[Closed] Two players with the same skill get very different results?

it depends on the skill level and the type of game I guess as well. I mean let's say for nl bss, if both players are bad, then they will put their money in in bad spots and therefore lose/win more due to variance. However, if they are good players, they'll put their money in when they're ahead and they will minimize their losses due to variance. I mean they can make different decisions on each street and reduce their losses/maximize their winnings that way.

The only thing that matters then is how many good spots they get. Some sessions you just wont have that many big hands run into one another, other sessions you'll win a few stacks in a matter of minutes. So I'd say the difference in overall results is not based so much on the negative variance of the game, coolers and suckouts and what not, but more on the positive variance of the game, aka how many good spots you get into, etc. (I mean stuff like top set vs middle set on a safe flop, obv you're getting ur money in and neither of you can do much about it since folding is out of question).

Due to that, any good player here can compare his great sessions with the good ones and see the difference. Ultimately similar difference would occur in this hypothetical situation in which we have 2 players of the same skill level.

there are 2 options in here.

player A plays NL100 with 1k roll where player B plays NL10-25, and player A wins AA KK and AK more then he is loosing it, and player B looses sets straights daily multiple times, where player A wins it 90% of the time. This way its not hard to imagine how the differences are so huge. I checked what my winrates would be if those aa/kk/sets/straights would win and not loose, the winning would go from around 3k to at least 12k, and i talk about daily "unlucks" 80 vs 20 and so on..

Easy answer -> If they have the same expected value, their current winning will be converging to that expected value as the number of hands grow, theoretically, if they have played infinite number of hands, their winnings would be the same, equal to the expected value of their winnings. That is the theoretical part, which is true. HOWEVER, there is a misunderstanding in the word convergence - it does not mean, that the difference will be decreasing in time on every step. This is kind of hard to imagine, but true.

However, no one can play infinite number of hands, so in a matter of lifetime, one of the players could be a loosing player (nasty but possible) and the other Tom Dwan. Sample size is hugely underrated in poker, I have done some simulations on SNGs some time ago, check my signature if interested. For example, 1 in 100 10player SNG players with expected ROI of 5% will actually be a loosing player over 5000 SNGs, which is about 250 000 hands, almost 1 in a 10 000 will still be loosing over 10 000 SNGs, meaning half million hands.

Variance is very huge in poker, even that huge that I would not ever thought it could be.

Over time ( a long fuckin time actually) they would have the same bankroll
IF they played under the same conditions and same players.

Enough said.

now you just killed the discussion duude

Originally posted by ihufa
now you just killed the discussion duude

Yeah well, you see... this discussion reminds me of this dog named Sparky, he was very friendly and loyal dog, one day he got old and his master shot him. End of story.

And you may ask yourself what does it have to do with this discussion, and that's actually a good question!

what kind of dog are we talkin here

....Or are we?

noone knows

I would suggest that the simulation that produced the graph above does not actually account for the real situation.
People are naturally results oriented. If one person finds acting a certain way in a certain spot is losing them money, regardless of whether it's the correct choice or not they will alter their behaviour in the future. This means that the original difference is purely in winnings but then their EV winrate changes because they alter their game. At the same time, the winning run could cause the player to question their game and plug some leaks, increasing their expected winrate. In other words, the expected winrate is also a variable over time and is affected by absolute results.

The picture also changes when limits are included (Although this may be negated by there being a maximum limit you can reach) - the expected difference between them is magnified as there is an exponential increase in the other players earnings as they step up a limit. This is equivalent to shifting the graph of the player who was initially unlucky in the negative x direction, as well as negative y. If you make the big blinds in the above graph have a higher value as time progresses based on how many have accumulated already you can see apparent divergence in their monetary winnings.

Originally posted by NightFrostaSS

Enough said.

What kind of thing would post that