KK vs AA No Limit Texas Holdem

Hello there,

I have a samplesize of 370785 Unique Online Poker Hands with one player / 6 max tables...

I Tracked KK HoleCards: 1720
I Tracked KK HoleCards SawFlop: 899
I Tracked KK HoleCards SawShowdown: 383
I Tracked KK HoleCards facing AA: 33
KK vs AA Equity Occured: 0,0367
KK vs AA Equity Expected: 0,0288 (is this correct when I have mainly headsup situations?)

I would like to ask for the probablility that KK vs AA is 0,0079 above expected. Why can't I just use BinomDist(33;383;0.0367;1) in this case? Has been a long time for me doing the math. Thank you...

I'm more a programmer than mathematician, and not skilled with binomial distributions, but off the top of my head, why does "KK seeing SD" (the 383 in your BinomDist call) have anything to do with how KK does vs AA ?

I'm not really sure what I'm looking at... your #'s appear to be saying that KK expects to have 2.88% equity in a pot vs AA, but hot/cold KK runs about 18%, and with those two hands, in short handed games, I'd expect most of that to go in pretty close to hot/cold #'s, no? If anything, I'd think the occasional set of KK's would play better than the occasional set of AA's, since the overpair can assume you hit a K and will want to get money in the pot while KK might want to see a SD when the A hits. But then your sample size goes from small to almost non-existant.

So, how is the expected equity calculated? Is this just from a larger DB of hands? Or am I (from US) confusing the notation?

Off the top of my head, that variance doesn't look like much with a 33 hand sample, but it's been a long time since I've looked at poker databases.