Normal Distribution -- Applicable to Winrates?

[Red_Diamond]

Borys, I have some issues on this win rate by mood issue.

Now I understand when on TILT, our EV generally falls. This is sort of common sense, but most literature tend to claim the opposite, when on the other side of the fence. they claim your EV goes up because your positive mood.

Let's say things are going our way, and we are feeling good about ourselves. In fact, instead of 'can't do any right', we are in a 'I can't do any wrong' state of mind. It would seem to me, that feeling TOO GOOD about your situation could be just as harmful. I think we've all been there.. "Oh my sets are running so hot, I'll call this bet even without odds because I'm hot right now... Oh I'll call that and see what happens on the flop, I know I can get away now if....."

Perhaps there is a big overlooked issue here. When we feel GOOD about ourselves and our play, it is because we have been winning. Even if we didn't deserve to win the last few hands, the fact we WON is what put us in that state of mind. We may be still playing bad, and playing worse the more we win (we are getting positive reinforcement for playing bad here). It is just 'luck' that we are sitting on +ev during that moment.

I think people go too one-sided in the mood status.

[Abbaddabba]

I think that more often than not, people say they play bad when they run bad because - well, they're being put in [censored] spots where you have to make difficult decisions. When you call down with a loser 5 times in a row, you cant help but feel down about your play. But the pots are often justifiably large that you dont have to be right very often. It's just that, for the several most recent events, you happened to have a loser. That doesnt mean you're playing poorly because you showed the hand down.

And then there are others who are just tilt monkeys.

[jogsxyz]

Assume your winrate is constant over time. This winrate was never normally distributed. The normal distribution is a good approximation of the distribution.
This is true of coin tossing. Coin tossing is a binomial distribution. The normal is used to estimate or approximate the binomial.

Now if you want to test if the game is getting tougher. Use hypotheses testing methods.