Slashdot Mirror
Online Poker Bots Becoming Problematic?
scumbucket writes "MSNBC has a story about how poker bots have started to appear on internet gambling sites and the implications. It also talks about how a 'master level' poker-playing bot already exists. Could this proliferation of poker-playing bots undermine the almost $1 billion online gambling industry?"
'master level' poker-playing bot
Stop calling me a bot.
I don't see what's wrong with it... if I could play against a poker-playing robot at a real casino, I would! People who write smart bots deserve the money, as far as I'm concerned.
Visit the Game Programming Wiki!
Here is the original post I sent on twoplustwo, as quoted at
;-) in the
the end of the news. Despite the desperate tone of the article,
I'm not depressed (;-) and I'd be interested to see bots fighting
on the poker server (Free Software) I work on at http://gna.org/projects/underware/.
---
Disclaimer: I do not favor bots, I do not develop bots, I won't
be happy if there are more bots than humans in online poker rooms.
From a technical point of view, no poker client will ever be able to
detect a bot that analyzes the window layout (to find cards, bet
amounts, player names etc). It could attempt this detection when the
bot runs on the same machine although it is likely to require frequent
updates (think anti-virus software). However, if the bot runs on
another machine and watches the display remotely, it is just
impossible (VNC is a example software that watches a display from a
remote machine).
From a legal point of view, international and national laws in most
countries (+200 of them, including US and all Europe) strongly
protects interoperability between programs. It means that the author
of a program whose sole purpose is to encode/decode the protocols or
file formats used by another program can never be sued on this basis.
Online poker rooms can forbid the use of any computerized assistance
(except the mouse, the screen and the operating system
terms and conditions that each player accepts when registering. A
contract is a powerful tool to attempt to force people to forfeit
rights that cannot be taken from them. Although the poker room may win
a lawsuit against a player using a bot that plays on his behalf, there
are more cases where they would lose.
For instance, if my only machine is running GNU/Linux, the court may
rule that I'm entitled to use my own client because there does not
exist a client except for Windows. Ruling otherwise would mean that
the poker room can force me to become a Microsoft customer. A real
world poker room can force you to wear a tie but cannot force you to
wear a tie of a given brand. This can have precedence over contract
terms and conditions. Furthermore, the features provided by my client
software (such as automated play or statistics gathering) cannot be
restricted by contract. No matter what is written, no third party can
legitimately control or restrict the software you run on your own
machine. If that was the case, no doubt a large software publisher
would state in its operating system license contract that all software
running on top of it must be purchased from them.
Summary:
. Bots can't be detected.
. Bots can't be outlawed.
. Poker room terms and conditions are inefficient to forbid bot
usage.
Will there ever be a widely spread bot able to beat most players
currently playing in online poker rooms ? I think so. It may already
exist but is kept secret. It's only a matter of time before a talented
poker player who also happens to be a good developer decides she or he
wants to be remembered as the author of the first bot that changed
online poker forever.
~
...with being an ass-hole."
-- Henry Rollins
I don't know about the rest of you, but I have little sympathy for the house when someone figures out a way to beat them.
I wouldn't be surprised at this at all - I've even heard rumors of people playing online chess while using Chessmaster to tell them their moves.
It was, um, a friend of mine...
"If you think you have things under control, you're not going fast enough." --Mario Andretti
Well black jack is different than poker. In poker you are not taking the casino's money you are taking the other players money. It is a problem for the online casinos only if nobody plays because it is to hard.
Poker is played against other players, not the house. The house makes money by taking a cut of every pot. Poker bots could undermine the industry by scaring off human players that can't play well against a bot. This will reduce the amount of pots that are being played, thus reducing the house cut.
I'm relatively sure that all of the online gambling sites use either Flash or Java applets to display cards and such. I wouldn't think they'd make it so easy as to give easy access to card names.
I've used the program. It's available at pretty much every torrent site. It's not so much a bot as something that tells you the odds of what you have a chance at getting versus what it thinks the other players might have. It then recommends what your next step should be.
It's not like this thing wins games for you. It basically does what the good poker player can do, look at their cards, look at the cards on the table, and then compute odds.
It's less useful than the article makes it sound.
I've just signed legislation that'll outlaw Russia forever. We'll begin bombing in five minutes.
Poker bots are problematic for other players, not he casino. You don't play against the house when you play poker, you play against other players. The house then take a 'rake' of the pot, that is a % of the pot.
So, yes bots are problematic in poker because if I play, I want to play against another human, not some computer simulation that can calculate the odds down to the decimal. Granted some humans exist like that, but not many.
-Mark
Dovie'andi se tovya sagain.
until someone accidentally uses an aimbot, and blows a guy's head off.
At first I was thinking that maybe this isn't fair to the other players... But then again if you think about it... As it stands poker is still a game of chance... A bot can only play optimum hands based upon the cards it sees and what it knows is still in the deck... This really isn't any different then a human player. If bots exist that are beating inexperienced players, how is this different from the poker pro who logs 10 hours of online poker a day?
When you break it down it still takes a skillful poker player to engineer a bot that can perform at a winning level...
Also the bots are betting someones money...
There is an inherent risk in online poker that the player at the other end of the connection has tools that he is using to gain a competative advantage, such as tools for counting cards, figuring odds and so on...
If you're looking for real human vs human action without worrying about cheat tools find a game in your neighborhood and go play there. Even though gambling isn't legal in all 50 states you can always find somewhere to play if you look hard enough.
I'm the head webmaster and programmer for a popular casino based in the Cayman Islands (although I reside in the US). We use VBScript for all our programming and backend work. We're smart and renamed the cards, so the "ace of spades" is really "5_hearts.jpg"
No one will figure that out.
it's easy to find software that is designed to follow along with the current game, giving you the statistically best move based on what is known, such as the face-up cards, who is on the button, and what your cards are. They also give you a gauge making it easier to decide whether to call or fold a hand. After seeing this type of program in action, it's turned me completely off of putting any real money into online poker.
First of, you DO get something when you gamble. You get the thrill of gambling. Anyone who walks into a Casino knows that chances are they are going to walk out with less money then they started. It is the fact that you might beat the odds and the thrill that brings that adds value. Saying that they are not giving you anything in return is like saying TV doesn't give you anything in return... sure it does. It gives you entertainment.
As to bots, they are not going to cause a Casino to loose money in any other way in that they might simply stop allowing certain games to be played online. If the game is a game where a strong pattern rec software can 'beat the odds' then they will simply get rid of the game, have their own bots play, or adjust winnings such that they still win in terms of dollar amount in the end. The only people who are going to come out loosers are people who who want to play online without a bot.
Well, many of these games are java based and require screen reading and physical clicking (or programmatic mouse clicking) to do what needs to be done.
I don't see why they make it sound so hard to code something like this. There are books out there that teach you strategy for poker and what to do based on when other things happen. If you could turn that into a programmatic routine, it shouldn't be hard to have a bot that wins more often than not.
Especially with online blackjack. Bots could make a killing on that. Between card counting and the what-do-I-do-when rules.
...spike
Ewwwwww, coconut...
...you don't trust other players to not be using bots, but you trust the house to not add their own player to every game and fix the host software to guarantee that the house's player wins???
"Freedom means freedom for everybody" -- Dick Cheney
Recently I was browsing www.suprnova.org for obscure torrents in their misc area, and I found this "ebook" on how to win texas hold 'em. I downloaded it hoping that it was some kind of strategy guide or an odds list or something
turns out it was just some product promotion for a company that makes gambling bots.
I downloaded the software trial out of curiosity (I've never played online poker before, I just wanted to see how the program was set up.)
The way it worked (or claimed to, I never tried it) it would monitor my poker game and make calculations based on other people's bids checks or folds and give me tips about whether I should fold, check, bid, or bid high. It kept a percentage rating for probability of wining and stuff like that.
Basically it claimed to play the game for me, which would suck as I was looking for a strategy guide instead. I can't remember which one it was that I downloaded.
here's a link to one of them
I have come to the conclusion that anything that has to do with money on the internet will eventually be hacked and exploited, why should gambling sites be any different?
"It's so convenient to have a system where everyone is a criminal" - A. Hitler
The University of Alberta does a lot of work on poker programs. One of their aims is to create a program that can compete in the World Series of Poker. However, poker is such a complex game (for computers, that is) that at the moment the research group would be satisfied if their program would survive the first round. You see, poker is more about beating another person, than about calculations. The best poker players win because they can entice their opponents to make mistakes. This is psychology, not math. Computers are notoriously bad at psychology.
No way! Coming from personal experience, I've played on-line and at the casinos. Bluffing is a big part of strategy in poker, and seeing the facial experssions is key. On-line poker could compete with webcams, but how many people would go through that extent? No one I know. The older folks probably want to get away from their spouse and kids. Also, the atmosphere at the casino is part of the adrenaline rush... hearing the constant noise, seeing tons of people, and getting free drinks. Playing in your pijamas without having to wash up might be alright occasionally, but I believe the vast majority of players will still drive to the casino to fully immerse themselves in the poker pit over.
Hmmm that's why fred#3079-beta1 would never answer any of my questions.
and explain why I am broke.
If you suck at playing poker and write a bot to do it for you, your bot will likely suck as badly as you do...
"The object of war is not to die for your country, but to make the other bastard die for his." - Patton
who the 'bots are. Strike up a conversation with your fellow on-line players. Something like...
Holden: You're in a desert, walking along when - Leon: What one?
Holden: What?
Leon: What desert?
Holden: Doesn't matter what desert it is, it's completely hypothetical.
Leon: Well, how come I'd be there?
Holden: Maybe you're fed up. Maybe you just wanted to get away from it all. Anyway. You're in a desert, walking along when you look down and you see a tortoise, Leon. It's crawling toward you. Leon: What's a tortoise?
Holden: You know what a turtle is?
Leon: 'Course!
Holden: Same thing. So you reach down and flip the tortoise over on its back, Leon.
Leon: Do you make up these questions, Mr. Holden? Or do they write 'em down for you?
Holden: The tortoise lays on its back, its belly baking in the hot sun, beating its legs trying to turn itself over but it can't. Not without your help. But you're not helping.
Leon: WHAT DO YOU MEAN, I'M NOT HELPING?
Holden: I mean you're not helping, Leon.
Despite what ESPN would have you believe, there are a large number of people who play games other than "No-limit hold 'em'", namely Limit and Omaha. Each of those games are based FAR more on mathematical odds, probability, and having the nuts as opposed to the bluffing and gamesmanship required for No-Limit. It would be much much easier to program a bot that could play Limit and Omaha profitably. No-Limit and preventing collusion are a much more difficult task.
sig--we don't need no goddamn sig
Oh wait, this was about poker? Sorry.
My beliefs do not require that you agree with them.
Picking up on a bluff is overrated. Most of the time, players are not completely bluffing (i.e. they have a lowly but reasonable hand). What constitutes a bluff? In Hold 'Em if I go all in on an 8-man table while only holding two pair, is that a bluff? Maybe and maybe not. The important thing is that you still have to beat me. If you're worried at all that I'm bluffing then you probably have a weaker hand yourself. The guy with the full house isn't worried if I'm bluffing or not. He's going to bet into me because he thinks he's going to win.
Bots won't help you decide if someone is bluffing, but they will help you decide if it even matters whether or not they're bluffing.
I had a debit/ATM card compromised somehow last year. There wasn't very much in the account at the time, so the guy set up an account "for" me at a poker site and tried to gamble my balance up. He lost a few hundred. I noticed the withdrawls a few hours later and called the bank, after finding that my wife and I didn't have enough money to go out to a nice dinner that night. (The charges hadn't posted and were labeled as "ATM/POS activity", so I didn't know how they were spent. I just knew it wasn't me.)
I called the bank and while I was on the phone with the bank rep, more weird charges were coming in! We were both watching someone gamble away all my money in real time. So he red flagged them all and gave me a claim code.
The next day the phone rings. "Hello, this is Planet Poker..." and without thinking I say "No thank you" and hang up. The phone rings again a few minutes later. "Planet Poker..." and I say "please take me off your list" and hang up, still thinking it's a telemarketing call. Which sounds stupid given the withdrawls the day before, but I didn't put two and two together. (It was Planet Poker calling me to welcome me as a new degenerate gambler / customer.)
The phone rings again. "Don't hang up we think someone used your credit card!" she says really fast. I said, oh yeah, I reported those charges to the bank yesterday.
Then she sounds sullen. "Well... I guess we'll be getting the chargebacks then..."
I said, "yeah, I guess so!"
Don't know if the guy was using a program to help him cheat, but he played really badly.
Casinos are out to make a buck, with little regard for the welfare of future of the players. They don't care if an addict comes in with all of his life savings, and blows it all in one sitting.
Actually, yes, they do...greatly. Casinos are based on gaming as entertainment and repeat business. They want to people to get hooked and then go to the casino once a month or once a year and keep spending money in their casino. Problem gamblers are regularly blacklisted since they drive away paying customers.
Sure, but problematic for whom? The casinos are notorious for putting a winning blackjack player on a blacklist, and not letting them in. Why? Did they cheat? No, they just won, and casinos hate to lose money
Casinos generate the majority of their profit from their interest on their cash reserves, not from actual gaming. The take on gambling alone would barely covering operating the average casino. They generate their revenue on the interest of their cash, an adequate reserve of which must be maintained to pay out all outstanding bets at anytime. The more bet, the more reserve, the more interest. The take from gaming is actually secondary now.
Blacklisting is reserved for cheaters and addicts, not winners. Casinos want winners to keep playing and turn their money over in the casino, game the system and bad things happen.
While 'bots are a serious topic, there is a solution. Make the "user" periodically enter in some text that has been graphically morphed... something that only a human eye can recognize. It can be done in seconds, and that would eliminate the use of 'bots.
What I can't figure out is how the gambling industry is going to fight "group cheating". Put 4 or 5 laptops together, and have several people cheat the rest of the table out of their money by sharing their hands. It's not hard to do, and it's impossible to detect. Especially with wireless access.
I have a couple of friends that refuse to play online because it's impossible to stop this behavior in online poker.
-- No sig for you!
Since bluffing is meant to fool human heuristics that judge the strenght of other players' hands based on those players' bets, why would a bot that works on probabilities NEED to consider bluffs?
From the article: argue the complexities of the game and the changing strategies ensure that creation of a program that can "read" opponents' cards using screen scanning techniques and respond in real time is years away at best.
well considering that my friend and i implemented a system that does just that (over a year ago), i call bullshit on this one. what we did was capture an example of each card off the screen (we use ParadisePoker). then we came up with a best-match algorithm that compares each card to our reference cards to decide which card it is.
it's the "react in real-time" part that is the bitch of it all... the "read" from the screen is, while i won't say trivial, doable. it took me and my friend less than a week to implement using java's "robot" class (to read pixel values from the screen).
some guys we know were just trying to put together something that calulated odds for 7-stud on the fly, but found that inputting the cards by hand took too long. so they asked us if we could capture the cards from the screen. we didnt know if we could or couldnt, but we looked into it and it wasnt so hard.
i also know of more than one person back at school (MIT) who is working on a fully-functioning bot. i dont think they are "years" off from being finished either.
. <-- the joke
o <-- your head
You: The University of Alberta does a lot of work on poker programs.
Eliza: Please go on.
You: One of their aims is to create a program that can compete in the World Series of Poker. However, poker is such a complex game (for computers, that is) that at the moment the research group would be satisfied if their program would survive the first round.
Eliza: Why do you mention computers?
You: You see, poker is more about beating another person, than about calculations. The best poker players win because they can entice their opponents to make mistakes.
Eliza:Oh, i see poker is more about beating another person than about calculations the best poker players win because they can entice their opponents to make mistakes.
You: This is psychology, not math.
Eliza: Please go on.
You: Computers are notoriously bad at psychology.
Eliza: Why do you mention computers?
I've always preferred making sure I have friendly players around the virtual table and sharing card information. Much, much easier to perform informed betting at that point. :) Plus, with that many cards on show you usually have a pretty good idea what the non-friendly players could have.
Not nice, but thankfully it's about as immoral as I get.
People that believe in their opinions don't post AC.
Seriously.
Foor the cool factor (Yay! its GNU!) consider "GNU Backgammon", the program uses 3 neural nets and humongous move databases. Backgammon compares well to poker btw, BG is ruled by dice and skill, Poker is ruled by chances and skill too. It is quite likely the strongest BG playing, ehm, thing, in the world.
Gnu BG plays an astounding 2200 rating on Fibs, if not higher if you get high end hardware, and give the bot a few secs between moves.
1800 is considered a worldclass human player, 1900 and above are grandmasters.
Friends, dont play backgammon online for money, and certainly not Poker. Instead if you must, visit tournies in the flesh.
Or get the bots, and a few spare comps... You will NEVER rob the casino thou, you will rob other suck^D^D^D^Dplayers.
"/Dread"
It doesn't "need" to do anything - however, if you come up with some ad-hoc or game-theoretic bluffing index, it can give advantage in spades. Kind of like a meta-heuristic.
Knowledge of bluffing is important, because coupled with a grasp of human psychology, it (potentially) gives more information about future game state than simple card probabilities will.
They actually use standalone .exe apps you have to install, and the thing in the article saying that a program is years away from identifying cards onscreen or whatever is ludicrous, because sites don't change how they display the cards.. If you wrote an app to watch a 7-card stud game on a certain site, you could easily just tell your program "this 32 X 48 pattern of pixels is the ace of spades, this one is the 7 of diamonds" and it could easily tell what every other player had by just looking for those patterns. Making a program that can tell this on EVERY site would be hard, but why bother, just make it for one or two of the most popular sites and you're fine.
sig:
See the "..for smart people" banners Wired runs here? Look elsewhere guys.
Poker is very different from Blackjack.
In poker you play against other player ; in blackjack you play against the house.
Blackjack used to be a game of chance with odds that could be slightly turned against the house in favour of the player if he played "perfectly". For this he needed to calculate odds given the number of cards left in the dealer's shoe and bet accordingly. If the game was played this way online, it would be a disaster for casinos as bots would rule the game. Online however, the deck is shuffled after each hand and there is no way of calculating the odds. Therefore, online blackjack has become a pure game of luck which is why bots are useless and why you shouldn't play blackjack.
On the other hand, bots can help you calculate odds in online poker. But that's only part of the game and they are largely ineffective against any decent players as they cannot understand human psychology as well as they can calculate odds.
Unlike chess - where bots are very effective - poker is not an information complete game. Therefore a player's skill depends strongly on his ability to "read" and bluff other players. Which is why poker bots will probably remain useless for a very long time. Probably until we reach hard AI.
The place where a bot could help in knowing if someone might be bluffing is that it could be used to instantly go over every hand you have played at the table. That data could be used to show your general tendencies while playing at the table, then give some sort of estimation of whether or not you're bluffing.
Knowing how many hands you went the distance on, your winning percentage, and your betting habits during each betting round are the things that a really good player knows about every other player at the table. Those are the things you see people thinking about while deciding to make a call on a bluff/non-bluff. Those are also the exact things a program could help track.
First of all, the house doesn't care in poker, because they get a share of the pot and aren't in an adversarial role. The only reason they might care is if this type of thing scares off real players and reduces the overall take in the course of an evening's play.
Secondly, depending on the casino, you can ask the dealer what the odds are, what the house's play would be, or you can even ask the dealer to play your entire hand as the house would. This varies from game to game and casino to casino.
Lastly, the house allows you to have certain materials available when playing. In Vegas, there are "player cards" or cheat sheets for black jack and other games that are about the size of an actual card, but show the plays that you should make based on certain stats (what you have, what the dealer/other players have, etc). Last I went to Vegas the only rule was that you had to set it down on the table before play began. As long as you weren't sitting next to some idiot that messed up the card distribution then it usually panned out. I believe there are some casinos that don't allow them, but most do, although you can't use them at the high minimum bet (>$50 or so) tables.
Basically I don't see the problem as long as this "bot" is really just a tool and the player is still interacting with the game, i.e., not automated play. I see it as a way to even the odds a bit and help the noobs not make so many stupid mistakes. Can't you remember a time when you just glanced at your cards and thought you had a certain hand and as you tossed them down in triumph you realized you misread the hand?... I can. Preventing those kinds of mistakes would save everyone some grief. Although I suppose poker wouldn't be the game it is if everyone didn't have a great story about a few hands.
Black jack is a game that can be beaten in the long run. That's precisely why casinos will blacklist good players... because the good players know how to consistently win. You can call it "card counting" but there is nothing illegal about it, and an average player can "card count" without much learning. Just keep track of how many high cards are floating around, and statistically when there are many high cards to be dealt, you are more likely to see the dealer bust... plain and simple. You don't play to win with your hand by hitting, you increase your bet when the dealer is more likely to bust and let them take the high cards.
Saying saying that blackjack "played at [sic] it's best... will translate into certain losses" is incorrect.
What's the point of saying you have a relatively "informed" view, when your view is wrong?
Most people play like bots? No, that's the problem, "most people" can't calculate perfect statistics on the fly, or do any of the other things bots can do in microseconds.
The 'human element' you talk about is present in face-to-face games, but in online games, you can't even see the other players! That's why bots can compete and disrupt the game. Maybe you can comment about playing poker in home/casino games, but I think your relatively informed view does not apply to this thread.
I suggest modding parent down.
As it stands poker is still a game of chance
Skill plays a major factor in the long term. If it wouldn't, poker professionals would not be able to exist. But they do exist and they do earn a decent living. The reason why skill is so important is that odds are only part of the equation. Straightforward betting on good hands and folding when the odds are not in your favor is easily exploitable by just not calling the bets and bluffing to get the opponent to fold. So a good player must use deception. That element of deception turns the game into an unstructured game that is very hard to beat algorithmically, so I have my doubt about being able to create world-class bots.
If bots exist that are beating inexperienced players, how is this different from the poker pro who logs 10 hours of online poker a day?
The difference is that the bot doesn't have to sleep, eat, pay taxes, etc so there are much lower expenses for a bot and it can work 24 hours a day. That means that if good bots exist, they can be let loose at tables where most people play for fun and where it's currently not worth it for a professional player to play. Then the poker games will split up in very low limit games that nobody plays seriously and the high stakes games where only the best professionals can live. There will be no middle ground, but that is where most money is made for the casino's and where most semi-serious players play. The result might be that online poker loses its appeal to 90% of the players.
If you're looking for real human vs human action without worrying about cheat tools find a game in your neighborhood and go play there.
That's not really realistic, is it? First of all, online poker is different from 'live' poker because you don't need a poker face and a lot of players like that. Also, you can play it whenever you want, without having to coordinate schedules with other people. You also don't have to play with the same 9 neighbourhood guys all the time. Then there are more games to pick from online. You can play big tournaments online. You can play freerolls online, where you can win money for free. I could go on, but I think you get the point. Online poker is just a different ball game.
The article mentions in passing one other danger.. "Team Edition". It doesnn't mention details but it clues you to the the obvious way of successfully cheating.
Imagine the advantage of having two machines side by side EACH playing a hand in the SAME game. Not only would you know more cards in play, but more importantly you could always have the ability to use the stronger hand as your main betting hand, folding the weaker hand to avoid wasting money on it. The mathematical advantage of that must be Very Large.
Seems like this cheat would be undetectable, easy to do (two internet connections so they can't compare your IP #s), and doesn't require any bot coding at all.. very adaptable to any casino or player changes or questions.
Summary: you can't trust any online betting activity.
In Poker, the house has no interest in the outcome (presuming the game is not crooked - that the dealer and a shill are not in cohoots). The house gets their money by raking the pot (taking a percentage) or charging a time payment from one of the players (each hand the person paying rotates). That's why Poker is about the only form of gambling where you as a player have a fair shot (notwithstanding the story of the MIT Blackjack team).
I can't imagine trusting online Poker play. Even if the site/house is honest, players can share information secretly or use aids to calculate pot odds perfectly. They can do that in offline games as well, but it is much more difficult to get away with it.
Beating a 0.5/1 table for 5-6BB/100 hands is considered destroying the game. You're claiming that you're beating it for about 40-50BB/100h.
Even if your bot is four-tabling, your alleged winrate is double what can be reasonably expected by an excellent player, simply due to the nature of the game.
It's a cute story, but next time try grounding your MIT tales of evil genius in a little reality.
Consider the following case in Hold 'em:
You are up against only one other player and are in last position. The river has come out and there is a queen (from the flop), a jack, and then 3 low garbage cards. You have Ace-Jack. If that player checks to you, and has checked the previous bets, you will probably bet that even though you only have second pair. If he bets high, and has been betting the whole time, you may want to REALLY think before calling, and especially before raising.
This is an example of how BETS come into consideration.
And of course a good bot, like a good player, would not be able to just take all bets at face value.
Since bluffing is meant to fool human heuristics that judge the strenght of other players' hands based on those players' bets, why would a bot that works on probabilities NEED to consider bluffs?
Bluffs are also intended to foil human intuitive statistical analysis. The fundamental problem is that the strength of a perfectly rational player's hand can be determined from his bets. So the player must introduce noise or bias into his betting strategy to maintain the advantage of hidden cards. Since computers are even better at statistical analysis than humans, bluffing becomes more important, not less. The problem is that a bluffing strategy is itself subject to statistical analysis. Probably ultimately, there is no constant bluffing strategy that consistently beats sufficiently randomization.
I remember the first time I went to Las Vegas. I saw signs bragging about how "loose" each casinos slots were. Some bragged that they paid as high as 97%.
I just loved it, they ADMITTED that for every dollar you gave, on average you'd get back only 97 cents. To put it another way, you get better odds from a change machine.
It's my opinion that the stupid and ignorant should not have money. Casinos do a great job at ensuring that.
If someone says he and his monkey have nothing to hide, they almost certainly do.
Yeah, the same way Internet porn has undermined the stripper industry.
--
make install -not war
There are a huge number of misconceptions about many aspects of this issue clearly apparent in this discussion. I'm going to go through some of the highest-moderated misunderstandings in hopes of reclaiming some of what this whole discussion is about.
I'm relatively sure that all of the online gambling sites use either Flash or Java applets to display cards and such. I wouldn't think they'd make it so easy as to give easy access to card names.
You do not actually need to break into the program in order to use some form of bot. Graphics recognition has advanced to the point where a hand can be analyzed on the fly by a concurrently running program. See Poker Office. Such programs can then immediately provide feedback based on the information they glean.
Could this proliferation of poker-playing bots undermine the almost $1 billion online gambling industry?
The end of the industry is not likely at hand. Poker is just one part of this industry, and the industry will continue EVEN IF bots are the only ones playing. The casino will just take the same percentage of each pot.
if a pokerbot can clean out the 9 other people at the table
Quite frankly it is ridiculous to think that a bot with perfect play can clean out any table. Good poker play results in a slow accumulation of profit at a faster rate than losses. A perfect bot will certainly not be playing more than 1 in 5 hands to begin with, and not win more than one in 3 of those. Good players can't just make the right cards appear, no matter what you saw in Maverick. They get the same crappy cards as everyone else, it's how they play them that differs.
you don't trust other players to not be using bots, but you trust the house to not add their own player to every game and fix the host software to guarantee that the house's player wins?
Yes, very much so. Contrary to popular opinion, most people are not complete retards. It's not difficult to tell when someone is consistently winning - certainly there are hot streaks, but any whiff of foul play will result in a huge exodus of players from any poker site. They have no reason to do such a thing, as profits are huge from both the rake AND the interest they are collecting on your bankrolled money.
_______
Any current bot is very likely for Limit poker - this is the 'easiest' style to play purely by the numbers. The state space required for a bot to make decisions in No Limit poker is absolutely huge- one poorly written part can get your bot cleaned out regularly.
Personally I would LOVE to be at a table where I have positively ID'ed a player as a bot because I could then run circles around it. There are a number of tactics that would play merry hell with a bot that plays the straight numbers, and even a bot that adjusts to my own play style is not difficult to take advantage of.
I play regularly online and I do not fear the bot. What I fear most is the bad player that will put all their money on a 20% draw, where any good player (or bot) would fold- because sometimes they hit, and that hurts.
Once they find a cure for bad players though, that's the end of poker, but I am content that that time is far in the future.
For Paradise Poker, the image reading would be easy. If you look in the Paradise Poker install directory, you'll find all the bitmaps the program uses to draw the graphics. Simply compare what's on the screen against the saved bitmaps until you hit a match. Identifying the edges of the cards would be the hardest task, imo, and that would also be pretty simple as the cards always appear at the same locations.
Some random links on the subject:
and a random company link (haven't read this one):
Why settle for the poker table, when the markets are much bigger? Playing the markets is probably more difficult, but you're the best coder around, aren't you? ;-)
Trusted Computing FAQ | Free Dawit Isaak!
(Side note: If anyone is interested in playing some online poker and wants a bonus on their first deposit, drop a reply to this with your name and email address, and I'll send a referral out. We both get a bonus from this.)
On the other hand, a network of several bots playing *at the same table* could do some fairly serious damage over time, and with relatively uncomplicated heuristics.
;-)
Something to think about.
"They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety."
Nah, you should haved used names like twisted.jpg and grind.jpg.
One line blog. I hear that they're called Twitters now.
Anyone who has spent a decent amount of time playing poker in the casino knows to be on the lookout for colluding partners at your table. (i.e. Two or more friends sitting at the same poker table that are able to exchange information about their hands between each other or even to use various combined betting styles to bluff or raise the pot to artificial levels). When done properly, having a colluding partner at a poker table is a significant advantage for you. On the Internet it would be very easy to have, say 3 or 4 seats of a 10-seat table controlled by you. Or hell, have 9 of 10. Just load additional casino accounts into web browers that go through different gateways. Or any number of dozens of other ways to make it appear to be coming from a different machine.
I'm a big tall mofo.
This whole idea is sort of irritating. I play a lot of online poker, and yes it is possible to write a poker bot to do some basic strategy. In limit hold 'em this may be particularly true, but in no limit? You have to deal with a real valued scale of bets every time... this is by no means an easy problem to deal with. Writing a player that irritates people heads up is relatively easy - lots of aggression. However, writing a full featured poker bot to play no limit hold em and fleece tables? I don't think so. However, putting aside all the computational problems with this - they act like this will KILL ONLINE POKER. It won't. Consider this scenario. Your poker playing program only needs to 1) show your hole cards 2) show the board. Encrypt the transfer to and from the application. If you change the keys enough, this will will force the user to type in his hole cards and the board cards as they come out. Also encrypt the traffic such that the only commands accepted by the server are those that come from their downloaded applications only. This may require more sophisticated authentication in the program but it makes sure that if anyone wants to use a bot they have to 1) type in hole cards and board cards manually 2) click the buttons in the actual application, or emulate a mouse enough to do the clicking for you. This forces a human to be involved, and would make it pretty obvious which players are bots. You can also monitor suspicious play and send a human observer around to interact with players in chat and see what they have to say, this is an annoying solution but still it would solve the problem. Anyway, these are simple off the top of my head solutions to the bot problem. I am sure there are more. This is just a hyped up story, and nothing to be worried about.
Here is an excerpt of my e-mail exchange with Mike Brunker (the author of the MSNBC article) prior to our phone interview. It might provide some interesting information on the topic.
- Darse.
[begin excerpt]
(1) It is not easy to write a good poker-playing program.
It took us (the Computer Poker Research Group at the U of A, http://games.cs.ualberta.ca/poker/) a few years to develop a program that could win consistently in higher-level games against opponents who took the game seriously. It has been successful against human players of average skill for many years now, but it is the only known program that can make that claim.
We operate a free poker server where people can play against our bots. Hobbyist programmers can also have their programs connect to the server and play in those games, and more than a hundred programs have participated over the past few years. None of them has come close to being a winning player, so it is clearly not a trivial task.
(2) Is it *possible* to write a very strong poker program?
Absolutely.
Poker is a challenging and rewarding field for research in Artificial Intelligence (AI). There are many aspects of the game that make it more difficult and more interesting than games like chess and checkers.
It isn't simply a matter of computing probabilities and other numbers. A good program has to *think* about the game in the right way. Master-level poker requires an understanding of how each opponent plays the game -- you must observe and adapt as you play, and that turns out to be a rather difficult learning problem.
Nevertheless, these problems will be solved eventually, and the technology will become available for others to use. It took more than 20 years for chess programs to finally become a serious threat to the best players. It won't take that long before we see elite poker-playing programs, but it still might be a number of years before they participate in online games.
Of course, a practical program doesn't have to be as good as the best players -- it only needs to beat a game with average players in order to win money.
(3) Are bots playing in online games now?
I expect there are a few now, yes. Perhaps more than a few. But are they a threat? Probably not. Many of them will be losing players, at least for a while. Their authors will either lose interest, or have to invest a lot of time and effort to improve their programs.
If someone does succeed in writing a program that can grind out a small win, what difference should it make? It will be like any other solid player -- playing a conservative style (only good cards and good situations), and slowly extracting a tax from the weak players.
Look at it this way. Most people who play online poker lose money. That's an unavoidable mathematical fact. Considering the house cut (the rake), perhaps 30% of players can stay in the black, maybe less. Many of the losing players will lose slowly, so the cost is a fair trade-off for the entertainment value they receive. Some will lose much more rapidly, and they really shouldn't be playing at all (unless they happen to be independently wealthy).
Of the players on the winning side, most will only eek out a small win rate. A winning poker bot would just be another solid player at the table. Probably less than 10% of all players have enough knowledge and skill to win a significant amount of money, and I doubt there will be any poker programs in that category for quite some time.
Will the existence of good bots radically change online poker? I doubt it. Look at casinos (real and online) that offer the traditional gambling games like craps and roulette. Those games cannot be beaten -- there is no skill that can be applied to avoid losing in the long run. But that fact doesn't seem to harm the popularity of their business.
Instead of fearing bots, people should use them to help learn more about the game. Our research program is com
char*p="char*p=%c%s%c;main(){printf(p,34,p,34);}"