Student Creates On-Line Poker Playing Program

Psyber Samurai writes "Online-Athens has this story about Paul Apostolik, a cognitive science and computer science student at the University of Georgia that has created a program that plays the risk-free online video poker sites. The program takes over the screen and actually reads the cards off the screen then determines which cards to hold and which cards to discard. I've run it, and it is extremely fun to watch it play. Yes, it does win over the long term as it collects many, many tokens."

Comments

Re:Debunking online casinos.

[Jonathan]

It really proves that to get the best chances, GO TO THE CASINO! At least you can see if they are swapping card on you or something.

But this article is talking about *video poker*, which if you haven't been to Vegas, is played using electronic machines even in the casinos.

Re:OT: Your sig...

[SilentChris]

I have windows and it still didn't do anything. Was it supposed to?

Re:I did this with real money, and lost.

[Coolfish]

hrm, i'm no expert at resume writing but yours could use some work. your objective is weak and too you-centric, it does not emphasize what you could bring to the company, what you could do there, etc. in the hobbies, the "learning" is kinda lame. in your work experience, the "I did this..." and "I did that..." does not sound very professional. Likewise, you mention many times why you moved on, and it often is the company folding. I don't think an employer would find that attractive... "Great... we hire him, and we're next on his list.."

anyhow.. good luck with the job search!

What about brute force?

[JoeShmoe]

I've always wondered if you could use computers to generate a perfect list of moves which would, given enough time, always turn a profit.

Consider the game of blackjack...imagine a grid that has all the possible blackjack hands (from 2, 2 to Ace, Ace) versus all the possible dealer shown cards (from 2 to Ace). Each spot on this grid can hold one of two values: Hit or Stand. For example, when the dealer is showing a 6 and you are holding a 10 and a 6, that spot on the grid should tell you whether to hit or stand (definitely stand!)

Now, here's the tricky part. Make a routine with a giant loop. The loop should assign values to every spot on the grid (starting with all stands, for example) then play a few thousand hands using those rules. After a few thousand hands, change one spot on the grid to hit and repeat until the program has tested every possible combination (i can't even begin to imagine how many calculations this is but hell, there are several programs trying to crack huge key values and this is certainly a more financially valuable goal).

At the end, the program should print out a grid that contains the values that generates the highest number of wins. Then, if we human players were to memorize that grid and then use those rules when we play in the casino, shouldn't we have a good chance at making money? Dare I say, even finding the perfect combination that always turns a profit?

I dunno, but rather than trying to invent a blackjack genius that you would have to continuously update and modify, just brute force the game and be done with it.

Anyone feel like writing this routine in a p2p fashion? I have a few cycles I'd like to donate...

- JoeShmoe

Re:What about brute force?

[BlackjackInfoKen]

This is exactly the way that blackjack's Basic Strategy was created. For example, I wrote a complete combinatorial analyzer for the game to create the charts available at BlackjackInfo.com And, yes, the game is beatable using card counting. You don't have to be a genius. I've had a net casino profit for the last 7 years in a row.

Re:What about brute force?

[JoeShmoe]

I feel so dirty responding to such a clueless moron.

You'd be more convincing if you had the guts to say that under a login.

But your advice is crap. Thousands of simulations and patters of probability are all still just theories. Same thing for the little charts. They are all pretty good, but not perfect.

Look at chess programs. They play ahead X moves and use that to make a decision, but are still limited by modern technology to just a few moves. It is a poor substitute for knowing every possible chess move and thus never losing.

Chess maybe be too complex to brute force, but a simple game like blackjack isn't.

- JoeShmoe

Re:Possibly more common than you think

[Eslyjah]

your friend, huh? wink, wink...

Stock Market Predicting Computer

[Robotech_Master]

Actually, according to PBS's Robert Cringely, such a thing does exist, and presumably people do indeed know about it:

But can you really program a computer to find patterns in stock prices? Yes, you can, according to Jim Hall, who did just that. Hall was an engineer at Deere & Co. in Moline, Illinois, who was asked to build a combine that didn't need a driver. Deere was worried about the rising cost of farm labor, and so asked Hall for a machine that didn't require skilled labor -- a machine that would automatically learn the field and harvest without supervision. Hall built the combine, which was never marketed, then converted the software to learn another field -- the stock market. Hall became the manager of Deere's pension portfolio, successfully harvesting capital gains instead of wheat.

As for what effects it's had, well, probably not many, or else people would have made more noise about it. :)

--

ARGH! - this is not A.I.

[jamiemccarthy]

I'd like to blame it on Spielberg's latest, but the fact is, movie or no movie, reporters will take any damn program that someone writes and turn it into the latest example of "artificial intelligence." Yes, this is my pet peeve. Do reporters think nobody will be interested in reading their story unless it mentions computers and "intelligence" in the same sentence?

In this case, the program reads the screen and moves the mouse to click on the proper places. Pretty clever I suppose. But once you get past the fancy I/O code, it's just applying simple formulas to decide what cards to hold. There's nothing "intelligent" about calculating that holding the pair will give an expected return of $0.21, while going for the flush will give an expected return of $0.08.

The only thing about poker vaguely related to "intelligence" is its condensation of the social relationships of zero-sum games into the formal interface of betting. Placing bets is much more amenable to mathematical analysis than the fuzzier stuff we do when we're playing similar zero-sum games, say, negotiating the price for a used car: furrowing our brows, clearing our throats, poking fingers at each other's chests, yelling, pretending to think over an offer, saying "my manager will have to approve this," and so on.

When you simplify the interface between entities to just a few operations (raise, call, fold), suddenly mathematics can play a role.

John von Neumann was one of the first to recognize this, and it was the game of poker that he focused on, in his study of game theory:

The nominal inspiration for game theory was poker, a game von Neumann played occasionally and not especially well. (A 1955 Newsweek article appriased him as "only a fair-to-middling winner" at the game.) In poker, you have to consider what the other players are thinking. This distinguishes game theory from the theory of probability, which also applies to many games. Consider a poker player who naively tries to use probability theory alone to guide his play. The player computes the probability that his hand is better than the other players' hands, and wagers in direct proportion to the strength of the hand. After many hands, the other players will realize that (say) his willingness to sick twelve dollars in the pot means he has at least three of a kind. As poker players know, that kind of predictability is bad (a "poker face" betrays nothing).

Good poker players do not simply play the odds. They take into account the conclusions other players will draw from their actions, and sometimes try to deceive the other players. It was von Neumann's genius to see that this devious way of playing was both rational and amenable to rigorous analysis.

Though much more interesting to analyze than, say, blackjack, playing a good game of poker still does not qualify as "artificial intelligence." It's been pretty well analyzed since, if I'm not mistaken, the 1920s:

[A particular type of deception] resembles bluffing in poker. Poker can be quite complex, in part because it usually has more than two players. Von Neumann analyzed a simplified form of poker. In outline, his conclusions apply to the real game. He showed that you should always bid aggressively when you have a strong hand. With a weak hand, you should sometimes bluff (bid aggressively anyway).

Von Neumann distinguished two reasons for bluffing. A player who never bluffs misses many chances to call the other player's bluffs. Suppose that both you and your opponent have bad hands. You don't bluff; your opponent does. That means you fold and your opponent wins without a showdown. Had you also bluffed, your lousy hand would have been compared with his lousy hand, and you might have won. The bluffer can exploit the nonbluffer; ergo, von Neumann's rational player must bluff. [...]

At the most abstract level, game theory is about tables with numbers in them -- numbers that entities are efficiently acting to maximize or minimize. It makes no difference whether you picture the entities as poker players who want to win as much money as possible...

(Quotes from Prisoner's Dilemma, William Poundstone, 1992, pp. 40, 60-61.)

A program that applies von Neumann's minimax theorem to place poker bets is still just doing simple math. But Apostolik's program -- cool though it is -- doesn't even do that. He hasn't incorporated betting yet. "That would be an interesting thing to play around with," he says. True! But even that still wouldn't be A.I.

Jamie McCarthy

Re:ARGH! - this is not A.I.

[god_of_the_machine]

As this poster has already mentioned, our own intelligence works in a similar way. There is nothing that makes human intelligence "special" except the level of complexity of the input and processing that we do in a unique way compared with the current crop of computer technology.

Personally, I would classify intelligence as a scale. At one end of the scale would be simple formulas such as the ones used in this poker-playing machine, and in a simple organic creatures such as a fly, which has pre-programmed responses to stimuli (if food, eat(food), else flyaway()). At the other end of the scale would be the human brain with a learning neural structure that accomplishes much more than its evolutionary programming could have hoped for. But in the end, it's all the same intelligence. Since it's lunchtime, I'm going to find food -- just like the fly would. =)

-rt-

Re:ARGH! - this is not A.I.

[Jace of Fuse!]

But once you get past the fancy I/O code, it's just applying simple formulas to decide what cards to hold. There's nothing "intelligent" about calculating

Humans are just a collection of sensory organs connected to a huge response-to-stimuli databank, and for some reason the media seems to think Humans are Intelligent.

Go figure.

"Everything you know is wrong. (And stupid.)"

Re:ARGH! - this is not A.I.

[Jace of Fuse!]

Science *explains* nothing, it only describes the world according to how well data correlates to a model.

Science explains plenty. Not all models are correct, and some are in fact very wrong. However, many are useful none the less. The useful ones are held into value by their own merit, and those that are largely inaccurate eventually come to be replaced.

People like me? You know nothing of people like me.

I believe science works towards solving problems. And that's the key, it works towards solving problems.

Science is the search for answers.
Religion is the belief that one already has the answers.
Law is enforcing those answers upon others.

Of the three, at least Science makes no oustanding claims.

I think my SIG says it as best as can be said.

"Everything you know is wrong. (And stupid.)"

Re:ARGH! - this is not A.I.

[DannyO]

For those interested, John von Neumann and Oskar Morgenstern covered this in significant detail in their book The Theory of Games and Economic Behavior (1947).

I analyzed and programmed the modified bidding poker game as part of my Master's thesis in 1988. I created many different players, including the "perfect, but don't bluff" , "always raise" and "always fold" bidders. I also created a player that played the von Neumann strategy. I filled out the roster with a couple of "decent" players with varying strategies.

Over time, I had each of these play. The von Neumann player used the bluffing strategy a percentage of time calculated from the relative weight of the "high" and "low" bids, in conjunction with the value of the hand he held compared to the possible number of hands.

The von Neumann strategy killed all of those players. I then programmed an adaptive player who learned how much he won or lost against each of the players for hands of a given value while playing a particular strategy. As von Neumann postulated, his strategy lost to the adaptive player due to gaps caused by the small differences in rational fractions of his bidding strategy as opposed to the continuous calculations I was able to use with my player.

All in all, it was a great experience to walk in the shoes of such a giant and it gave me a lot of practice with my Turbo Pascal.

Re:ARGH! - this is not A.I.

[Kunta Kinte]

Hmmm, I sense a little tension,... Yes, yes, I think I do.

Jamie. What is AI?

I'd argue that the reporters may be correct to say this is AI and you are wrong.

Maybe AI is not just "learning algorithms", or simply what you'd normally study in a AI CS college class, but AI is "whatever seems to be intelligent, but's not natural". Actually have come across some interviews of very prominent AI researchers have equally loose definitions of AI. I tried to get the real video AI piece from CNN where the director of MIT AI labs gives a similar defination of AI, but I can't find it. It was one of the stories after the recent release of the AI movie.

Any algorithm that passes the turing test, no matter how it was built, is intelligent. Even if the programmer predicted *every* possible reaction to his program and just reacted to that in a human fashion, then his work is intelligent.

To simply put my arguement...

Intelligence is as the intelligent does.

Re:Finally!

[Arctic Fox]

If you did that in a casino, the would haul you off in a heartbeat. Try tapping your foot irregularly at a blackjack table, and see if they don't ask what you are doing.

A friend of mine has written code that can play Minesweeper and FreeCel. It's amazing to watch it play freecel. I think the streak record for it is like 250 games.

can't take it anymore

[Anonymous Coward]

This is really getting to me because I'm seeing and hearing it more and more: the practice of referring to people as thats, not whos.

a cognitive science and computer science student at the University of Georgia that has created a program no, no, no, he's a student who has created a program. All of a sudden everyone is referring to people as thats. Or should I say, everything is referring to people as thats.

Re:java golf

[colmore]

"Computers playing games are always much better than humans if their AI is done correctly© I realized that I had ruined the game for everyone else©"

Interestingly, despite some rather serious attempts, nobody has yet made a good Nethack playing program© And if someone did, I would imagine that it would just be a mess of switches and ifs designed to mimick some good human player's specific style, rather than an algorithmic approach like good chess programs use©

Nethack is probably the closest thing in a non-network computer game to approximating the complexities of AI existing in real world situations, where everything is a special case, and regular patterns and strategies are only minimally useful©

Link?

[nasteric]

Is this program publicly available via the internet? (i.e. download) If so, does anybody have the URL? -n

Re:Link?

[tristan f.]

I clicked on the link in your sig and nothing happened.

Oh wait, I'm running Windows 2000.

Re:It's called card-counting, and if you ask me...

[Stiletto]

10-15 decks? If your casino is playing with more than 6-7 decks go to a different casino!

Why did it loose? General case, or specific hands

[SuperKendall]

That's pretty interesting - did you ever do any examination of the hands played in the real game vs the free one to see if odds for secific cards/hands were less for the pay version?

It seems like somehow you could use a bit of counter-programming to adjust the program to win with the adjusted system. Like you said though, it would probably take a lot of money to figure out what was really going on and develop an effective counter.

---> Kendall

reminds me of my prog

[elpool]

reminds me of the simple program i wrote for class registration at me college. it keeps trying to register for the full classes you want, and when someone drops the class, you are always the first to register. :)

Re:Debunking online casinos.

[Anonymous Coward]

But this article is talking about *video poker*, which if you haven't been to Vegas, is played using electronic machines even in the casinos.

This is a common misconception. The video poker machines are actually miniature kiosks with midgets inside who shuffle and deal the cards. Surveilance cameras are set up inside so that casino employees can look in on and make sure the midgets aren't cheating, but if you're lucky you can find a machine which if you whisper "Hey midget, i'll split my winnings with you" into you will find yourself winning a lot more than probability will allow for. Of course you don't actually have to split your winnings with the midget, he's a midget what's he gonna do? beat you up? Ha!

Re:java golf

[ArticulateArne]

Actually, that sounds a lot like Tom Clancy's Debt of Honor. IIRC (and it was a while ago when I read it), somebody made the right anouncement somewhere and started a trend, which all the expert systems picked up and sent the market into a terminal plummet. The guy had also put some code into the main trading system that kept track of the actual transactions that caused the system to stop logging the transactions, and succeeded in screwing up the entire NYSE really bad, to the point where they had to backtrack and pretend that several days never happened.

Re:Why did it lose? General case, or specific han

[GlenRaphael]

It lost because the low-odds, high-payoff hands didn't come out as often as I needed them to.

My real goal at the time was just to see if the player-bot could be done and to practice C++ programming. Yes, it would be interesting to check whether the free program's odds are accurate. But running under emulation was _really slow_ at the time and the client software tended to crash a lot so it seemed like more trouble than it was worth.

Regarding "developing an effective counter": if the house is cheating at all, there is probably no way to develop an effective counter. It would be stupid to assume the only way they cheat is to tweak the odds in mechanically predictable ways. People have suggested modifying the bot player to make mistakes, but what if their cheating algorithm is win-based rather than strategy-based? It would be simple to have two payout schedules and switch to the bad one only after the player has - through luck or skill - won $500.

Or perhaps the payout schedule was just wrong from the beginning.Drop the chance of a Royal Flush to nothing, and no amount of skilled play will consistently win.

Gamblers Ruin

[sheetsda]

all you have is 10 steady declines of bankrolls

Yup. It's called Gamblers Ruin.

"// this is the most hacked, evil, bastardized thing I've ever seen. kjb"

Re:java golf

[Carbon Blob]

Hmmm.... There is a program for Angband called the Borg that plays it automatically, and wins frequently. Nethack can't be that different...

Re:It's called card-counting, and if you ask me...

[mindstrm]

Like I Said... I'm going on memory. I'm not a gambler. You're probably right.

Gambling is only a sport for people who can't do math.

Re:Personal Experience

[mindstrm]

Though I have no personal proof, I am fairly certain that if this is true, they are making the 'free' game easier, to attract players, rather than cheating on their live gambling. Again, I have no proof of this, but it makes sense.
The fact that you can consistantly win should tell you something; unless you are a card-counting prodigy, you cannot consistantly win.

How a casino audit works.

[mindstrm]

The most common auditor being the very well-respected accounting firm of Price-Waterhouse-Coopers. Would you accuse them of being paid to lie? I doubt it.

One of the *main* things looked at, in these casinos, is the payout rate. You don't analyze every single game and audit the code, but rather, look at how much is gambled -vs- how much the house is profiting. The common average for online casinos is around 98% payout I believe. The audits show exactly how much.
If they were cheating, you'd know.

Re:How a casino audit works.

[aka-ed]

The most common auditor being the very well-respected accounting firm of Price-Waterhouse-Coopers. Would you accuse them of being paid to lie? I doubt it.

I would not accuse them of it, but why are you so convinced that the possibility is ruled out? That is stunningly naive. Didn't they "guard the questions" for one of those rigged quiz shows in the 50s?

I want to get drunk with Hoagy Carmichael and

You may not necessarily have lost..

[CDanek]

Just never truly won. You overlooked the fact that the "fake" or "free" rooms in online casinos use grossly different odds to lure people in.

"Wow, I just won 100,000 bucks.. Man, if this were only the real version... <yells> Honey?? Where's my credit cards?"

Re:and this point is...?

[synaptic]

Actually, the pogo.com games are Java based so he has to be capturing regions of the screen and comparing the blob to a set of blobs he has representing the entire deck to determine which cards were dealt.

I was playing at pogo.com for a while and wrote a short and sweet little program in Perl that would continuously play the spooky slots and keno. Ended up with several hundred thousand tokens that I ended up being too lazy to do anything with.. hrmm. /me goes to enter the contests..

I had thought about trying to expand to the Poker games and actually did a little work on it using Image::Magick but hit a wall somewhere and/or lost interest..

Too bad there wasn't more information about how the "AI" learned to play better. The pogo.com, gamesville.com, and iwin.com games probably have exploitable flaws with randomness or maybe some easter eggs from the programmer. :)

The idea of using this kind of thing to play "for money" online casinos is rediculous though. The amount of cheating by the house going on at these sites must be mind boggling.

Re:Roulette?

[Gannoc]

And a 10% chance of losing $34.

Alright!

[nexex]

Time to take the old laptop to Vegas! I'm gonna be rich!

I did this with real money, and lost.

[GlenRaphael]

I wrote a program that played online video poker several years ago when online casinos were just starting to become popular. I had been making money at blackjack and this seemed like a potentially worthwhile time and money investment. It was very easy to write the program. I tailored my program to a site called InterCasino because at the time they were offering a video poker game that was positive expectation for players with perfect basic strategy.

I only had a Mac handy but the casino client software was windows-only, so I ran the client under SoftWindows emulation figuring this would make it harder for the client to determine something untoward is going on.

Recognizing the cards involved nothing more complicated than hit-testing for colored points along lines with certain offsets. The card pictures never changed, so it was just a matter of partitioning the set appropriately. For instance, a "9" of anything has a pip in the top-left corner; an ace does not. The only thing that was at all tricky there was differentiating a couple of the face cards. For instance, I seem to recall that their Jack of Spades and Jack of Clubs looked very similar to my algorithm. Differentiating spades and clubs generally was hard but it was especially hard with the face cards.

As for the play, that also required no original work. There are people who play video poker professionally; you can order basic strategy charts for all the major games that detail what payout schedules have what expected value and tell you how to play the hands. Some common casino machines are positive expectation and the casinos don't care because the average player makes enough strategy errors to be a net loser. My program had to recognize the card values, categorize the hand, then run down a chart (either a table lookup or a series of if-thens) to determine what strategy to employ. Then it had to simulate a mouse click on the appropriate "hold" buttons, then on the "draw" button.

To tune and debug the program, I had it play the hands but NOT hit draw until I verified the right cards were selected. I played through many hundreds of hands, discovering and fixing several bugs along the way. There were a couple of card-recognition bugs and a couple of play-related bugs, but eventually I got to the point where every time I questioned a play and looked it up in the rules, the computer was right. My program could play faster and more accurate video poker than I could.

InterCasino had two modes. You could play just for fun with fake money, or you could play for real money. To shake out the bugs, I played in the "just for fun mode", letting it run through the night while I slept. I had it preprogrammed to pause at random intervals for random amounts of time to simulate bathroom breaks or whatever, but mostly it just played.

In the morning, I'd won about a thousand dollars of fake money. As long as I played in "free" mode, I seemed to be earning about what was expected, occasionally hitting the big payoffs that make up for the long term negative grind.

So I switched to the real-money mode. I figured it was easily worth investing $500 or so of my blackjack earnings to see if I could institute a real money machine. The dream was to have a bank of PCs in my closet raking in the cybercash while I sleep, eat, read, or go to work... So I deposited the money and set the thing to work. Again I ran it in semi-interactive mode for a while before putting it into automatic.

And it lost. So long as I played with real money, it just wasn't hitting the big payoffs as often as I think it should have. I switched back to free mode and won again, switched back to paid mode and lost again. After losing several hundred dollars I decided to pull the plug.

I did not play enough hands to be sure I didn't have a run of bad luck. Video poker is a game with a very small edge and almost all of the edge you have comes from hitting that once-in-a-blue-moon royal flush or equivalent. Maybe if I'd been willing to lose $10k I'd have eventually gone positive. But there's no way to prove that the game is honest, so I couldn't count on the laws of probability being in my favor. My only protection was to have a reasonable stop-loss figure, and I hit it.

So I won't be retiring to the carribbean just yet. :-)

(In fact, I'm currently unemployed due to a recent dot.bomb event. Need a lead QA engineer, software engineer or otherwise just generally smart guy? My resume is here; check it out. Especially if you're doing something with handhelds.)

Re:I did this with real money, and lost.

[graystar]

I wondering if the cook the cards in the free game. So when you are "practicing" it seems like you win, they suck you in, and take your money on the real game.

Re:I did this with real money, and lost.

[Anonymous Coward]

Gee, then how'd you get hired? Modern law's purpose is to make everyone a criminal. Just because something is criminal doesn't make it wrong.

Re:I did this with real money, and lost.

[mrwhite]

A few friends and I did the same thing. We were all Computer Science students, and, back in 1998, we created a program that would search the cards for a few specified points. I personally lost $250 bucks. :(

It was an interesting program and an interesting experiment, but we just weren't hitting the payoffs as much as we should have and we ran outta money. (College students.)

Re:java golf

[colmore]

the big difference between the stock market and casinos is that the house ¥the fed, the market, the man, whoever earns big when the players ¥funds, politicians, you earn big© It's non-zero sum©

If it could have been better orchestrated, the tech bubble, which grew huge for no real reason, could have kept on going© Everyone involved was making money on IPOs and all that jazz, all they had to do was ignore the failure of advertising and the failure to actually move products and turn a profit©

Re:Debunking online casinos.

[QuantumG]

Why not reply to my post? "Perfect play" is a well defined term in gambling, and in some games skill is actually a factor. I point the interested reader at the rec.gambling faq. We now return you to your regular ignorant trolling.

Exists...

[Speare]

Blackjack strategy cards have been around for ages. They come in laminated business-card size, so that you can put them in your wallet to study them anywhere.

There are dealer rule variations, which change the chart somewhat. For example, dealer hits on soft 17 means that if the dealer has 16, or has a 'soft 17' including an ace, but doesn't have 17 with non-ace cards or higher, then the dealer must hit. This is usually printed on the felt of the casino's blackjack tables, so it's hardly a secret.

There are other "moves" besides HIT and STAND. For maximum advantage, you have to know when to double down, or split, also. You don't need to know anything about Insurance-- it's a bad bet and the house makes a lot of profit from it.

The odds at perfect play are just barely in the house's favor, if you don't count cards. If you do count cards (even a simple increment/decrement on high/low cards), you may swing the odds to be slightly in your favor. The fewer decks the dealer uses, the worse it is for card counting, and the dealer can shuffle more often if they think you're counting. Some casinos will also ask you to leave if they analyze that you're a card-counting savant.

For an example, I just searched Google, and got this link.

Re:Exists...

[JoeShmoe]

Well, the problem with the other moves, like split and double-down is that that you can't continue to track the win count. Splitting is basically playing two hands in the space of one and double down makes your win or loss count twice as much.

In order to track these other options, you would have to use real money, say, bet $100 each hand and then find the pattern that produces the greatest profit.

The odds are always going to be in the house's favor because you always bust first (so even if the house would have busted, it doesn't) but still i think it should be possible to find a series of rules that makes the advantage as slim as possible.

- JoeShmoe

Re:Exists...

[nachoworld]

Your basic strategy chart based on number of decks, DAS, DA2, etc.
The best blackjack site for info.
The most informative site on counting cards for the beginner

Just a few points to help you on the field. Insurance is considered a good bet when the true count is 3 or above. This is regardless of your hand, of course, because whether the dealer has a face in the hole has nothing to do with your hand.

You're right about the odds just barely being in the house favor (at Foxwoods, CT, where I play, they're 0.33% on the 6-deck high limits and 0.36% on the 8-decks).

Which brings me to my last point. Fewer decks is considered better, but what matters most by far to a card counter is penetration, where the dealer places that colored card to signal shoe reshuffle. A penetration that's closer to the front of the shoe, the worse it is for the counter.

BTW, simple increment counting systems, while simple and difficult to get good at, are the systems that the vast majority of counters use.

---

Re:Very cool book about using computers for roulet

[strags]

Yeah, it was an interesting book but, as you state, more so for the descriptions of the builders' environment than the actual machine/theory behind it.

I'm fairly sure their premise was unsound to begind with. Roulette really is a random game - the whole point is that the outcome, much like rolling a die, is based on the composite effect of millions of forces along the ball's path - air pressure, friction, rebounds. They were trying to predict an outcome based on only a few observations, which, furthermore, were subject to human judgement and reflexes!

Re:java golf

[tomblackwell]

Randominity?

Re:OT: Your sig...

[joto]

Actually, that was quite fun, because the machine would lock up in such a funny state. Any key you pressed on the keyboard or the mouse would result in a clicking sound, and even moving the mouse would result in this sound for each "tick". Thus, you could make funny roaring sounds simply by moving the mouse around. It's almost sad they took it away. I'm not sure if it could cause any serious damage, but I guess the press this eventually got forced them to remove the "feature".

One exception - Go

[invenustus]

Computers playing games are always much better than humans if their AI is done correctly.

Actually, computers have, as of now, never been able to master the game of Go. For an intro to the game from the angle of Game Theory, check out http://eksl.cs.umass.edu/~heeringa/home/doc/go_pap er/. A human being with a mastery of the basic strategy can beat most of the best artificial intelligence simulations. Fascinating stuff....
----
"Here to discuss how the AOL merger will affect consumers is the CEO of AOL."

Re:and this point is...?

[joto]

Actually, the pogo.com games are Java based so he has to be capturing regions of the screen and comparing the blob to a set of blobs he has representing the entire deck to determine which cards were dealt.

Or disassemble the java code, and rewrite it with some AI, which, depending on your hacking mindset may or may not be easier than reading the screen and moving the mouse. But I digress... it was a wonderful hack.

Re:Personal Experience

[enjo]

From personal experience I can tell you that the free version of Blackjack for "Casino On Net" is MUCH easier than the actual paying version. I can consistently win with the free version. I can consistently lose when actual money is on the line. While I have no scientific backing, this has been far to frequent to be anything other than true.

pseudo random

[main()]

I have a friend who works for an online casino here in the UK (on the helpdesk phones).

While talking to him, I wondered (and asked, although he didn't know the answer) exactly how these online blackjack/poker/chance games come up with a random sequence of cards.

As many of us will know, it is incredibly difficult (if not impossible) to generate true randosity (?) using a computer (look at TCP sequence number spoofing for an example).

So, given that you are able to observe (with a low stake) the random choice of cards an online game will generate, how difficult is it to predict the next sequence of cards?

I assumed at the time that online casinos must realise this and have some way of defeating it...

?

Si

Re:pseudo random

[Erik Fish]

While the casino is surely not generating truly random numbers I would guess that they are using many different methods of generating pseudo-random numbers and switching methods pseudo-randomly. I'm not an expert, but this seems like a sound method of making the patterns difficult to detect.

Good truly random numbers can be had here:

http://www.fourmilab.ch/hotbits/

Re:Bah

[thefallen]

There is nothing wrong with the assumption that matter (and thus simply a long row of "formal methods") can be used to reproduce 'intelligence' or 'consciousness'. It is simpler assumption than to assume some mystic "soul" for which there is no other proof.

To say that "since you can't be sure matter can be used to simulate consciousness, you must assume it can't be" (like you're saying) is a bit like saying "since you can't know gravity isn't caused by little men with green hats, you must assume it is so".

Until you tell me exactly what is wrong with the mechanistic consciousness assertion, when it fits everything else we know about world just fine, and why I should assert the existence of some "soul", just shut up. Please.

Finally!

[SaDan]

Someone using their computer for something useful! ;-)

Actually, that's kinda neat. I'll have to try it out sometime.

So, what are the applications for this? It'd be neat to see if this program could take input from a video camera, (mounted in your hat, of course), process the data on a tiny wearable PC of some sort, and display the proper course of action to your VGA glasses. Off to the casinos! ;-)

Interested in weather forecasting?

There's no point in rigging gambling...

[Garin]

My background? I'm a Sysadmin for one of the biggest on-line gambling companies in the world. I work in the software development house, advising on sysadmin & security issues. We do casino games, sports books, and parimutuel. Sure, people always whine and complain about companies "fixing the odds" but I'll tell you, we don't even bother. The games that we create are straight odds exactly like you'd get from a fair dealer, perfect roulette wheel, etc. No "real" on-line casinos will change their odds, because there's no point in fixing the games. The real money in online gambling comes from the hard-core players. They're always the ones who bet lots of money, and play religiously. You don't mess with these guys, you just let 'em play. See, the great thing about being running a gambling outfit is that statistics works for you. You win some, you lose some, but after a few thousand customers it all evens out. In order to make money, the casinos don't actually pay out at 100% -- they usually pay out at something in the high 90s. It's not much different to Joe Gambler, but after thousands and thousands of gamblers go through the casino, that couple of percentage points adds up to give the casino a net win. So in a sense, I suppose they do "rig" the games, but they tell you right up front exactly how they do it, by telling you the payouts. Moral of the story? Casinos always win in the long run. They really have no reason to cheat.

Re:There's no point in rigging gambling...

[QuantumG]

If you are actually putting games of skill on the net unchanged, you should know in games of skill the player can get a slight advantage over the house. So maybe you'll get 1000 poor players but you will get 10 or 20 "professional" players who will write programs to play better than a person ever could and they will milk you. Maybe you dont care because you've got those 1000 poor players to suck cash off, but most casinos, including real ones, dont appreciate professional gamblers. This is all old hat, I suggest you read rec.gambling if you're actually in the biz.

Re:There's no point in rigging gambling...

[Garin]

Yes, that's true. But, it all comes out in the end, and we still consistenly pull in that couple of percentage points to make money. I'm not actually one of the math guys that works this stuff out, I'm a Unix geek. I deal with logistics and security. But we certainly don't neglect these things -- there are staff members whose sole responsibility is to figure this out for us. Real-world casinos can get away with neglecting this, because the "pro" gamblers are relatively few-and-far-between. On-line casinos, however, can't. All it takes is one pro with an automatic player running 24hrs/d for a week to make a bazillion dollars. So, yes, we're more than familiar with "perfect" poker playing strategies, and that sort of thing. We do it ourselves, actually. Part of our company's intelligence strategy employs a group that does nothing but play "perfect" poker etc. on all the on-line gambling sites, to see which companies account for it and which don't. Sure, this group (and the automation software) is expensive, but they have no trouble making cash on the side when they find someone that has neglected these details, so they're happy :) When you've got millions of dollars going through the system every day, you catch all the subtleties, and you can easily build up your own statistics on what people really do. We've been doing on-line gambling for three or four years now, and we've put manymany millions of transactions through, so we've mostly figured it out. Sure, there were some slips in the early days, but these were soon recognized and remedied.

Re:There's no point in rigging gambling...

[QuantumG]

Well it sounds like you're on the ball. I was talking about some specific posts to rec.gambling, not in general. You can definitely do some analysis that a regular casino cannot do so you can tune your systems on the fly. If you're not opposed to professional gamblers playing by the rules that you have set, then I dont see a reason to be against your analysis.

yawn

[QuantumG]

I know at least 5 people who have shown me code which does this. It's no biggy. What is interesting is that looking at their statistics it is clear that online casinos cheat when they detect perfect play, but it is very very hard to prove.

Re:yawn

[Anonymous Coward]

Well, even worse, the free versions of their games have different odds than the pay ones -- even though they make them out to be the exact same game. A.K.A. fraud.

Re:yawn

[mindstrm]

Can you back that up?

As far as I know, at least the reputable online casinos have code audits/verifications frequently; if it got out you were fixing your games, goodbye ALL your customers.
How do the 'odds' of poker change from place to place?

Re:yawn

[aka-ed]

Who do you suppose pays the auditors? What is the extent of oversight over these "auditors?" In whose interest do you suppose they operate?

There may be very good answers to these questions, but my instinct says "Follow the money."

I want to get drunk with Hoagy Carmichael and

Re:yawn

[dfenstrate]

Changing the rules on the fly is generally illegal in US casinos- the state gaming boards require certain average payoff levels in all games of chance, and none can be changed after manufacturing.

They go so far as to periodically show up and probe all one-armed bandits and similar devices to make sure they're within the rather strigent rules. If, on teh other hand, you're playing over the net with a company hosted in pango-pango, i doubt they have such zealous gaurdians of the public.

Not bad...here's some better stuff.

[ruebarb]

The only thing new about this is it's interacting with online gambling. Interesting if it could be used in a monetary sense - like Texas Hold Em or 7 Card Stud to actually win money. (if you're playing video draw poker, you're playing to lose. The house has the best odds) - In fact, I half wonder if it's being done right now...people playing online and feeding odds and stats into a computer to see if they've got the odds to win.

FYI - the most commonly recognized poker playing software is Wilson Software - www.wilsonsw.com - they design poker programs with very detailed programmable profiles...That's how I improved my game...a thousand hands a day no problemo. (it's also good for simulation - wondering if you'd save or lose money raising AK preflop? Run 100000 hands and find out!)

The most interesting thing is this...Poker is actually a lot like a casino. You think you're playing each other, but if you get the ten best players together, then all you have is 10 steady declines of bankrolls. The house rake sucks it all away at 10% a hand. The best way to make money at poker is to find a fish - which I'm thinking he could program this thing to beat no problemo

Re:Not bad...here's some better stuff.

[Atlas]

AK preflop is when your two hole cards are Ace, King, like you guessed and the flop has not happened yet. The flop is when community cards are turned over as in Texas Holdem or when you get face up cards as in 7 card stud. For instance in Texas Holdem you are dealt 2 cards face down and three community cards ard also dealt to the middle of the table but also face down to start. You view your hole cards and a round of betting goes on. Then the flop occurs and the three cards are turned up.
--------------------------------------------- -----

Re:Not bad...here's some better stuff.

[Rudeboy777]

So sue me for being a poker newbie, but what is an AK preflop? (I'm assuming AK stands for ace-king)

Real Poker

[NatePWIII]

So is this kind of like Big Blue, I mean if I were to play some of the best poker players with this little programs assistance would I win a majority of matches?

I could see how something like this could really spoil things for Las Vegas...

Nathaniel P. Wilkerson
.biz and .info for $13

Re:Real Poker

[s390]

"So is this kind of like Big Blue, I mean if I were to play some of the best poker players with this little programs assistance would I win a majority of matches?"

In college we played hi-low poker for table stakes and it was... interesting. $100 might change hands in a single "tap" - a lot of money, then. Later I played 5-stud in $2 limit games (that's a loser to the house rake). Win some, lose some, but I do like poker. Haven't played much for years, though. If it's recreation, then don't count the chips. If it's for real money, it takes as much focused concentration as any profession. And as professions go, it's a fairly stressful one. Hacking MVS, networks, Linux is lots easier.

It's also an unfortunate fact that people do cheat at poker. I sat in a low-ball game once where the guy to my left said "watch this" before dealing me a low wheel (A2345, perfect hand). I raised the open and everyone folded; I took the pot and left. When there's real money on the table, the odds of people cheating in various ways go way up. Be aware. There's even more danger of cheating online in virtual casinos, where you don't know which "players" are shills and which aren't; the server (house) knows your cards.

Poker isn't like chess, in that it's not just a deterministic game. A simple poker 'bot will be too predictable (below 50% fold, above 75% raise, etc.). The reason it's an interesting game is that human strategies, emotions, and intuition all play at some times in any session. Poker, like playing the stock markets, is 50% money management and 50% picking plays. The similarities are very striking.

Poker 'bots have been written and played against each other by academics for decades. Best results are obtained by heuristic routines that analyze the moves of all other players and learn from the progression of play, building a 'book' on each from which to calculate odds about bluff or hold. There are many books out about poker (most published to scam wannabe winners out of a few dollars), and not a few academic papers - use Google to find them. Probabilities are easy, strategies are hard, for many reasons.

But poker is a complex game, especially with human factors added. The complexities are infinite and arguably incalculable. There is no perfect way to play. One of the best session strategies is to bluff on small pots early, have 'em on big ones, then later "on towards morning" reverse the strategy - but like all poker, it's risky and won't work repeatedly with the same group of players, like most poker session strategies. Or, one could reverse the above strategy. These are the two basic possibilities for any one game. If one plays consistently though, others will see it and learn when to duck or push. There are levels within levels, in each session and venue.

I've been killed at poker, and I've also shut down a small poker parlor by taking every player until no one would sit down with me (400% profit). One good piece of advice I'll offer is... don't play for blood with friends, or especially, coworkers. Those relationships are more valuable than money, and some people take lasting offense at a sandbag. If they give you hands, OK - but don't extract it.

Re:What about blackjack?

[Erik Fish]

Yeah, several years ago I remember reading a .plan update from one of the id software guys (Carmack?) about how he went to Vegas and card counted his way to a fairly large sum of money then donated it to some charity or other.

He's Lucky he's not an Aussie or....

[jimbojames]

The Australian Government would write a program to automatically arrest him for gambling online
----------------------------------------------

E.R.N.I.E. and random numbers. Re:pseudo random

[matt_wilts]

Years back, ERNIE (Electronic Random Number Indicator Equipment) was the system used to pick Premium Bond numbers here in the UK. (Premium Bonds are basically "lottery tickets" sold by the UK Government - your investment doesn't make interest, but some of the interest gained by the scheme is payed back in the form of monthly prizes - currently the top prize is £1 million)

http://www.xrefer.com/entry/107493

ERNIE used to produce random numbers by measuring the fluctuations of light intensity in a number of neon tubes, in other words it was pretty well "white noise". I believe another similar "random" source is the junction of a zener diode.

Matt

Link?

[pclinger]

The article doesn't have a link to the program anywhere, and neither does the /. story. Where is this guy's Web site so that we can download and try out the program?

But ...

[psergiu]

> and actually reads the cards off the screen
...
> Yes, it does win over the long term as it collects many, many tokens.

The question is: can this program Punch the monkey to win ?
--

And this is why . . .

[hawk]

. . . we have the Nevada Gaming Commission. Every single gaming design approved for use in Nevada has been massively tested to make sure that it does indeed do what' it claims.
(Although about 10 years ago, there was a scandal where someone had a special key sequence that would cause the game to leave the "fair" power up mode and go to a mode that prevented a couple of the big payoffs--tough to find, as when you seize the game, it cuts power) . . .

hawk

OT: Your sig...

[SaDan]

I clicked on that link, but nothing happend.

Oh, wait, I forgot. I'm running Linux. :-)


Interested in weather forecasting?

Re:java golf

[Steeltoe]

Often when the stockmarket starts to get predictable in some way, there's a force in it looking for randominity. The knowledge of a "perfect" computer playing on the stock-market, could disrupt the market into even more illogical ups and downs than it already has. Which is not good for the market of course.

As it is now, everytime an analyst predict upgoing trends on a stock, it's time to sell and vice versa. Of course they use this to sell their own because of it. However, as closed-analysis becomes even more advanced the stocks will seem even more random. So the stockmarket is reduced to a pure gambling pot. What this could do is reduce the incentive for people to gamble on it, slowing down the economy.

Analytical reports are designed to manipulate the market. This is unfortunate because the people being suckered in, as always, are of course the small-time gamblers. It's also unfortunate because what should drive the market is knowledge of good companies to invest in, not obscure ideas in an analysts mind or manipulation to get an unfair advantage. So the system is already cracking up as it is, affecting the commercial psyche. Companies are met with unreasonable expectations every single year, some develop obscure bussinessplans or decides to play hard ball with their customers because they need all the economic growth they can possibly grab.

- Steeltoe

Debunking online casinos.

[ImaLamer]

What someone posted:
"What is interesting is that looking at their statistics it is clear that online casinos cheat when they detect perfect play, but it is very very hard to prove"

The thing is, wouldn't this system prove that it's impossible to win?

I'm sure they use some sort of algorithims to deal the cards and can never really be random. And what dictates 'perfect play'? Counting, in poker? Wouldn't a trained player play similar to the computer? Garbage in, garbage out.

It really proves that to get the best chances, GO TO THE CASINO! At least you can see if they are swapping card on you or something.

and this point is...?

[abes]

The article gives very little details about the actuals concepts involved making it somewhat uninteresting. I'm sure most people here could easily write a program to do the same things. Does it use predefined weighted logic, back progation NN, recurrent NN, alpha-beta, genetic algorithms? Even if it did, making AI games isn't really that interesting anymore, unless someone comes out with a Go player that can beat a master.
<p>
Besides, wouldn't it be easier to just parse HTML rather than try to read the screen? I mean, if the point of the program were to just try to understand writing, <i>that</i> would be interesting. It always irrates me when newspapers write stories like these, when there are many other people doing much cooler things that go unrecognized. At least they should ask someone who knows something if this problem is interesting... Then again, I've never heard of the newspaper, so maybe its all the new they could get to print.

Re:and this point is...?

[James Foster]

The point of reading the screen rather than parsing the HTML, I believe was to make it adaptable to different online poker sites, I suppose?

Very cool book about using computers for roulette

[Anonymous Coward]

The Eudaemonic Pie, by Thomas Bass, about these chaos theorists in the 80's who built computers in their shoes at a time when that was technically a difficult feat (no pun intended). They used a 6502 microprocessor potted in a blob of epoxy along with toe-operated switches and vibrating pins that buzzed different parts of the wearer's foot to tell him/her what numbers to play, and the circuit was stuck into the hollow heels of custom made shoes originally designed for dope smugglers.

The way it worked was casinos used to let you place bets when the roulette ball was already rolling. By clicking the toe switches the first few times the ball went around, the shoe computer could calculate the deceleration rate of the ball and predict where it was going to end up. It couldn't predict perfectly of course, but it was enough to give a substantial advantage over the house.

The circuitry was flaky but worked well enough to prove the concept was sound. The builders claimed they didn't actually win much money with it, but they sound a little bit coy. More interesting is the book's description of the looney, communal environment that the hackers lived in when they built the thing.

Bass's later book "The Predictors" is about the same guys using similar techniques to program computers to automatically pick stock market trades (www.predict.com). They were the first to do that and were apparently pretty successful, but now I think all the brokerages are doing it and there's not much advantage to be had any more from those methods.

MOD THIS TO THE MOON!

[suitcase]

FINALLY, a wesley willis fan on slashdot!

hell yeah!!

suck a polar bears funky asshole!

Re:programs that play well against humans

[eisbaer4]

Thanks for the mention. Poki probably plays better than the average casino player, but it still has a long way to go to compete with the best in the world.

You can now play against Poki (and other humans) on our web applet. Just go to the University of Alberta GAMES group web page and click on the red "Play Online" button in the Poker section. Have fun.

- Darse.

It's called card-counting, and if you ask me...

[mindstrm]

The casinos do everything they can to make it out to be 'illegal' or 'bad behavior'.

Though I can fully understand how they can't afford to let these people play blackjack forever, as they will always, statistically, take more than the house, it's silly to paint them as 'bad' just because they can beat you at your own game.

'Come play our game! Win Win Win! Unless you actually CAN win, then we don't want you here!'

ALso, they don't necessarily shuffle when there is only one deck left; they shuffle at a fairly random point when the pile gets a bit low. If memory serves (and it may be exaggerating on me) she reshuffled when there were about 4 decks left,l and they were playing with about 10 or 15 decks (but as I said, my memory could be playing tricks on me)

Re:It's called card-counting, and if you ask me...

[nyet]

They also have the "infinite deck" machines where delt cards are rotated into the shuffling machine as they are played.

These are sadly becoming more and more common.

Re:It's called card-counting, and if you ask me...

[ardiri]

ALso, they don't necessarily shuffle when there is only one deck left; they shuffle at a fairly random point when the pile gets a bit low. If memory serves (and it may be exaggerating on me) she reshuffled when there were about 4 decks left,l and they were playing with about 10 or 15 decks (but as I said, my memory could be playing tricks on me)

it all depends on the casino/place you go to play in. in Sweden, there are two ways:

mode 1:
the deck is reshuffled at the end of the round in which the 'void' card (normally red) is brought out. when the game is over, the decks are shuffled and the person who got the void card cuts the deck, the dealer then places the void card approximately one deck from the end of the cards.

mode 2:
the 'void' card is placed at the end of the deck, once card follows. when the 'void' card is dealt, the last card is shown to everyone and all cards stay where they are while the dealer reshuffles the remainding cards. the first 5 cards are shown, and put to the side.. dealing then continues.

the rules vary from place to place, and sometimes, the dealer does it the way they want to :) but of course, in Vegas.... thats Vegas!!!

Re:Complete and utter bullshit.

[barfy]

You're right, your response is complete and utter bullshit... There have been several REAL video poker games with positive expectations with correct play. Correct play is often NOT simple, and when video poker machines first came out, the correct plays were not known, and when discovered led to quite a high expectation. (on par with perfect play at Blackjack) For "Play" money video poker often have positive expectations for what should be obvious reasons. Finally, it is a cool hack because of the ability to read what the cards are.

Re:Possibly more common than you think

[catch23]

I friend of mine wrote some cheapy macro scripts in Win9x to beat some basketball shooting game to win himself a Dreamcast.

Re:java golf

[Speed Racer]

Randominity?

It's a perfectly crumulent word.

Re:java golf

[Steeltoe]

Everyone involved was making money on IPOs and all that jazz,

You're right, because you don't make any money before you sell a share. Whoever bought shares without profit, lost on the IT-craze. They got suckered in. However, that's their own fault for being greedy.

- Steeltoe

Re:OT: Your sig...

[Jester998]

It's supposed to crash your machine. Anytime Windows 95 and 98 (First Edition) tried to access a directory with two device names in it (\con\con, \nul\nul, etc), it supposedly crashes the OS. I've never seen it happen, but I've seen the postings on many security lists.
Apparently, it can be used to crash a system remotely too, if they're running an FTP daemon (I know the security bulletins speicifically mentioned Serv-U FTP was vulnerable.)
I think that this problem has been fixed in Win98 SE, because my Winblows machine doensn't do anything when accessing those types of files/directories either. Of course, none of my Linux machines have this "feature". :)

Home town hero.:)

[Kintanon]

I live in Athens, in fact, I live on the UGA campus. I'll have to hunt this guy down and talk to him....
Just basking in my indirect association with a now famous person.

Kintanon

Re:java golf

[tomblackwell]

Sounds like a mathematician's gum to me.

Re:Very cool book about using computers for roulet

[jefflinwood]

Unfortunately, the Eudaemonic Pie book is out of print now. I tried Powell's, and Amazon.com. I thought Predictors was interesting enough (even with the whole VC/stock market stuff), and the roulette story would have been even better. Anybody have any good bookmarks to casino cheating stories? Probably easier now with PalmPilots than back in the day with giant boxes strapped to your waist and leg.

But does it know...

[mikosullivan]

... when to walk away, and know when to run?

Does it know that you don't count your money when you're sittin' at the table? (There'll be time enough for countin' when the dealin's done.)

Miko O'Sullivan

Time's Arrow

[Dungbutter]

The first thought that came to mind was a scene in the Star Trek: The Next Generation episode "Time's Arrow Part 1",
where Data plays poker against huslters in 19th century San Francisco.
He ends up winning many hundreds of dollars.

God, what a geek I must be: someone mentions poker, and I think of Star Trek!

The resume, and the PDA industry in general

[GlenRaphael]

Great... we hire him, and we're next on his list [to fold]

Regarding companies folding: It's really bad out there now, but it's been particularly bad in the PDA field, which has been my specialty until recently. The last four major companies I worked for (the Newton Group of Apple, the DataRover group of General Magic, PocketScience, and Red Jade) are all dead or on life support today. Generally I left a company because I could see the crash coming despite my best efforts to prevent it. Although in one case -- the Newton group at apple -- I left because I was actually optimistic about the product and wanted to have a bigger stake in its success than I could get as a low-level Apple employee. (So I became employee #4 of LandWare, a company which narrowly managed to survive the Newton crash by diversifying into Palm and Windows CE).

My logical next move would be Palm or Handspring, but they've both got hiring freezes on and Palm just had a couple rounds of layoffs. Had I gone the Windows route it wouldn't have been any better; none of the major players in the early days are still thriving today. The only player that has been somewhat stable throughout is Microsoft itself. I've thought about that too, but on the other hand, "once you start down the dark path, forever will it dominate your destiny." :-)

i'm no expert at resume writing but yours could use some work...[suggestions]

Thanks for the comments. I've changed some of it. I usually save the "what I want to bring to YOUR company" for the cover letter and just let the objective say what general area I'm interested in. But what I bring is the ability to spot and solve problems the company doesn't know it has. Problems with the product, problems with the process for producing it. I am really enthusiastic about producing a high quality user experience and have the tech background that lets me understand and figure out how to help the engineers fix the problems that might prevent that.

Something will turn up. And if not, there's always acting! (In the last couple months I worked as an extra on the two Matrix movies currently being filmed. I left that off the resume since I couldn't figure out how to make driving on a fake freeway relevant enough to a real job to be worth the distraction factor.)

And bluffing?

[Stonehand]

I'd think that'd be where the *real* research issues are -- with the added complexity of knowing when to check, call, raise or fold... Not just, say, figuring out which cards to replace...

Re:java golf

[Perdo]

get this: macro express for windoze. I can't even program. I just made it mimic me playing the game then optomized the timings.

programs that play well against humans

[Silmaril]

Darse Billings and others at the University of Alberta Computer Poker Research Group have done quite a bit of research on computer poker. Their program, Poki (source here) plays a reasonably good game of Texas Hold-Em. This is the primary game at the famous World Series of Poker.

Vegas blackjack for business travelers

[Go_Ask_Alex]

For those of you that regularly get stuck in Vegas for days... Yeah, it's possible to win as long as you know how to bet and when to walk away. Winning involves basic strategy, some degree of "counting" (more like trend analysis with multi-card decks) as mentioned, as well as a betting strategy. You don't want to run out of money, plain and simple. Casinos make their money when some idiot comes up to a table with $100, places $50 bets, losing it all within 2 to 4 hands. A general observation from actually playing on the strip in Vegas and a handheld unit for practice was this: start with a large pool of money (say $100) and place bets of 1/20 of your pool ($5 of your $100 pool); if you play basic strategy, not even counting, your pool will fluctuate about +/-50% (roughly). Say your up to $150 having started with $100, you likely won't go higher. Say you're down to $50, keep your cool with the $5 bets and basic strategy (don't deviate), and you'll likely be back up to $100 with time and patience. Some years ago I had a DOS based blackjack program that also ran simulations, card counting and basic strategy exercises. I threw my observation at the DOS simulator with 1000s of hands played and my basic betting strategy was roughly correct. In the simulator the house still has the advantage and you will eventually bust, but in real life you can increase your bets as you are winning, decrease your bets if the trend looks bad and maybe supporting a decision to just walk. Eventually you should get comfortable with blackjack to play to float around your pool. So say your pool is up to $150 from the $100 you started with. $7.50 bets are a pain, so just add another $50 out of pocket and you now have $200 to play $10 bets out of (always remember that it was ultimately $150 out of your pocket). Give it 30 to 60 minutes at a fast table and you'll be up to $300, then you can start doing $15 dollar bets, and so on. I try to keep the chips that represent my out of pocket money separate, so if I do hit a bad losing streak I'll easily know. That's when I leave, but usually after a few hours of playing and lots of free drinks (minus tip money) before I call it a night. When in Vegas for days I allow myself to lose up to my trip Per Diem. I have had only one net loss of a couple hundred bucks (I played craps), while on all my other trips I either broke even or more often came out ahead $300 to $700 (my first trip was +$500), subsidizing my sidebar trips to Grand Escalante (slot canyons, gotta go), Zion, Bryce, etc. This does take hours, trending of the shoe, and playing tips for especially good dealers. I'm not advocating anything that'll get you rich; I still stick to $5 tables, $10 if I have to, although I'm getting the confidence to move up to the $25 tables. If you know basic strategy you should be able to play for hours at least to break even, or after hours not loose much in the process. The entertainment comes from finding a cool table and meeting different people from all over the country, maybe from around the world, some boring, some really interesting, and getting free drinks too. Never play craps even if playing along with a bunch of old guys that act like they know what they are doing. Any other game is a waste, although I may be able to apply a similar theory to slots, namely those machines that have lower payouts but are perceivably looser (pay out more often); definitely not the networked machines with $million jackpots. Many of the slots are set to provide hours of entertainment to one person who never moves from the same machine, up and down, eventually taking it all away. In one experiment so far a friend and I came out ahead from a networked group of nickel slots, playing for the smaller wins and moving to the next machine.

java golf

[Perdo]

While not as great an accomplishment, I wrote a macro to play a java golf game online. Since there was some randomness to the game and my reflexes are not that good, I figured it would be a way to get my name in the top ten. after three days running the macro, I had all ten of the top ten spots. The game was pulled on the third day. It has never been back. Computers playing games are always much better than humans if their AI is done correctly. I realized that I had ruined the game for everyone else. Deep blue beat Kasparov, My AMD box beat java golf, Aimbots beat me at TFC, Paul Apostolik beat pogo poker, and the power producers are beating California (with a great weather model/consumpion prediction algorithm) on the energy trading floors.

Are we are all cheating? Cheating only applies to games and applying enourmous computer power to commodity markets is simply shrewd? Will we one day wake up with zero value in stocks because some bright day trader wrote a miraculous piece of code that makes a billion perfect trades in the course of a day? Cheating by attacking protien folding with petaflop power?

What are the moral causes for the application of analytical force? When is the right time to spray for moths? When is the right time to let a hummingbird fly over japan? (obscure reference, chaos theory: the ammout of air a hummingbird moves in flight over japan is all that is required to generate a hurricane a month later in florida)

With IBM predicting a teraflop available on your (or Saddams) desktop by 2004, these questions need to be addressed yesterday.

Re:java golf

[K.]

When is the right time to let a hummingbird fly over japan?

Not a problem that could be solved with current
analytical methods, even if you threw the
computing resources of the planet at it.

K.
-

Complete and utter bullshit.

[nyet]

The ONLY "house" game you can play and WIN at with perfect play is Blackjack. You can learn to do it by counting cards, or you can program a machine to help you do it. Both are against most states' gambling laws. Mathematicians MUCH smarter than you have already analyzed Blackjack to death, and the "counting"/"tens richness" strategies are well documented.

There are no other games, other then the odd progressive jackpot game (when the jackpot becomes high enough, the expected value becomes positive). AND this already assumes "perfect" play, which the average human playing video poker can easily learn and be 100% accurate at. Video poker is an EXTREMELY simple game.

Craps is close with pass/don't pass come/don't come, but it is still negative. There is NO way in hell any gambling machine (unless poorly programmed) would give a statistically postive payout, unless the programmers are complete idiots.

I see so much uninformed BS flying around it is fucking amazing. Slashdot may be a good place to get facts on computer/technology based topics, but when it comes to statistics and gambling.. well this is just plain pathetic.

All video poker machines have a house advantage.

[nyet]

I can only assume that the programmers of the online video poker games are morons.. all COMMERCIAL video poker machines have net house gain.. tweakable by the individual rules. Video poker machines offer among the WORST payout expectation in casinos, except for certain progressives.

And you don't need a computer to play "perfect" video poker. It is an extremely trivial game.

Possibly more common than you think

[Anonymous Coward]

I have a friend who has written several perl scripts to automatically play online games, the kind that reward you with pseudo-money. (Sometimes you can trade the pseudo-money for prizes, sometimes you can use it in auctions.)

Some of his programs are quite primitive; he randomly 'clicks' on links that could pay off, and a small percentage do. He wins simply through the massive number of clicks.

But he has at least one program that plays a trivia game. If the question is new, it guesses, but it "learns" when it has won the question. If the question ever comes up again, it now knows the answer. Pretty impressive for a few hundred lines of perl. (And he's not even that good of a perl programmer!)

He has been caught several times, and usually gets a 'cease-and-decist' letter (for violating the terms of service you have to sign to get an account), but has never gotten in any real trouble. (A key to not getting caught is to never do TOO good; never be the top winner of the day.)

Moral of the story: Programmatic attacks on games go on all the time, and my friend has dozens of free t-shirts to prove it.

Posting anonymously to protect the guilty.

Re:OT: Your sig...

[SaDan]

Yeah, I know. I've just been working on my computer for about four hours straight, and drinking heavily. My humour gland is not functioning properly, or something, because that reply seemed funny at the time.

Oh, well. Don't drink and post people... Innocent karma gets killed that way.

Interested in weather forecasting?

Solitaire

[rchatterjee]

Good thing this thing doesn't play solitaire for you, if it did what would all the execs and marketing people at my work do all day? Not actual work that's for sure. :)

What about blackjack?

[neonstz]

It would have been more interesting with a program that plays blackjack. In blackjack it's actually possible to win. There are people who're able to to this (without computers), I read about one guy in a norwegian book about games by the sci-fi writer Tor Aage Bringsvaerd, and that guy was banned from several casinos. Doing this requires extraordinary memory and math skills. You have to remember how many of each card that has been drawn during the game and calculate probabilities in your head (well, it's actually a bit easier since a lot of the cards are worth 10 points, so you don't have to remember everything). The clue is that in blackjack the dealer uses more than one deck of cards (which are all shuffled together), and he doesn't shuffle the cards until there is only one deck (52 cards) left (correct me if I'm wrong, I'm not sure about the exact numbers).

Re:What about blackjack?

[Sycraft-fu]

In blackjack it's actually possible to win.

Not in a casion. Why? They pay 3 to 2. Yes, supposing that you play for 2 to 1 odds and the dealer plays by normal dealer rules (hold on 17 and over, hit on 16 or less) and you are good, you'll make money. However casinos pay 3 to 2 odds. So if you bet $100 you only get $150 back. That's enough to tip the odds back in their favour.

Re:What about blackjack?

[ardiri]

Not in a casion. Why? They pay 3 to 2. Yes, supposing that you play for 2 to 1 odds and the dealer plays by normal dealer rules (hold on 17 and over, hit on 16 or less) and you are good, you'll make money. However casinos pay 3 to 2 odds. So if you bet $100 you only get $150 back. That's enough to tip the odds back in their favour.

sorry, have to correct you here - they do pay 3:2 for blackjack (ace + ten) - but, if you bet $100, you actually end up with $250 (+$150) - not as you originally say with +$50. if you win a hand, then you get 1:1 odds, which means if you bet $100 you get $200 (+$100).

Macroing

[Anonymous Coward]

I used macroing to win on Jackpot because I had thought it was a total scam (this was last year, in the spring). One morning, I woke up with notification that I had won 5 bucks.

I didn't believe I'd get the check, but it DID come, amazingly enough. As long as you aren't using your computer, it got a little money using your idle cycles. Of course, calculated out by time, I made about 8 cents an hour, but that's as much as a Nike employee makes.

My friends in the dorm did similar; one won one check for $10.

Cheaters sometimes prosper.

Impressive

[sg3000]

> After losing several hundred dollars I decided to pull the
>plug.

>I did not play enough hands to be sure I didn't have a run
>of bad luck. Maybe if I'd been willing to lose $10k I'd
>have eventually gone positive.

Wow, it sounds like not only did you write a program to simulate video poker playing, but also to simulate a gambling addiction!

Re:Impressive

[GlenRaphael]

>Maybe if I'd been willing to lose $10k I'd
>have eventually gone positive.

Wow, it sounds like not only did you write a program to simulate video poker playing, but also to simulate a gambling addiction!

No, the fact that I cut my losses at something that for me was perfectly affordable indicates I was following a rational speculative investment strategy. I decided upfront how much I was willing to lose, and when I lost that I quit.

But skilled players who want to make money playing video poker really do need a huge bankroll and need to be willing to withstand a long losing run before they go positive. That's not wishful thinking, it's just the way the math works.

To quote from the rec.gambling video poker FAQ:

The main difficulty with playing video poker is that it takes an average of 60 hours of rapid play to hit a royal flush, and it takes a _huge_ bankroll to survive long enough to win. During this time, the casino enjoys an advantage of approximately 5%. Straight flushes can be expected about once every 6 hours on average, but these contribute only about 0.5% to the player's return. 4-of-kind hands occur only about once per hour, and these hands account for about 5% of the player's return.

What this all means to the video poker player is that you will be playing with about a 10% disadvantage while waiting for an occasional "boost" from a 4-of-kind or straight flush. On average, it will take a bankroll about as large as the progressive jackpot to survive long enough to hit the royal flush (and this assumes that the jackpot is large enough to give the player a reasonable edge over the house).

In an honest game my "expected income" was positive but I had a low chance of a high payout and a high chance of a low loss, unless I was willing to invest many thousands of dollars. Were it just a matter of the money, I could have formed a partnership to spread the risk. But given my suspicion of cheating and the lack of reasons to think the game is honest, the smart thing was to just let it go.

My Gambling Adventures

[Jesus IS the Devil]

Once a blue moon I had an idea. Here where I live casinos aren't allowed to play as the bank (win people's money), so they have to let the people be the bank, rotating every few turns. These casinos make $ by taking in a "collection fee" of $1 per hand.

Anyways, some of the games definitely gave the "bank" an advantage. Pan-9 because you could see if people are hitting or not. If they're not, and you have a low-scoring hand, you'll know to hit. I'm too lazy to get into the details, but trust me, there's an advantage for the bank.

Then there was Pai Gow. Here the bank has this "copy advantage". If a player has equal rank with the upper 2-card hand as the bank, the bank will win that set. Blah blah...

One more thing. You'd also have to factor in the collection fee into your chances. You'd need to put in enough money per hand to offset the disadvantage created by this fee or you'll lose. Also, I made a vow to NEVER tip those greedy-ass, good-for-nothing, I'll-sell-your-mom-just-to-make-a-dime dealers.

Anyways, I wrote this program to simulate playing the game from beginning to finish. I ran in on an Apple II GS. Yeah very primitive heh? That's all I had at the time. I did this just to prove that I was right that the bank did have an advantage.

Since that computer was slow and lame and didn't have a hard drive, I had to have everything printed out to me for record-keeping.

So day and night this Apple ran. A few days go by so I decided to analyze the results.

What do you know? There were distinct patterns repeating over and over again. Then it occurred to me, the damn computer's random number generator wasn't really random after all! I had wasted my time with that cheap computer.

So anyways, off I went to the casino to test my theory out. With $3000 in my pocket I started gambling away day after day... Up, down, up, down.

I was actually up $5000 for a while (and this was after I'd logged over 150 hours of gambling), but then I got greedy and started betting too big. This proved to be my fatal mistake. Wham in one night it was all gone. Just like that... After a month of hard work it all disappeared. >(

Anyways, I'd gotten sick and tired of the constant smoking and rude atmosphere. Besides, I wasn't learning anything and it was a HELL of a job (if I can call it that). So I quit cold turkey and went back to school.

To this day I'm sure there's an advantage there. Maybe one day when I'm really really bored or really really rich I'll try that again.

What a life.



---------
Did you just fart? Or do you always smell like that?

Give me your email...

[thefallen]

...I won't sign it up in goatse.cx mailing list, but it'd be much more convenient platform for, mmm, discussions.

I said nothing of soul but please show your proof of it's absence or that it is "simpler." There is no reason to think you are equipped to explain the world or even to comprehensively describe that which you can measure. This is very different from a soul.

You didn't say soul but I'll bet my ascii doughtnut you sure as hell meant it. Mechanistic model is simpler because so far everything else in world can be explained with mechanism. So, we have two otherwise equal explanations for consciousness, other relying on mechanism which is for other areas proven and working method, other is something mysterious which doesn't manifest itself in anything else and which even its proponents don't really know anything about. That's what my analogue was about; we have two models for gravity, other fits nicely with everything else we know, and other just conjures up something ridiculous (little men with green hats) and says "it could be this!".

Oh, really? Explain consciousness then.

What's the to explain? Person receives information through senses. Person processes that information with a complicated neural network. Person sends signals to muscles to interact with the world. The problem is? (and don't give me that "but where comes the I feeling then?!" crap... feelings are just internal feedback routines in our brain, and that I feeling is fundamentally no different from that dog is biting my leg, it hurts feeling.)

Fuck off, Mr. Science. If I wanted to listen to some moron spouting his faith, I'd go to church. Please learn a thing or two about the foundations of your beloved science.

Am learning all the time, while you rant about why mechanism is flawed without even bothering to tell us how it's flawed. And very convenient quote you picked; the "just shut up. Please" part had a fucking conditional. By picking up that part you basically say "no, I won't" to my request "please tell me what is wrong with the mechanistic consciousness".

Why do you think it is random?

[Ord]

There is no reason to believe any of the following: a) the deck has 52 cards b) every card in the deck has an equal probability of being selected It doesn't matter what random number algorithm is used, these two possibilities greatly outweigh the random number generator's effects. It's entirely possible to make 4 of the 5 cards necessary for a royal flush come up say 50% of the time, and then make the 5th card come up something like 0% of the time.

Are all2� the keyboards br///oken there too~

[Owensellwood]

I don't want to go to your CS department if the keyboards aren't broken like this guy's. Debugging inline assembly is no fun w/out a broken keyboard> This post is © Kyle.

Why I am convinced.

[mindstrm]

A couple of reasons.
Firstly, because it makes logical sense. Yes, they COULD be rigging all the games, but if the casino still maintains a 98% payout (ie: They are taking, on average, 2% of what's gambled), I find it hard to see how games are 'rigged'. If they fixed one game to win, they'd have to fix another to lose to keep the numbers working, and what would be the point?

Thirdly, because I work for an online casino/sportsbook, and if I thought for a moment they were rigging the games, I'd leave.

And finally, because there is not NEED to rig the games to win. Profit margins for online casinos are higher than those of meatspace casinos; overhead is much lower, and if it ever got out that the house was cheating (and it WOULD get out) you could say goodbye to your business.

Also, if this was some unheard of firm, I would have suspicition. These casinos go out of their way to have their things audited by the most respected firms out there; those firms do NOT get respect by being anything but impartial; they certainly wouldn't ditch that respect for an online casino; they'd stand to clients that are orders of magnitude larger (like the Fortune 500 companies, say). Online casinos are generally NOT multi-billion dollar operations like you might think.

My cheating experience..

[Sarin]

I used to play a game called alien online where you raise stats on your alien by battling other peoples aliens etc. like a bbs game there were many sub-games where you could make money to buy better equipment. There was a nice hangman game, and after playing it a couple of times I realized I sucked at it. So I wrote a program (costed me a couple of days of freetime) in PHP using the excellent Snoopy class file (to navigate over the pages) that could beat any hangman game not by storing all answers but by generating an answer using a large dictionary and making intelligent guesses, my success rate was 85%, not bad.. After running the script for 2 hours I realized their dictionary was very limited and there were only 500 guesswords so I could only make something like 1500 gold.
I hadn't thought of that, I would had made far more when I just played regular. But my friends were impressed and we had a good laugh.

ANd yes..

[mindstrm]

I made several grammatical and spelling mistakes; I'm not fully awake yet.

That last bit should read 'They stand to lose clients that are orders of magnitude larger .....'

Roulette?

[mecredis]

If anyone has played roulette, they will know its quite easy to make a couple bucks after playing for a LONG time. Even though the stakes are against you, if you bet on 34/36 of the numbers, you have about a 90% chance of winning 1$. The only problem is, it takes a very long time to accumulate a decent amount of money.

This is where a computer could come in handy. Just repeat the steps over and over again (changing the unbetted number every so often) and eventually, you'll end up with some cash. Run it every night for a month, and you'll end up with a good hunk of change, around 700 bucks. Then the only trick is to prevent the online casino from figuring it out..

-Fred

online Poker a cheats heaven

[Tarquin Sidebottom]

I've never played on-line poker for money, and I've often see advice saying that you should just stick clear even if its a reputable 'casino'. Apprently its common practice for groups of friends to all play at the same table and share their card info. Imagine the results if you did that using this program!

Re:Very cool book about using computers for roulet

[Chris Hiner]

One time in Vegas, while playing roulette, I had a dealer tell me how he won several thousand dollars while playing at another casino. He had some friends in from out of town, and of course they wanted to go gamble... He noticed that the dealer wasn't changing the wheel speed, and wasn't adjusting his ball speed like they're supposed to. So apparently the outcome became much more predictable that it should have been.

One reason why the house wins

[WyldOne]

I saw this explained somewhere. Dont have a link but here goes.

Roulette: 100 numbers 2 numbers house wins always. Laws of chos tells us that the house will win at least 52% of the time. (ave. 50% + 2% for the house always win) This is for long term runs

Apply the same priciple for 21. AFAIK the house always wins a 21/same count hand if you both get one. This is the percentage buster.

I would not be surprised that the games are looser for 'fun' vs. 'real' money games. (How would I know/prove?) and to make you think you might win.

It worked for me.

[perlchimp]

The first real useful application of programming for me was writing programs to determine optimal seating positions or hands for all those made up poker games like 'Screw Your Neighbor'. I'd brute force the games on a couple of million decks and later use the stats to fold, stay or just sit out. It was time well spent.

All this talk of gambling addictions...

[nick_davison]

All this talk of gambling addictions, I'm concerned I may have a slashdot addiction when the first thing that comes in to my head is... "I wonder if I could write a PHP script that analyses the topic of each new subject then generates a karmawhoring reply for me?"

Definitely time to go cold turkey. I'll just read one more story before I go.....

Re:All video poker machines have a house advantage

[GlenRaphael]

I can only assume that the programmers of the online video poker games are morons.. all COMMERCIAL video poker machines have net house gain...

I'm sorry, but your information is not correct. The "full pay" version of Deuces Wild and of Jokers Wild have net player gain with optimum play.

Video poker machines offer among the WORST payout expectation in casinos, except for certain progressives.

This might be true generally, especially for players who don't play perfect basic strategy. But players who are very careful about game selection and learn the proper strategy can indeed find games with a slim edge for the player in Southern Nevada. Perhaps there aren't any positive expectation games where you live, but that doesn't mean they don't exist.

(Another way people make money at video poker is to find a casino where the slot club offers money-back bonuses that make up for the house edge.)

The online casinos were just offering the same sorts of games that are found in the casinos. The game I attacked was a full-pay Deuces Wild.

And you don't need a computer to play "perfect" video poker. It is an extremely trivial game.

It's certainly true that you don't need a computer. You do need a chart or table of rules to follow which is different for each game and somewhat sensitive to the payout schedule, but it's possible to memorize this table, and many people have. (I haven't, and I doubt you have either.)

Anyone who is really interested in becoming a video poker expert will probably want to do one or more of the following:

Read the rec.gambling video poker FAQ
Buy Dan Paymar's book Video Poker Optimum Play
Subscribe to the newsletter Video Poker Times