I have to tell AS that 100K hands of Texas Holdem doesnt give the accurate results that you are thinking. 100K hands is a good measure but results from 100K hands of poker can be dramatically different if variance kicks.
I have also seen the case of a few players who dont know crap about statistics, probability theory or proper poker strategy doing quite well in a sample that according to AS standards is significant.
Run some numbers with your winrate and SD for the last 100K hands and you will find that running at 1/6 of that over another 100K hands is not that unlikely. Also a winrate based on 100K hands is less reliable than what you think.
In conclusion, 100K hands is nothing in Holdem
