Slashdot Mirror
Slot Machine with Bad Software Sends Players To Jail
dcollins writes "Previous discussions here have turned into debates over who is liable for faulty software: the programmers, the publisher, etc. Yahoo has a new option: perhaps the users are criminally liable for using the software. From the AP: 'Prosecutors are considering criminal charges against casino gamblers who won big on a slot machine that had been installed with faulty software ... A decision on whether to bring criminal charges could come in a couple of weeks, said John Colin, chief deputy prosecutor for Harrison County. He said 'criminal intent' may be involved when people play a machine they know is faulty.' Would your average user be able to distinguish 'faulty software' from 'lucky'?"
The Nevada Gaming Commission [PDF] (As an example, I know the article isn't about Las Vegas) heavily regulates slot machines, their software, and their payout schedule. Machines that deviate from the payout schedule are inspected and machines whose software processes are not open to inspection and audit are not allowed on the floor. In this case it would be, prima facia, a crime to install software that was not audited by the authorities onto a machine. IANAL, so I can't tell you if proving criminal intent would be required, but I suspect that the threshold would be minimal, assuming that it could be proven that the users inserted the bug.
In this case, it doesn't appear as though the bug was inserted by the users, just (sigh) exploited in order to win. These cases are well litigated in Nevada (though probably not in Indiana/Kentucky), and elsewhere. The trend seems to be (Scroll Down to "Overpayment to Patron") that if it can be proven that the gaming patron didn't involve him or herself in the actual flaw of the machine, then not only are they not liable, but the Casino must still pay out the winnings.
Yoda of Borg am I! Assimilated shall you be! Futile resistance is, hmm?
The woman who reported the machine to the casino has it right - the casino doesn't give you your money back if a machine jams, so it shouldn't work the other way.
I work in a casino repairing slot machines at the moment, and yes, if a machine jams, the patron DOES get their money back. The casino by law cannot pick up change off the floor nor can they keep any money that was put into a machine but the machine didnt register. The patron will always get their money back after an investigation is completed and it is determined that the money really does belong to the patron.
This should be treated much like counting cards at blackjack. If you do it in your head with no mechanical help, you are playing within the rules, and you are not cheating.
The issue of whether or not card counting constitutes cheating was and is a hot topic in the gaming industry with strong opinions on both sides of the issue. The courts have consistently ruled that it is not cheating, from the legal definition of cheating (i.e. the one that results in a criminal prosecution), to count cards provided that one is not assisted in this endeavor by any sort of device (i.e. mechanical, electronic, electro-mechanical, etc). The casinos on the other hand, not surprisingly, consider all forms of card counting, even the type that courts have ruled legal as "intelligent play" (i.e. using your brain), as "cheating". This is where it gets interesting. In Las Vegas the casinos have the right to refuse service to anyone for any reason and it is not unheard of for card counters to get the back-room treatment (i.e. casino security goons try to verbally intimidate you into not coming back again and in the old days that did more than just talk if you know what I mean). However in Atlantic City they are not allowed to refuse someone a game, provided that the casino is open to the public and the player is not creating an overt disturbance, so they do things like switch decks frequently, deal from multiple decks, deal only partway through the decks, use automatic shufflers, etc...to discourage card counting.
The gaming industry is not a nice industry to be in, either as a player or an owner, being classically considered as a vice industry which means that you are going to attract many of the "wrong sort of people" if you know what I mean. The casinos deal harshly with anyone they perceive to be hurting the bottom line precisely because the entire atmosphere of gambling is already highly adversarial and involves the one thing that everyone wants more of...money...and lots of it. Personally, I don't much care for gambling, even as a form of entertainment, but if you plan to try card counting then all I can say is watch out...either you wont be any good at it and the casinos will encourage you to "count" more often or you will be in which case you will end up in their database and be bared from just about every casino on the face of the earth.
Gambling machines are weighted to weighted to cheat in favor of the casino.
Gambling machines are highly regulated in the states where they are operated. The machine processors are checked by the state and held under lock, key, and surveillance until installed in the machines which are also under lock, key, and surveillance. The people who service the machines are certified by the state. Those employees fill out about an inch of paperwork when they are hired in order for their state and federal background checks, which includes taking and running their fingerprints through the criminal databases. Cheating and allegations of cheating are taken very seriously by the states and the feds in this business, which at one point was run by the mob. The machines do not cheat.
That said, those machines are programmed to benefit the casino, which is not the same as cheating. The casino actually sets the payout amount to a certain percentage of every dollar, in accordance with state law. For example, machines that bring in little money for the casinos, such as penny slots, generally have the worst payout to the player. For penny slots, that payout could be $0.85 for every $1. For more profitable slots, such as $5 slots, that payout ratio could be $0.97 for every $1.
The payout ratio depends on the casino and what they want to do. If they want to attract customers, then they increase the payouts. If they have a good amount of customers, then they'll decrease the payouts to increase profits. Customers are very sensitive to these payout amounts. If players sense a casino is being cheap, aka "the slots are tight", then players will leave that casino. If the players sense a casino is giving money away, aka "the slots are loose", then those players will get out the word and people will flock to that casino. It's all supply and demand.
I worked for a lottery corporation for a short time after college until I found a real job. I was offered a pretty good paying job in the "slot depot", but turned it down strictly due to ethical reasons.
As far as I understood it, all the slot machines would be programmed for a 95-98% pay back to the player, but there would always be one or two in a casino that is programmed to a 105% payback. The higher pay machine would always be changed so that if you went back for you "lucky" machine, you might not get it.
Nasty. That's why I didn't take the job.
Using more decks doesn't alter the relative proportions of cards (and hence the probability of any particular value turning up). In fact, if anything, it makes the probabilities conform better to the empirical expectation. Since card-counting is all about calculating the probability that any particular card will help you (by increasing your score without busting you) or hinder you (by taking you over 21), anything that makes the observed probability closer to the expected possibility favours you.
Most probability studies assume that the random-number generating mechanism has no memory. This is usually correct. Throwing six sixes with a die does not alter the probability of throwing a six next time: it's still 1/6. But when playing 21, there is a sort of memory effect going on; because cards that have already fallen will not show up again. And it's upon precisely this memory that the card-counter relies. Once KS has been drawn, the probability of the next card being KS is zero. If you have an infinite number of decks (equivalent to returning each card to a random position within the deck after use), the probability of any random card being the KS is always 1/52. With a finite number of decks, the memory effect is reduced as compared to a single deck but not eliminated altogether.
You can memorise the order of a single deck, but that's not the way most people do it. The "classic" method is mentally to divide the card ranks into "high" (8-K, likely to bust you whatever you've got), "low" (A-3, good for completing a five-card trick) and "middling" (4-7). Now you know in any deck there are 24 high cards, 12 low cards and 16 middling cards. By knowing how many cards within each band have fallen, you can determine how likely you are to get a card you want. If, say, a bunch of high cards come up, it's not unreasonable -- because of the memory effect -- to expect the next card to be low or middling. And you can bet accordingly; low when the cards don't favour you (or when they favour the dealer), high when they do (or the dealer stands a good chance of being busted on the next card).
The only way to disrupt card-counting (unless you have an infinitely large casino with room for an infinite number of cards; but then, you'd have no room for any players to sit at the card table -- even if you had an infinite number of seats, they would all be full of nothing but stacks of cards) is to return each card to the deck immediately it has been played, and always draw each card from a random position (or shuffle between each deal).
Je fume. Tu fumes. Nous fûmes!