Pool Regular Winrate and Mathematical Meaning
I'm not very good at applying mathematical concepts to real life. Right now I'm dealing with a big downswing, and as always, I'm trying to inject logic into it.
I filtered the 1000 regular players with the highest hand count from the GG RNC 200 Pool data I have. Their average win rate is -2bb/100 and the average standard deviation is 90bb/100. Lets say that 1000 new regular will play RNC 200 for 1 year to reach 1mil hands. After that 1 mil hands they will share their results with the community. I try to understand that how many of them will be stay in different winrate range. According to my calculations, the results will be as follows at a 95% confidence interval.
If I understand correctly, the four players who are essentially identical in skill and have a win rate of 1-2bb/100 will be considered the best players in this pool, and people will be scrambling to get lessons from them.
The 68 players with a win rate of 0-1bb/100 will be considered very good players, while those with a win rate of -1-0bb/100 will likely be considered average players.
Consider that one player with a win rate of -7-6bb/100. They'll be labeled "Fish" and might spend a lot of money to get coaching from those four players with 1-2bb win rates, even though they have exactly the same skill level. Or they might quit the game altogether.
I'd appreciate your comments on whether I can think of it this way.
8 Replies
Forget to write that. I assume that 1000 Reg have the same winrate -2bb/100.
Have you played around with https://www.primedope.com/poker-variance...
I don't think your numbers are correct. The probability of ending up above 0bb/100 should be 1.3%.
=1-norm.dist(0,-20000,9000,1) = 1.31%
Sorry my bad. Calculation was wrong but it will not change what i try to say.
Players who have 0 to 1 bb/100 winrates will be count as best players in that limit.
-1 to 0 good players
-3 to -1 meh regs who tries to understand how can a guy can have possitive winrate in that game
-6 to -3 do i need to quit poker or maybe i can get coaching from that 13 players who have possitive winrate.
Yes, that's true. Patrick Howard wrote an article about exactly that:
There is a lot of variance in poker for sure.
I'm not very good at applying mathematical concepts to real life. Right now I'm dealing with a big downswing, and as always, I'm trying to inject logic into it.I filtered the 1000 regular players with the highest hand count from the GG RNC 200 Pool data I have. Their average win rate is -2bb/100 and the average standard deviation is 90bb/100. Lets say that 1000 new regular will
Hello
We understand what you’re trying to do — using math to make sense of your downswing and look at your results more objectively. That’s actually a very solid approach, especially when variance feels overwhelming.
The key idea is that even over large samples like 1M hands, variance still plays a huge role, especially with a standard deviation around 90bb/100. A 95% confidence interval doesn’t mean players will “finish” in those ranges — it only shows where their true winrate is likely to be based on observed data. In reality, the spread of results will still be wide, and some players will run far above or below their true EV.
This is exactly what we focus on in our analysis. We don’t just look at results, but help separate variance from real leaks, understand realistic winrate expectations, and identify where EV is actually being lost.
You can check examples and feedback here:
https://forumserver.twoplustwo.com/290/c...
If you want, we can review your database and help you understand whether it’s mostly variance or actual leaks. First look is completely free.
RnC is played for rake back, so their ture wr/income should be shifted for like 4bb/100 or whatever rb is.
Your general idea is correct. If you play in game where your edge is close to zero, variance will be huge and you'll need milions of hands to converge to true EV. Maybe go to regular tables and game select. RnG is for unique player type, reg who can play 100k+ hands every 30 days with very high B-game. Most players after while will start to play terribly due to huge swings or burnout. It's important to be honest with yourself on whether you are that player type or not.
Sample size matters a lot here. Most people way underestimate how many hands they need before their winrate converges to something meaningful. Even at 100k hands you can still be running significantly above or below your true rate. I wouldn't read too much into anything under 200k honestly.
