He said: "On the other hand we have you saying:

I've proven it's rigged, you're all too stupid to understand, even though I don't know what standard deviation means and couldn't even be bothered to google it.
I'm even more unlucky if I use some other data."

In his impersonation of me, he based his argument on personal accusation (e.g. not capable of knowing what STD means), while taking away focus from, or totally avoiding logical argumentation to actually prove that I e.g. don't know about STDs. So that is ad hominem fallacy.

My logics is objectively superior to what others have provided in terms of arguments. He is taking away focus from my logics by refering to me as feeling superior, to the point where his logical fallacy even convinced you. That is ad hominem fallacy.

It's not an ad hom logical fallacy as he's not making a logical argument there. He was mocking you, yes, you may even argue unfairly, but ad homs are about argument structure and he wasn't making an argument.

An ad hom is not defined because it takes away focus from your argument - the ad hom logical fallacy refers to using a personal characteristic to refute a person's argument. Again- I suggest you re-read the wiki that you yourself cited:

Claiming that I don't know about STD's is an objective prediction. Doing that while mocking, is a personal accusation. Creating such assosiations between personal accusation and objective prediction is ad hominem fallacy.

Well that's exactly what he's done. He refers to me as superior, and uses that to refute my logics. He refers to my logics slightly hidden by saying "when you talk down to everyone about a subject". Then he refers to my characteristics in order to refute mentioned logics.

It is a claim not a logical argument.

Again not a logical argument.

Also not a logical argument. If he didn't make a logical argument he can't have made a logical fallacy DUCY?

No, he doesn't call you superior to refute your logic. He suggests that you feel superior without merit. He doesn't say your argument is wrong because you feel superior, he simply says that you are not justified in your superiority complex.

I considered logics to just be objectively related. However, since you insist on the exact meaning of logics it becomes irrelevant, since the definition of ad hominem is independent of the word logics.

ad hominem: argumentum ad hominem, is an argument made personally against an opponent instead of against their argument.

Dictionary.com gives the following meaning of "argument":

1.
an oral disagreement; verbal opposition; contention; altercation: a violent argument.
2.
a discussion involving differing points of view; debate: They were deeply involved in an argument about inflation.
3.
a process of reasoning; series of reasons: I couldn't follow his argument.
4.
a statement, reason, or fact for or against a point: This is a strong argument in favor of her theory.
5.
an address or composition intended to convince or persuade; persuasive discourse.

As we see, argument only needs to be an objective statement.

Thus my ad hominem accusations are correct.

I quoted the relevant part. The wiki is not terribly clear I agree though. Here is a better example:

http://www.nizkor.org/features/falla...d-hominem.html

So you test one thing and perform somewhat extensive calculations and find you run slightly bad, but not so bad that it is at all suspicious, remarkable, or noteworthy .... so you then say, without doing any calculations, that you run bad in other areas as well and so adding those it is suspicious, remarkable and noteworthy, but you aren't going to do those calculations?

What was the point in doing the 1st set of calculations in the first place if you are just going to assume the rest of your results that weren't measured in that first calculation show that the deal is rigged?

LOL that Faen carries on for several more posts about ad hominem, refusing to see that he's wrong - seems like a familiar theme with him. Probably still thinks I was "unintentionally lying" as well.

Exactly. I asked him why he doesn't just post all of his data, and he claimed that he had. I can't recall him posting much data of substance aside from his analysis of a couple different hand types, but he was in full swing of his condescension and calling everything an ad hominem at that point, so I decided I couldn't be bothered to ask him to link the post(s).

Douche itt....

lol addhomaments

That was a great link that dealt with exactly what I was referring to, thanks. That said, have you read it, because it seems to somewhat counter your statement in the bold.

"We wanted to come out of this saying one of the following things: "There were players on Cake who could see other players' hole cards, and Cake has compensated affected players" or "Given the data we've received, we believe that there is a high probability that nobody cheated significantly on Cake network by exploiting the security vulnerability". Unfortunately, identifying players who can see another player's hole cards is extremely difficult. (See this thread in which both Stars and FTP security failed to identify cheaters by analyzing hand histories in spite of knowing with near certainty that they were cheating.) In order to have such confidence, we have to be sure that our method is extremely good at catching cheaters. In order to do this, we have to test it on a large sample of hand histories with hole cards exposed played by known cheaters of varying levels of sophistication."

Correct me if I am wrong but it seems these individuals, with the assistance of the pokersite, in this case Cake, devised a method or methods to detect Superusers. So theoretically if the pokersite, was the one cheating, how could a project like this work either without the assistance of the pokersite or if they claimed to "help", with the "rigging" pokersite tainting the investigation?

So in order for the "rig" to be exposed to be outside the acceptable range of variance, one would need the full cooperation of the pokersite using the "rig", right?

I did calculations for sets, overpairs and top pairs in general. I were unlucky in all of them. I even posted screenshots of those before I posted of AA stats.

As usual you provide no intelligent arguments but provide empty criticism.

It seems that you are talking about 2 different things: (1) rigging the deal and (2) people cheating by being able to see all hole cards ... but it seems that you are talking about the them as if they were the same.

That pretty much sums up what I would've said. To expand tough, superusers would be detectable by their winrates. No one on earth is going to use that power to win 5 bb/100 or something, especially not when you can have pretty normal stats otherwise and still absolutely CRUSH winning 25+ bb/100.

A non-superusing rig would be detectable because any change to random cards makes them non-random, and non-randomness is reasonably easy to detect.
Redo the calculations excluding hands you folded, just for funsies.

I'm not an expert on statistics, but if we're talking solely about having players for the house that could see all hole cards, I think that could be done in a way that would be very difficult to detect. But it would require a very intelligent team behind it to make sure they were only using their information in ways that wouldn't draw attention to themselves. When players suspect something and start pooling data, what might normally seem like innocent plays could get noticed if a pattern started to emerge.

And there's the problem IMO.

I seem to recall a gang of superusers getting caught by their statistics. Funny, that company is out of business now too. Weird how that works.

Fair enough so I will attempt to clarify. If it's possible for a Superuser to keep their stats within a difficult to almost impossible to catch statistical range without help from the pokersite, can't the same thing be done for a "rig"?

Theoretically, if a pokersite was using Superusers to cheat gamers, for whatever reason, it seems according to this project, it would have been difficult, if not all but impossible, to catch them, without assistance from the pokersite. Likewise, if a pokersite is utilizing a "rig" that works on the same principles of being difficult, if not all but impossible to catch without assistance from the pokersite, by having the stats fall within a perceived acceptable range, how would the "rig" be caught?

What's even more corollary, imo, is the way they state that even when it was known that cheating was occurring, meaning is was clearly detectable by some form of perception, on Stars and FTP, it couldn't be proven with HH. Which seems to fly in the face of this oft repeated notion in this thread that a "rig" can't be perceived by players and currently undetectable through HH.

I don't see the sets one, but are the following two posts what you mean for top pairs?



If so, I don't see any calculation regarding how likely such a thing is to occur whether in percentages or using standard deviation.


And backing up: Your initial range assumption that you use to determine that you "should win 80% of the time" is flawed. You're assuming that every one of your opponents in those hands had the same range of hands despite the fact that people play differently and despite the fact that each hand presented with different pre-flop action and dynamics. And furthermore, the range that you assumed is a pretty wide range at that of about 34.5% of hands.

Also, you are using one type of top pair (a big pair [a Q] with an A kicker) and one specific board to determine the percentage you "should win" and then you are applying that to various different types of top pairs on various different types of boards such as QT on a T high board or 77 as an overpair.


I took the QT top pair hand you posted and plugged in the same range you used for your opponents (I think I used the same range correctly, it is hard to see your screen shots ... correct me if I missed something) and your winning percentage is considerably different ... only 68%:

338,580 games 0.000 secs 67,716,000 games/sec

Board: Td 5d 8s
Dead:

equity win tie pots won pots tied
Hand 0: 68.385% 67.25% 01.14% 227679 3860.00 { QcTs }
Hand 1: 31.615% 30.47% 01.14% 103181 3860.00 { 22+, A2s+, K8s+, Q8s+, J8s+, T8s+, 98s, 87s, A4o+, K9o+, Q9o+, J9o+, T9o }



I did the same for your 77 hand and you are only 60.5% to win:


2,358,180 games 0.020 secs 117,909,000 games/sec

Board: 5s 4c 2c
Dead:

equity win tie pots won pots tied
Hand 0: 60.539% 59.75% 00.79% 1409085 18540.00 { 77 }
Hand 1: 39.461% 38.67% 00.79% 912015 18540.00 { 22+, A2s+, K8s+, Q8s+, J8s+, T8s+, 98s, 87s, A4o+, K9o+, Q9o+, J9o+, T9o }



One more ..... On the KQ hand vs. the range you set you are not 80% to win, but rather are only 62.8% to win:


329,670 games 0.010 secs 32,967,000 games/sec

Board: 5c Jc Qc
Dead:

equity win tie pots won pots tied
Hand 0: 62.878% 60.39% 02.48% 199103 8185.50 { KdQs }
Hand 1: 37.122% 34.64% 02.48% 114196 8185.50 { 22+, A2s+, K8s+, Q8s+, J8s+, T8s+, 98s, 87s, A4o+, K9o+, Q9o+, J9o+, T9o }






Your range assumption is flawed and you calculate an expected winning percentage based on 1 type of top pair hand and 1 specific board and then apply such expected winning percentage to various different types of top pair hands on different boards and, therefore, the amount you calculate as the percentage of the time that you "should win" is flawed and, therefore, the rest of your calculations are flawed and, therefore, the result you arrive at is flawed.

No. Two totally different things. One refers to statistics of a random process (the deal), which has a very well known expected distribution. The other refers to statistics of human behavior, which have a very wide range of expected distribution. The cheater just needs to stay away from the extremes, but that isn't possible when skewing a random process. It will always become extreme if consistently repeated.

That's what I am asking about, concerning the bold but using an AI program, algorithm or whatever to accomplish the same thing as a Superaser in the sense of being difficult to catch but to manipulate cards.

This is all off the cuff but for example a if there are two hands where the flop, turn and river will have the exact same 5 cards, would the order of those 5 five cards make any statistical difference in the HH?

So if the hero has AA, is called with 39os, the flop is 933, turn A, river Q, that is the same deal statistically in HH if the flop were QA3, turn 3, river 9. However the order of those cards would likely illicit different outcomes.

No, it absolutely is not the same,and the difference is detectable. You are skewing three different streets of the deal away from random. Every street should be random on its own and eventually have a statistically equal occurrence of every card in the deck (as well as the sequence being random in each position individually too, which is measurable).

Strike two for your terribly wrong assumptions, just in your last two posts.

Edit: actually 3, I see where you claim players can perceive more information about the dealt cards than what appears in hand histories (of the same cards seen by players). Do you know how ******ed that is?

It depends on the statistical analysis - what you were testing. So yes it could be shown.

To hark back to your previous posts, I think it is agreed that a very careful superuser would be extremely difficult if not impossible to detect. There have been superusers detected; some such as nionio & potripper that played absurdly making huge and implausible hero calls, and others whose winrates generally or in certain situations were outside normal bounds, or whose pattern of play was unusual.

It is definitely possible though that only the 'bad' superusers have been caught, and that sophisticated ones have got away.

I gave some thought to what Monteroy had to say and he has a point.

Far be it for me to not oblige so ...to the rigs, I give you my graph for the last 731 games, no filter.

What I don't get is ...no bonus to clear and yet ...no rig? Huh...

And I've been with Black Chip Poker for about a year and yet ...no rig?

When does this rig kick in again, guise, I can't seem to find it!!

This is across two networks, mind you (BCP used to be Merge, now it's Yatahay) so I guess I'm dodging multiple rigs! I bet they somehow know that I'm really quux on 2p2 and so they turn it off because they're afraid if they rig me, I'll post my database.

That must be it.

Forgot to add rakeback to the chart. Oops.

How exactly or could you point me to a link that explains it. I referring to a "rig" in that example, not a Superuser, if that wasn't clear. I understand that the odds for winning the hand are different for which order the cards come out but not the overall result of the hand. So if AA beats to 39os 87% of the time or somewhere within that range, the order of the community cards wouldn't effect that outside an acceptable range, would it?

To put it another way, if the RNG or whatever deal method is not actually generating cards in real time at the moment of the deal for a 9 player table but is instead dealing the next "random" 23 cards (2 hole cards x 9 players + 5 community cards) that are already generated in que but manipulating the players the hole cards are dealt to and/or the order of the community cards. How would that be distinguishable from being random using HH?

I meant manipulating the order in which the holes card are dealt to players, so players end up with different hole cards than they would have had if the cards had been dealt in the originally generated order.