in a way that they went all in any more often than, say, once in a million hands, then the assumption of my having a 3 BB/100 advantage would be false...it would be much greater than that.

For me to have only a 3BB/100 advantage we must rule out extremely high variance plays on the part of my opponent since those would lose too much EV.

And when you bring down the variance to "normal" levels, then the random walk model can be used with high precision even in NLH.