is on-line collusion sound?

open question:


my friend, who got laid off a busted broadband provider, wants to put a poker-playing program on my computer and several other people's. i'm not really sure how it works, but apparently all the computers read gif files for composition and instant-message each other, and determine the probability of a crap hand drawing out... he says it can be done, and quite quickly.


he may know how to program, but i am wondering if his poker is sound. suppose you are sitting in five seats at once in an on-line poker table. can you see enough extra of the deck to really improve expectation on pre-flop calling decisions on most hands, and at the same time have enough opponents left to not just be paying the rake to bet against yourself? would it improve your bayes theorem enough on later rounds to make up for the fact that most of the time you don't even have an opponent to read?


-bz

Seems to me you would have to do an awful lot of betting against yourself to keep up the image, and keep the "outsiders" in the game.


- Louie

The previous post was an imposter. At least its polite.


The advantage of taking 2 seats in the same game is NOT so you see a couple more cards and can therefore better calculate your hand odds. Its [1] so the weak partner can raise the strong partner trapping others for multiple bets, and [2] much more important its so the good enough but weaker hand can fold when partner is playing a stronger hand: if I flop top pair partner should fold 2nd pair.


Fortunately on-line programs monitor for these sorts of activities and I am confident they will ferret out all but the most subtle cheats. Having said that, the original idea of seeing more cards to slightely improve your calculations WOULD be impossible to detect; but then again it would rarely matter and when it does it would be for only pennies.


- Louie

how do you mean imposter?


you mean the first thing you said was just a stand-in for what you meant to say?


at least you were polite:-)


is it polite to put, like, a placemarker for your next post?


were you meaning to make some kind of example? are you trying to fake me out?


i don't get it. please explain imposter, so i can do it too!


bz

also, the plan has been explained with the following examples:


1- i have A5s, in my callable hand, but also 1 ace, 3 cards of the same suit, and a 4, in my folding hands. i fold all hands.


2- i have high pair against a possible flush draw or high pair worse kicker. if i folded the same pair card in a hand with a worse kicker, but folded no cards of the onboard flush suit, his relative odds of having me beat have to be triple.


i can't remember the other examples right off, but you get the idea. it seems that in most pair-on-pair showdowns, you can increase your odds by much more than pennies. it seems you can dramatically change your preflop calling decisions as well as your river raising decisions when you get up to four seats, no?


i want sklansky's opinion.

"PENNIES"???


I L-O-V-E PENNIES!!


YOU DON'T PLAY MUCH BLACKJACK, DO YOU!

OK, Louie, I here's a proposition, which I am sure you will take me up on! Either:


A) You meet me for a heads-up 20-40 freezeout where 1) I pay you 23 copper pennies each time a hand is dealt, and 2) I get to see the 7 cards on the bottom of the deck after the hole is dealt before deciding whether to call the flop (but you don't), or


B) You sit at a 20-40 HE table with me and 4 other people. Before each hand is dealt, we each give you 4 pennies. Then, whoever stays in among us, we all get to sort through and keep an eye what cards are in the muck pile - but you don't.


Where do you want to meet?


-Cal

The point is the rake. If you are playing 5 of the 10 hands at a table which is raking $3 from each pot, you will have to pay on average $1.50 per hand instead of $0.30 in rake. In addition, you will only actually win money whenever you go up against the 5 hands you're not playing. Therefore, if you have a win rate X when you are playing one of the 10 hands, your win rate without the rake is then approx. X+$0.30, but when you pay 5 of the hands, you will only have a win rate of 5X/9 - $1.20 after the rake. X - (5X/9 - $1.20) = 4X/9 + $1.20 is what the extra knowledge would have to increase oyur win rate by. This seems like a bit of a stretch.

I don't understand this post. The first response above with my name dated 29 Sep 11:28pm was not posted by me. Does that answer your question?

With two hands face up, on the flop the flush draw has 9 outs out of 45 unseen cards or 36:9 or 4:1 against making the flush on the next card.


If you have seen 8 other cards and none are of that suit then there are 37 unseen cards and 9 outs or 28:9 or 3.11:1 against the flush getting there on the next card. Hardly "tripple".


The difference is reasonable but KNOWING it only matters when other factors suggest you need between 3:1 and 4:1 to call: the rest of the time knowing the missing cards does you no good since you are going to bet or to call regardless. And if you make a "mistake" because you didn't look at the other cards it doesn't cost much. So in this example it doesn't matter often and when it does it doesn't matter much; thus the reference to "pennies".


Better examples abound such as when considering calling with a small pair yet one of them is dead. Or not fearing an Ace flop when you have KK because you know 2 aces are gone. Or NOT drawing to a flush since its so dead.


- Louie