And in 1970 it was accepted wisdom that a chess-playing computer would never beat the best human players, or at least not for the next 100 years.

Plain and simple bots are very wrong in a game which is human vs. human.

The continued advancement of bots has the capacity to destroy online poker due to many factors that have been mentioned elsewhere in the thread.

The only thing I agree with is that hud's should be banned.

Bots are a world away from hud's tho and it doesnt matter that most botters lose - what does matter is essentially you are working towards the destruction of a game we all know and love.

I also love how you all try to justify your actions - sounds like your trying to alleviate a guilty conciense for something u, I and everyone else knows is a POS thing to do. As far as Im concerned you should all go live in the world of ultimate bet where morality is a non-issue.

I'm glad someone got my point. I fear that my arguments will fall on deaf ears because what we have here is retroactive justification.

Bots will only get better. They won't get worse. Do you understand the nature of bots? You think that they will never be a threat because you are

[x] a fool
[x] and/or a botter

What in god's name are you talking about? Some of the most genius people are self motivated and self employed. This is a huge fallacy and actually quite ignorant.

Doubt it, Poker Pros bring new players (rake.) Bots do not. Not many people say "gee I heard there is this site where you can play for real money against a computer at NLHE."

lol @ the rest of your post, it isn't even worth my time to make you understand your stupidity.

What I think these people don't understand is that people who oppose bots are not talking about the now, but rather the later. We aren't talking the present, but the future. Where does this lead the exponential enigma known as "technology." Once precedent is set, it will be hard to reverse. The wheels will already be in motion. I'm not about to suggest that it is 100% preventable, but welcoming it with opening arms is -EV IMO. If you can't see a few steps ahead it is no wonder the chess analogy flew right over your head.

Yes, and I exlpained it very simple how bots are irrelevent not only now, but for the future too. Look at my previous post. The examples I gave are so simple, and the logic is so simple, that if you cannot understand this - then it is really useless to discuss it further (I'll just wait for some people with higher IQ to pick up on these arguments).

Repositng, and this addresess the chess, and some other of the posts which ironically came after mine.

I don't know why we expect the younger players to see that bot's, multi-accounting and collusion are cheating when they grew up playing video and computer games and the people who designed them made millions more by selling cheat codes to help people beat the game. To the young people any game is made to be beaten and any edge is a fair one, as long as everyone has a fair access to it.

This statement shows exactly that mentality:

Crude, bots, on one hand, and collusion, multi-accounting, database sharing + poker tracker & HUDs on the other hand are different things.

While bots, on the one hand, does not provide unfair advantage (hence they are cheating only is they are made against the ToS - like other things, over there, e.g. that you should not tell something bad for the pokersite in a forum). Bots do not cheat with respect to the game.

All other things, on the other hand, provide actual unfair advantage from gametheory standpoint.

That's pretty much a difference, for both young and old players, although one need to be capable to understand this.

As with regard to the statement, when a game gets sufficiently solved (like chess, blackjack and so on) - the outcome I talk about is immenent.

That statement is so wrong. Any AI that is used to make decisions is cheating in my book. Bots are no different than HUD's. It is something that we use that can always either help us make the right mathmatical decisions or can make them for us.

Can you beat something that always makes the right call mathmatically? Maybe in the short run, but in the long run the AI will always win.

First off it's laughable that you insult someone's IQ just because they don't agree with your skewed view of cheating. Lets take a new approach since you are not getting the current one.

Bots don't get tired. They can play hours upon hours without making a "mistake." They will always stick to their coding no matter how long they have been playing. And thus, we have another unfair advantage. Humans get tired, their decisions start to get outside of their game plan, ect. This is indisputable, and if you can't see how this is NOT GOOD FOR POKER then you are probably an idiot. Plain and simple.

Lets play a mind game. Assume we have two players, one is AI another is actually a human being. They are both marginal winners with the exact BB/100. They sit at the same table. They play for 12 hours because the bot can and because the human loves to play poker. Assume they encounter the same amount of "variance" over that 12 hours, if we assume all other factors are held constant (the Latin/economic term is "Ceteris Paribus"), who will we expect to win more in that session? - the BOT. Also assume from that point forward that both the human and the AI log 12 hours per day, 5 days a week. Whose BB/100 will start to rise relative to the other player? The bots, because the bot isn't affected by "the human condition." The bot will not make bad decisions as it gets "tired" because it DOESN'T GET TIRED. It also doesn't go on TILT which every human does at some point in their poker playing.

Here are three points to marinate on:

1) Bots don't get tired, thus they have an inherent advantage relative to players who need sleep to function
2) Bots can put in more volume, which translates to more profit. This is closely related to the first point.3
3) Bots don't go on tilt, thus they have another advantage over humans

Botfanboys, the thing is I can list lots of reasons why you shouldn't use bots to play humans for real money, but you really can't provide one solid reason why they should be. You counter with weak arguments that is more akin to two little kids saying "Well he did it, so I should be able to do this too" That **** is weak and indicative of immaturity.

Regards,
Chris
Software Engineer
http://www.accuvant.com/

Personally I think bots are the worst kind of cheating. I will make an analogy using the popular game Counter-Strike.

Aimbot : pokerbot :: wallhack : PokerTracker

Aimbots aim for you and your fatigue never is a factor in your accuracy.
MUCH LIKE BOTS PLAYING POKER.

Wallhack lets you see through walls - it reveals information but you are still required to aim the mouse yourself.
MUCH LIKE POKERTRACKER, IT REVEALS READS ON PEOPLE BUT YOU STILL HAVE THE MAKE THE PLAY.

I played that game a bit, and if I had to choose between the two evils I would choose to share a server with a wallhacker over an aimbotter any day of the week.

Regards,
C

Chess is a high profile, socially respectable game...
And Fortune 500 companies like IBM...
Spent tens of millions of dollars...
For the Public Relations benefits...
Of defeating highly respected Chess Grandmasters.

Also... because chess is a game of "complete information"... defeating a Grandmaster was purely a function of processing speed... so it happened the day after computers became fast enough.

In sharp contrast...
Poker is a relatively sleazy casino game...
That is illegal to play online in the USA...
Because there are legitimate questions about "moral hazard"...
And no Fortune 500 company wants to be associated with poker...
Nor will anyone spend tens of millions of dollars...
In order to defeat Phil Laak or whomever.

Also... poker is a game of "incomplete information"... and Elite poker play has little to do with computer speed... but, rather, is entirely reliant on various AI algorithms... none of which are even remotely close to competing with the human brain. As a Computer Science grad... I've been reading about the AI field for 30 years... and it's been mostly hype all along the way... like Flying Cars and Time Travel.

I would agree with all of the above except for heads-up poker where the best bots are clearly ahead of humans. In fact I would not be suprised that many of the top HU online poker players have bots making the decisions for them.

Of interest it was research at IBM that developed one of the first world class bots for backgammon. The original paper is here: http://www.research.ibm.com/massive/tdl.html

Most of your argument is salient - but this is not. It is not illegal for a US citizen to transfer funds into or out of a poker site.

It is, however, illegal for a US Financial institution to do so.

AB

Please get your FACTS STRAIGHT. Poker is NOT illegal to play online in the united states. It IS illegal to play online in some states, like Washington - but not in the whole of the US.


ORLY?


Harrah's Entertainment is ranked 254 in the Fortune 500 - and I guarantee you they are associated with poker (ahem, they host the WSOP).

MGM/Mirage, Starwood, and Hilton are all also in the top 500, and all also associated with poker and even degenerate gambling like blackjack.



AB

Isn't it theoretically quite easily possible to make botting extremely difficult with just a bit of creativity? I do believe that AI can play decent NL and bots could be a huge drain on the poker economy, but there are some areas where AI simply will not be anywhere close to even a handicapped human mind in hundreds of years, and they could be used to hide the input from the AI. Something along the lines of, the user is only able to decipher the actual nature of the cards in his hand (i. e wether it's an ace of clubs or six of spades) visually (instead of server sending the client the string "Ac" or "6s" or whatever), and the visual signal is varied constantly (can't think of a good practical solution to how éxactly do that off the cuff, but to illustrate the idea think the server having prepared thousands of card pictures and different strange fonts, which also get changed constantly, and instead of the easily detectable string "Ac" or "6s" or display this and this picture on the hard drive, the client gets transfered these pictures, which are incredibly difficult to decipher for the AI, but any normal 3 year old human can recognise in 0.1 sec. Of course this example would be a horrible bandwidth drain but you get the idea, workarounds can be achieved. For a hilarious illustration imagine instead of clubs and spades and diamonds there being monkey faces and chess pieces and adidas shoes, after all it doesn't matter at all what the icon is, as long as there's 13 of each it could change every hand for all I care.). I mean it would be a lot of fuss but definitely doable. ...Right?

Started to read this thread, very interesting.
2 Indiana:
If I understand game theory correctly since in poker everybody has equal opportunities for every strategy you develop one can find a counterstrategy which will consistently beat the former one. Ok, may be not to beat, but at lease to have zero EV. There is also something like the law of irrelevance which roughly says that if a strategy is optimal then no matter how you play against it if you don't do big mistakes you won't loose. Therefore to make consistent profit the strategy has to be not optimal, on the other hand the strategy should depend on some patterns of mistakes that the opponent do consistently. Therefore ones somebody knows how your bot works one can build a strategy, that works may be only against this bot, and totally crash him. Am I right?

Interesting Thread.

I did see what computers can do in Chess, now they beat every Humanplayer.
I did see what bots can do in Onlinegames.

But in Chess, a computer can calculate every possible move. In a Online-game there just a few actions he have to do.But thats not AI.

I think such a bot could do well in SNG´s but not in cashgame. Maybe he can play lowlimit Sngs with a ROI about 10%.
He would play very TAG and will play at higher plays with perfect ICM.

So he maybe could do 20k $ a year, per account. I think thats much money...

Such a Bot plays just as well as his programmer do...

Im sorry for my english.

From the game-theoretic point of view poker is just a game of mistakes. If everybody played perfectly there would be no poker - everybody would get 0 EV. Therefore poker is completely different from chess. You don't want to find in poker the optimum strategy. You want to play the strategy that uses other's mistakes. It is very often in such games that strategy A beats B, B beats C and C beats A. I think it would be fair to make bots open source for purely scientific reasons then people would understand better their mistakes.

The current best PC chess engine is a program called Rybka. It was written by one guy with no budget. This engine (running on suficiently fast hardware) would probably beat any Grandmaster in a match. This program actually inspects far fewer positions per second than it's rivals, yet plays far stronger - so it is not just about processing power.

A poker bot has nothing whatsoever to do with AI. Poker is a pretty simple game of probabilistic decision making. It has a rigid set of rules, just like chess. Computers are pretty good at probabilistic decision making, given the right models. It might be difficult to make a bot that would regularly beat top pros, or that couldn't be exploited with the knowledge of it being a bot. But on the other hand, I don't believe that a bot that consistently beats online NL100 (say) is beyond the realm of possibility.

Small correction. Rybka reports that it inspects fewer positions but in fact it inspects as many if not more as other engines. I suspect this was initially done to deceive other engine writers as to where the improved strength came from. The main strength of Rybka over its' rivals comes from its search algorithm (as opposed to position evaluation) which uses very aggresive pruning.

Poker bots do beat the top pros at HU poker and are practically unexploitable by humans. With any more than 1 opponent bot strength and exploitablility drops off considerably, the "search space" just grows too fast.

Here is the strategy against bots.
Play Nash equilibrium.
With time the bot will switch to some exploiting strategy because he would gather some information about you which would be there because of some random fluctuations but the information will not be true since you know you play equilibrium. As soon as the bot deviates from the equilibrium you start winning!

Hi guys,

Whilst looking for a simple chart on the net to help me explain the odds of me drawing certain outs in different situations, I found Poker Tracker.
Is this common place on pokerstars and am I the only mug not using this software these days?

I understand it can only give mathematical percentages but surely this helps.

p.s I am such a newbie to the politics I would appreciate if somebody told me what the following terms meant.

b)collusion
c)multi-accounting
d)database sharing
e)HUDs
f)10bb/100

Many Thanks
Marsh

b) 2 or more people working together at the same table
c) one person using more than one account
d) people gathering information about players at the table in the form of a database of hands, then sharing this database with another person
e) heads up display (a visual display of the info in the database, readily available on top of the tables)
f) 10 big bets per 100 hands (winrate, the player is winning an average of 20 big blinds per 100 hands)

I like that perspective.


As far as superturbos go, bots could totally destroy those games - they are simpler games, and seem to require perhaps more math than poker skill (but it's hard to know which is more important).

HUDs are kinda shady too. IMO poker is about using your brain, not using better applications. I have used PT before - it's cool and useful and stuff, but the datamining factor isn't really fair. I guess what I mean by that is it is perhaps ok to use for remembering VPIP/PFR/AF for the current game, but when you keep those numbers and eventually end up playing with the same person again, you may have a very unfair advantage.

Shadiness is all about finding loopholes - you can observe this in law, business, etc. - and I think that these tools can be quite similar. I can see how building and using bots could be fun and intellectually rewarding. But if you are that smart I don't see why you can't just learn to crush certain games, at the very least.


Use your gut and use your brain.