ClubWPTGold - RNG discussion

Exactly.

This is even more pronounced with higher win rates but higher standard deviations.

Regs don’t punt 200bb into the middle with bottom pair and a 9hi flush draw vs your top set, so variance is reduced but winrates are lowered.

Whales do so variance is increased but winrates are higher.

Over the long run you win more vs the whales and realize a higher winrate, but in the short term you can absolutely be blasted by variance and lose massive piles vs very weak holdings that you would theoretically never have the chance to vs winning regulars.

Here we have a 6bb/100 winrate with a 150bb/100 STD which is widely considered normal for PLO

Here we have a 20bb/100 winrate but a 500bb/100 STD which is what my tracker shows is the standard deviation for my live 8max PLO games over the last 3 years of tracking.

Notice how deep the downswing can be over 1000 hands despite having a 3-4x winrate vs absolute whales.

Over the long run I will absolutely print piles vs the whales, but in the short term I can experience swings that are way more severe than playing regulars.

This is the same on WPT. Massive winrates, much higher short term variance.

But the human brain can’t comprehend this without seeing the math. You all just think you are being cheated lol. It’s actually the opposite. You are sitting in one of the best game environments of the last 20 years and have the ability to print piles if you can just stomach the short term variance.

This discussion is getting a bit to nerdy for me so I called on Chat GBT. I will look at what you posted a bit more when I get a chance and give a human response. But this is what Chat GBT and Gemini both basically said about your this specific part of your response:

That response is incorrect—they’re mixing up concepts.

�� The core mistake they’re making
They’re claiming:

“Standard deviation is exactly the same regardless of edge”

That is false for a ±1 outcome coin model (which is exactly what your example is).

�� What’s actually true (simple math)
If each flip is:

win = +1

lose = –1

Then:
Mean (EV):
μ=2p−1\mu = 2p - 1μ=2p−1
Variance:
Var=1−μ2\text{Var} = 1 - \mu^2Var=1−μ2

Plug in the numbers
65–35:

μ = 0.30

Variance = 1−0.302=0.911 - 0.30^2 = 0.911−0.302=0.91

55–45:

μ = 0.10

Variance = 1−0.102=0.991 - 0.10^2 = 0.991−0.102=0.99

�� Result

The 55–45 coin has higher variance than the 65–35 coin

So:

❌ SD is NOT the same

❌ variance does change with edge

✔ closer to 50–50 = higher variance

�� Why their intuition is off
They’re thinking:

“The outcomes are still +1 or –1, so spread is the same”

But variance isn’t just about the range of outcomes—it’s about:

how often you land far from your average

Example:
65–35:

average = +0.30

outcomes are closer to expectation more often

55–45:

average = +0.10

outcomes deviate more relative to expectation

�� That increases variance

⚠️ Important nuance (this is where confusion comes from)
If instead the game were defined as:

win = +1

lose = 0

Then variance = p(1−p)p(1-p)p(1−p)
In that case:

still highest at 50–50

still decreases as edge increases

�� So even in that model:

their claim is still wrong

�� Clean rebuttal you could use

“That’s not correct. For a ±1 outcome, variance depends on the edge. A 55–45 coin has higher variance than a 65–35 coin because it’s closer to 50–50. Standard deviation is not constant—it decreases as the edge increases.”

�� Bottom line

✔ They are confusing EV with variance

✔ Standard deviation is NOT constant across different probabilities

✔ Maximum variance occurs at 50–50

If you want, I can give you a super short one-liner reply that hits them cleanly without getting into equations.

Here are a few clean one-liners you could use, depending on how direct you want to be:

�� Simple and neutral

“Standard deviation isn’t the same—variance is highest at 50/50 and decreases as the edge increases.”

�� Slightly more pointed

“That’s not correct—variance isn’t constant. A 55/45 coin has higher variance than a 65/35 coin because it’s closer to 50/50.”

�� Very concise

“Variance is maximized at 50/50, so a bigger edge actually lowers it.”

�� If you want to tie it back to poker

“Closer equity spots have higher variance—so a 55/45 situation is actually swingier than 65/35.”

If you want one with just a hint of edge without being toxic, I’d go with the second one ��

I mean…I just showed you the graphs

LOL "This discussion is getting a bit to nerdy for me" and than that load of gibberish

If you dont understand its fine but its not wise to rely on chatgpt
But im also not sure you asked it the right question, and if you did it clearly didnt interpret it correctly (a common thing for AI)
Because it says standard deviation isnt the same as variance... Correct, but nobody ever claimed that
And it says "They are confusing EV with variance" but thats exactly what you did :p

But you can just take one look at the confidence intervals in the posts above and see for yourself if you want to remain the statement that you have less variance versus whales.
Without getting too nerdy of course 😉

I put in your exact quote: "What you explain in your coin flipping example is not variance, its expected value/result.
The 65/35 coin gives you a bigger edge, but the standard deviation is exactly the same so it does not affect variance at all."

All the AI's say that variance decreases the further you get from 50-50 and increases the closer you get to 50-50. So are you going to just double down on that statement? It is very easy to ask yourself if you think I may have prompted it in a way that makes you look wrong.

Genuine question about your primedope example. I am not even necessarily trying to argue here. I am actually trying to understand and may be asking something stupid.

You show two scenarios. One with a 2.5bb/100 win rate and a 100bb/100 SD. The other with a 5bb/100 win rate with 140bb/100 SD. Where are you deriving the SD numbers from? Can't anyone can just plug in whatever numbers they want for this and see the results? Is the higher SD with the 5bb/100 win rate that is plugged in just based on your own assumptions about what the SD is in fishier games? Can anyone go to primedope and do the reverse? Say plug in a 140bb/100 SD for a 2.5bb/100 win rate and a 100bb/100 SD for a 5bb/100 winrate and show the results the same as you did here? If so then how is this actually proving something about whether variance is higher in fishier games?

You are right in saying a 50-50 flips are more variance heavy than a 49-51 flip or 20-80 flip or whatever as shown here

If I said it wasnt before I must have misspoken or was thinking I was referring to something else, my bad

I thought the point made (by you but also by others) was that playing versus fish is less variance heavy because you end up winning more often
And the point I was making is that isnt the case, but its generally not grasped properly
Like in this statement

I will have to say I am not very familiar with Lalibertes style of play but if he is somewhat of a typical whale than yes, variance is higher with 7 Lalibertes than with 7 Galfonds.
As is shown by the variance calculator examples.

To answer your question about the primedope examples and st.dev, which is not a stupid one to ask at all btw
SD numbers are as difficult as "true winrate" to call exactly, but you can get very close.
Every tracker like HEM or PT4 has a st.dev stat so you can look it up for your game and style of play to see how it differs. If you have a tracker with a lot hands it might give you some more information.

For myself I noticed that st.dev on zoom games where quite lower than on normal cash games, even though my normal cash game winrate was significantly higher and those games were fishier. And this is over a sample of millions of hands on both formats. So higher winrate did in fact combine with higher variance.

In theory yes, you could plug-in any old number but the numbers I chose are very realistic and taken from experience, they arent plucked out of thin air.
If you have many hands on different formats and table sizes you could also see that st.dev changes a lot depending on nlhe vs plo and fullring vs 6max vs HU.
For PLO this is because you generally play bigger pots and have more combinations that want to continue or shove which increases st.dev and therefore variance.
The same rhetoric goes for fishy players in comparison to regulars.

I hope this cleared something up.

Thanks for the response and I think I understand what you are saying here. I honestly am unsure if we actually have that much disagreement. Some of it might come down to just definitions and possibly not explaining well. I sort of already did this in an earlier message but let me try to give a rundown of how I see things and please tell me where you see I could be wrong. Hopefully this isn't boring you and I think its possible we both might learn something from each other. While I am certainly no trained mathematician I do think I have good intuitions about these things.

I think regardless of how you want to specifically define variance in poker it is incredible complex and nuanced with too many variables to really have clear cut answers. I may not have been accurate in an earlier message that said "variance is lower against worse players". I may have also been too confident in the Guy Laliberte example. I actually don't think you can say variance is higher or lower against fish in any blanket statement like that. I think it can be either or depending on so many variables. Some of the variables that come to mind are game type, buy-in sizes, if you always quit after getting to a certain size stack ( especially in PLO), both your own play style and the play style of the fish and other opponents, ect ect. So that being said I don't think there is any poker variance calculator that can give any concrete answer as to whether variance is higher against fish or not. Even if I accept they are doing their job correctly with the inputs they are given.

The reason why I started off with the weighted coin flip example is because that is the most basic model of variance without all of the variables that come with poker. In poker if you were simply playing only all-ins with the same stack sizes vs one opponent as a 65-35 favorite in one game vs a 55-45 favorite in another the variance would be lower in the game where you were always getting it in as a 65-35 favorite. Variance doesn't care if it is a coin flip or a PLO flip. At least I think that is the point I was trying to make there.

I think one of the main reasons why standard deviation per 100 hands can be higher against fish is because many fish types provide more opportunities to put your money at risk. If say for example at a table of tight pros you on average find opportunities to risk only 100bbs/100 but with a table of fish you find more frequent opportunities and are averaging opportunities to risk 200bbs/100 then we can say with certainty that you have a greater chance to lose more against the fish simply because you are putting more money at risk over that sample. If I flip the coin 200 times in the 65-35 scenario and only flip the coin 100 times in the 55-45 scenario I have a chance to lose more in total in the 65-35 flips. But my likelihood of winning is so much higher in that 65-35 scenario that I have difficulty saying there is more variance in that example. Even if the total SD of the 65-35 scenario is technically higher.

I will stop here because this is already too long. But there is much more to say especially when it comes to the real world application of this in actual poker games. And how it applies to what people like myself have experienced on ClubWPT.

Yea I think we might have hijacked this thread :p So I just say leave a brief reply and than let the RNG discussion continue, if they want to

Not every fish is the same so you cant say variance is higher versus a fish per se. But in general, versus whales (who generally like to play too many hands, make rando bluffs and continue weak etc) you play more big pots which means more variance. But not every fish fits this profile but generally it is true. And yes variance is also affected by stack size, game, table size, style of play etc as I have mentioned before. A reggy/fishy game is just one of those variables

There doesn't seem to be much activity in this thread anyways. The actual evidence that I presented earlier has been largely ignored. And this conversation is pretty relevant because people like Echemondo are insistent that people's perceptions of something being wrong with ClubWPT is because there is more variance against bad players. I can only speak for myself and say that is not the explanation for my experience on ClubWPT. My personal results are not the only factor in this. It is the very consistent pattern of the terrible players not losing at the rate they should. And possibly even winning at impossible rates like in the examples I provided.

Since the start of online poker there have always been a lot of dumb and unfounded claims on this topic. So I can understand why people might be automatically dismissive of there being something wrong here. But ClubWPT is not like other poker sites in the past. They don't even technically define themselves as being a poker site. They actually promote themselves as having the worst players and make it part of their marketing campaign. So yes there is a lot of incentive for them to have the bad players keep their money for as long as possible. Whether it is simply handicapping the players deemed as fish or in some way implanting their own fish to lure people in to playing. I still can't say for sure. But something is wrong here.

Even though I am barely playing on their anymore, I have noticed significantly less PLO tables running the past few weeks. Again I won't pretend to know what that means but a sudden decline like this is interesting and must mean something.

In regards to "significantly less plo tables running"

The density of VIP's has notably dropped. Myself, one of my very successful hs reg friends, and both of our friends/students have reminded ea other just how high rake is and how a game that's "just weak regs and a weak but not vip fish" on here may be not nearly as good as that same description on much lower rake. I personally know myself and many of the regs playing less given the rake and declining density of VIP's to justify it. I do think we've overshot it but I would suspect a ton of regs are smart enough to start picking up on this.

I read back quite a bit in ITT but im afraid I never saw anything close to "actual evidence", was it posted elsewhere?

I did read stories about bad beats and the same players winning while being very bad, but I can read those about literally every poker site.
But they always are just that, stories. And im sure you understand that stories without any foundation arent actual evidence...

I think he's referring to his first post in this thread at https://forumserver.twoplustwo.com/showp....

OK yes I have read that.
Its all "I saw this" and "I saw that" but nothing for us to see

Nowhere near actual evidence, in fact I can find those stories with the same lack of proof about any site in the riggies thread

Well this is another follow up post with somewhat detailed stats about the second extreme godmode run I saw in two days. I included a few screenshots of the stack sizes at different hand intervals. Then I made an attempt to show how unlikely it was based off of some AI calculations. Then someone else who I don't know chimed in corroborating these events. I literally took over 300 photos on my phone of the all the hands played in the first god mode run in the first message. I have been meaning to go through all of them and figure out showdown vs non-showdown wins because that could be even more damming.

So I provided evidence of two separate millions or billions to 1 coincidences happing in favor of the same player types in he same games on back to back days. I am not sure what else I can realistically do on site that has no hand histories. I have not heard any retort as to why this is not evidence that something is wrong. If someone has a logical explanation for it then of course I am open minded for my conclusions to be wrong.

Tinfoil hats are on, baby

ive never once thought a poker site was rigged but something off about wpt gold

the player pool is suss and the rng is beyond suss

and i cant even download hh to provide proof of said sussness

anyone know who owns it?

I have been unable to find clear answers as to who owns it. Grand Cru Concepts is the company but I don't believe we know who owns that company. I believe the WPT brand we all know just licenses that name to Grand Cru Concepts. They do not own or run ClubWPT Gold.

yooo what's up DO IT! LETS GOOOOO!!!!

Ran into him again tonight. Lost to him in two back to back hands where I flopped top set on both. In the last hand, he called a 40bb bet on the flop with absolutely nothing and get it's all in on the turn to hit runner runner straight. I left the table right after. Here he is after just 45 minutes at that table. This is the third time I've run into him and every time it's like this, the first time he had over 4000bb's playing pretty much every hand just like he is tonight.

Less than an hour later - I'm posting these as it's happening.

now that's pretty sus

I’m up around 45k on the site but I do not believe it is legit. Can you win money if you play well? Sure. Are you being cheated by the house and/or other players? Probably.

Something is off, I’ve played millions of hands mainly HU Cash, I crushed for awhile but now players making big fundamental mistakes with high VPIP low SPR and small samples are coolering me in 3B/4B pots flop set over set, top two getting cracked by straights on turns in rivers that villain shouldn’t have in their range and SPR is 1-1 1/2. It’s to the point I know the set up before it happens and can say “I know he has X, but I can’t fold, folding is a mistake” call and see my instincts were right. Idk I don’t want to be this “fish” online is rigged guy” but…

played over 8m hands MSNL+ online, lots of 6max but a good chunk of 9max and HU across many sites. Never have I questioned a RNG until now lol. Just watching myself and other good regs seemingly on the wrong side of things vs fish a disproportionate amount of the time. Winning at a good clip on this site over 120k hands in the 1/2/4 - 2/4/8NLH 8 max games but I am suspicious as it gets