The great "Poker is rigged" debate - Collected threads edition - Page 284 - Internet Poker

**Markusgc** · 06-05-2009, 03:06 PM

Quote:

Originally Posted by K13

"Maybe you should stop playing with AA, KK, QQ because all they do is lose"

never heard that at a live card room. that's all the proof I need!

LVGambler · 06-05-2009, 03:33 PM

Quote:

Originally Posted by Arouet

I think K13 has said he plays mainly SNGs. I wonder what the makeup is of rigtards: cash vs. tournaments. I'm betting more rigtards play tournaments, SNG's etc. where the immediate consequence of a bad beat can mean elimination. Where in cash you just move on to the next hand knowing that if the other player stays around long enough you should have a chance to return the favour.

**Markusgc** · 06-05-2009, 03:37 PM

Quote:

Originally Posted by Arouet

Where in cash you just move on to the next hand knowing that if the other player stays around long enough you should have a chance to return the favour.

not if you have your entire bankroll on the table you can't.

toltec444 · 06-05-2009, 03:41 PM

Great laughs up here.

toltec444 · 06-05-2009, 03:42 PM

Quote:

Originally Posted by KingOfFelt

There should be a rigtard phrase or word(s) of the day. This one should win for today imo. It's rigtastic!

Great laughs up hre. Keep the fun comments going on, you should be humourist.

**Markusgc** · 06-05-2009, 03:48 PM

Quote:

Originally Posted by toltec444

Great laughs up hre. Keep the fun comments going on, you should be humourist.

burden2 · 06-05-2009, 03:51 PM

"rigtard" as I would be called here. I prefer skeptic.

I started my painstaking hand analysis. So far I am just looking at pf all-ins (this is all heads up) and just have a few days of data. My record is 48-56, my average equity 54%. I have to admit, though its running bad, I thought my record would be much worse and my equity a little higher. This does include 3 cases of QQ against a higher pair (bringing down my average equity). There were no case of my going all in preflop with a bigger pair than my opponent. I have about 45 days worth of data, looked at 5 so far.. I will be looking at big pairs (1010+), and flopping two pair or better, which all seem to lose a lot more often than they should given that I am playing heads up. I know that I am a net loser when flopping trips, and with KK (huge loser with KK heads up).

So far I am surprised at the mildness of the running bad but I don't think I have hit a bad stretch in my data yet.

**Markusgc** · 06-05-2009, 03:58 PM

Quote:

Originally Posted by burden2

"rigtard" as I would be called here. I prefer skeptic.

that's healthy, but different from a Rigtard. the difference is a skeptic might lean towards that belief but would reserve judgment until verifiable evidence is presented. Rigtards just go on a "feeling."

Quote:

So far I am just looking at pf all-ins (this is all heads up) and just have a few days of data. My record is 48-56, my average equity 54%. I have to admit, though its running bad, I thought my record would be much worse and my equity a little higher. This does include 3 cases of QQ against a higher pair (bringing down my average equity)... I have about 45 days worth of data, looked at 5 so far...

sick sample size

batair · 06-05-2009, 04:49 PM

Quote:

Originally Posted by K13

Its so ****en bull**** how you can have KK, QQ lost to Ax hand after hand after hand.

That's actually the biggest flaw in the PS RNG along with flush draws, you play somewhere else and the RNG plays completely different. So which is "random"?

Quote:

Originally Posted by K13

you can't even ****en play poker on PS even when you know what people have.

Another KK lost to Ax.

EDIT

and another and another

stop ****en defending this ****. Its beyond ****en probabilites.

play Ax fold AA KK QQ you will thank me later, i would like a cut of your profits for my excellent tip.

spadebidder · 06-05-2009, 04:51 PM

Quote:

Originally Posted by burden2

So far I am just looking at pf all-ins (this is all heads up) and just have a few days of data. My record is 48-56, my average equity 54%. I have to admit, though its running bad...

Not all that bad for the sample size. In a sample of this size, your chance of losing >=56/104 with 54% average equity is about 7%. That means every time you look at a 104 hand trial with 54% average win equity, 7% of those trials will have 56 or more losses. When you get to 500 allins it should be pretty close to equity. Make sure you don't cherry pick sets from a larger population, you must use all that you have records for, or pick smaller samples totally randomly. Even picking the smaller sets sequentially as you appear to be doing, introduces biases and isn't representative of the whole population.

DMoogle · 06-05-2009, 05:01 PM

Quote:

Originally Posted by spadebidder

You're quite wrong here. We don't need to know any hole cards at all to analyse the board and determine if the community cards follow their expected distributions. Having all hole cards would allow a determination of whether all 1326 possible hands are dealt evenly, but that isn't of much use by itself anyway, as the site could switch hands between targeted players and not change the hole card distributions. Knowing all the seen community cards and all the hole cards that made it to showdown, is sufficient to do a very comprehensive (and conclusive) analysis of whether the deal is fair.

The board might be *slightly* weighted towards low cards, because the more high cards that are out in players' hands, the higher the chance of there even being a board.

spadebidder · 06-05-2009, 05:12 PM

Quote:

Originally Posted by DMoogle

The board might be *slightly* weighted towards low cards, because the more high cards that are out in players' hands, the higher the chance of there even being a board.

That is a well-known effect and it shows a predictable slope, with the difference being very small. In large samples you'll see dueces hitting the board spots about 7 in 10,000 more often than Aces, and the rest are sloped in between. Knowing that, it is easy to evaluate other skews or discrepancies.

Here's a small sample from 2 million hands. This is also part of my larger analysis that isn't ready to publish:

Code:

[Card]   Flop1   Flop2   Flop3    Turn   River    Mean

[ 2c ]  1.943%  1.925%  1.964%  1.953%  1.954%  1.948%
[ 2d ]  1.958%  1.939%  1.931%  1.965%  1.952%  1.949%
[ 2s ]  1.933%  1.957%  1.947%  1.968%  1.965%  1.954%
[ 2h ]  1.963%  1.931%  1.944%  1.983%  1.956%  1.955%
[ 3c ]  1.929%  1.936%  1.947%  1.961%  1.961%  1.947%
[ 3d ]  1.963%  1.947%  1.947%  1.946%  1.949%  1.950%
[ 3s ]  1.945%  1.946%  1.959%  1.939%  2.004%  1.959%
[ 3h ]  1.960%  1.942%  1.952%  1.962%  1.927%  1.949%
[ 4c ]  1.952%  1.922%  1.938%  1.952%  1.937%  1.940%
[ 4d ]  1.919%  1.931%  1.936%  1.946%  1.982%  1.943%
[ 4s ]  1.944%  1.953%  1.950%  1.946%  1.947%  1.948%
[ 4h ]  1.939%  1.945%  1.941%  1.939%  1.959%  1.945%
[ 5c ]  1.953%  1.917%  1.932%  1.929%  1.923%  1.931%
[ 5d ]  1.947%  1.935%  1.915%  1.934%  1.940%  1.934%
[ 5s ]  1.950%  1.954%  1.940%  1.955%  1.936%  1.947%
[ 5h ]  1.949%  1.946%  1.921%  1.936%  1.954%  1.941%
[ 6c ]  1.952%  1.929%  1.926%  1.944%  1.916%  1.934%
[ 6d ]  1.942%  1.922%  1.943%  1.934%  1.949%  1.938%
[ 6s ]  1.911%  1.927%  1.932%  1.920%  1.953%  1.929%
[ 6h ]  1.934%  1.938%  1.940%  1.946%  1.928%  1.937%
[ 7c ]  1.924%  1.912%  1.935%  1.916%  1.909%  1.919%
[ 7d ]  1.933%  1.935%  1.944%  1.932%  1.925%  1.934%
[ 7s ]  1.905%  1.932%  1.926%  1.926%  1.939%  1.926%
[ 7h ]  1.931%  1.939%  1.924%  1.896%  1.952%  1.929%
[ 8c ]  1.927%  1.917%  1.917%  1.916%  1.942%  1.924%
[ 8d ]  1.938%  1.931%  1.924%  1.949%  1.924%  1.933%
[ 8s ]  1.899%  1.898%  1.905%  1.929%  1.919%  1.910%
[ 8h ]  1.924%  1.938%  1.936%  1.906%  1.936%  1.928%
[ 9c ]  1.915%  1.936%  1.917%  1.920%  1.924%  1.922%
[ 9d ]  1.899%  1.927%  1.916%  1.932%  1.914%  1.918%
[ 9s ]  1.917%  1.918%  1.936%  1.929%  1.928%  1.926%
[ 9h ]  1.926%  1.929%  1.910%  1.932%  1.917%  1.923%
[ Tc ]  1.908%  1.919%  1.932%  1.903%  1.934%  1.919%
[ Td ]  1.917%  1.921%  1.927%  1.917%  1.925%  1.921%
[ Ts ]  1.929%  1.923%  1.920%  1.920%  1.881%  1.915%
[ Th ]  1.894%  1.898%  1.926%  1.932%  1.926%  1.915%
[ Jc ]  1.915%  1.930%  1.913%  1.898%  1.883%  1.908%
[ Jd ]  1.907%  1.937%  1.918%  1.890%  1.907%  1.912%
[ Js ]  1.908%  1.912%  1.911%  1.909%  1.896%  1.907%
[ Jh ]  1.890%  1.906%  1.921%  1.913%  1.924%  1.911%
[ Qc ]  1.898%  1.896%  1.918%  1.894%  1.900%  1.901%
[ Qd ]  1.911%  1.898%  1.926%  1.910%  1.907%  1.911%
[ Qs ]  1.925%  1.921%  1.909%  1.911%  1.896%  1.912%
[ Qh ]  1.913%  1.891%  1.914%  1.904%  1.893%  1.903%
[ Kc ]  1.915%  1.893%  1.896%  1.896%  1.883%  1.897%
[ Kd ]  1.921%  1.920%  1.888%  1.898%  1.944%  1.914%
[ Ks ]  1.905%  1.914%  1.894%  1.905%  1.884%  1.900%
[ Kh ]  1.898%  1.921%  1.899%  1.915%  1.861%  1.899%
[ Ac ]  1.884%  1.890%  1.882%  1.838%  1.866%  1.872%
[ Ad ]  1.889%  1.881%  1.871%  1.847%  1.832%  1.864%
[ As ]  1.869%  1.878%  1.876%  1.899%  1.869%  1.878%
[ Ah ]  1.878%  1.896%  1.864%  1.862%  1.866%  1.873%
[ ALL]                                        100.000%

The more interesting stuff I'm working on is the board types analysis, with the frequencies of about 30 board types that affect hand outcomes.

Again the sample above is only two million hands, so don't try to read it too granularly at this point. I just wanted to illustrate the effect you refer to.

DMoogle · 06-05-2009, 05:20 PM

Is that all from FR? There'd be a significant different between a table of 9 players and a table of 2 players.

spadebidder · 06-05-2009, 05:25 PM

Quote:

Originally Posted by DMoogle

Is that all from FR? There'd be a significant different between a table of 9 players and a table of 2 players.

No, that example is mixed. The skew for heads-up is bigger, with dueces showing up on the board about 14 times per 10,000 more often than Aces. Still a very small effect. We're analysing by number of seats and a lot of other variables.

**Markusgc** · 06-05-2009, 05:26 PM

Quote:

Originally Posted by spadebidder

That is a well-known effect and it shows a predictable slope, with the difference being very small. In large samples you'll see dueces hitting the board spots about 7 in 10,000 more often than Aces, and the rest are sloped in between. Knowing that, it is easy to evaluate other skews or discrepancies.

Here's a small sample from 2 million hands. This is also part of my larger analysis that isn't ready to publish:

...

The more interesting stuff I'm working on is the board types analysis, with the frequencies of about 30 board types that affect hand outcomes.

Again the sample above is only two million hands, so don't try to read it too granularly at this point. I just wanted to illustrate the effect you refer to.

BORING!!!!!