• Welcome to the new Internet Infidels Discussion Board, formerly Talk Freethought.

The odd of winning... when computers are involved

When someone is on a loosing streak and leaves a machine that has wiped them out, that machine is 'hot' and is likely due to payout (it has to, eventually, by law).

Really? Do you state that there us a regulation stating that each nachine has to payoff a certain amount per day?

In short, yes. They are programmed to ensure a specific payout over time. The laws of numbers relating to the Gambler's Fallacy apply only in a pure random environment. The Gambler's Fallacy is no fallacy at all with respect to electronic slot machines, due to this law.
 
shuffling after each deal would reduce the overall take, since the total number of games per timeframe would be significantly less.
their solution to card counting is by utilizing multiple decks.

In games where multiple decks cannot be used (poker - 5 of a kind, anyone?), the pace of the game is mitigated by the odds.

For electronic games, the payout is regulated by the gaming board. The game programmatically ensures a specific payout. While each game is random, the daily take has to be less than 100% (probably what another poster was mentioning about the 95% rule). In other words, for electronic games, there is such a thing as a 'hot' or 'cold' machine that is 'due' to pay out or not.

Common technique for advanced slots players... observe the players around you. When someone is on a loosing streak and leaves a machine that has wiped them out, that machine is 'hot' and is likely due to payout (it has to, eventually, by law).
That sounds like an example of the gambler's fallacy. That is unless you are saying that the gaming commission requires the casino's to fiddle with the odds of winning by making the slot nonrandom rather than by setting the payout to give the house an edge.

I don't know but seriously doubt that the casinos are required by law to intentionally rig the slots to give nonrandom results.

Just addressed that with a previous post... but yes, it does sound like 'rigging'.. but the law states that the games must be 'rigged' to meet a certain payout. The commission does not care about being fair to the individual player, only to the players as a whole.
 
Really? Do you state that there us a regulation stating that each nachine has to payoff a certain amount per day?

In short, yes. They are programmed to ensure a specific payout over time. The laws of numbers relating to the Gambler's Fallacy apply only in a pure random environment. The Gambler's Fallacy is no fallacy at all with respect to electronic slot machines, due to this law.

Sorry if I'm not satusfied... Isnt it enough that the total payout of all machines together have a specified payout? That how its works in sweden.
 
that's a good question... I don't know if individual machines are audited and that the rule applies on the individual slot, and not the whole... but I imagine it does apply at the individual level... the purpose of the law is to protect the consumer from machines that are programmed to pay out less than the advertised odds (also a law - rules of the game, odds of winning, and exact payout amounts, all must be posted for each machine, and all must be in compliance with the payout law).
 
For electronic games, the payout is regulated by the gaming board. The game programmatically ensures a specific payout. While each game is random, the daily take has to be less than 100% (probably what another poster was mentioning about the 95% rule). In other words, for electronic games, there is such a thing as a 'hot' or 'cold' machine that is 'due' to pay out or not.

Common technique for advanced slots players... observe the players around you. When someone is on a loosing streak and leaves a machine that has wiped them out, that machine is 'hot' and is likely due to payout (it has to, eventually, by law).

Apparently there are teams of slot players who monitor the progressive payout slots and when the jackpot reaches a certain threshold where they deem the jackpot is high enough they will profit, they pile on banks of machines and play like crazy. They even intimidate people to leave machines so they can hog the all the machines and get at the jackpot.
 
that's a good question... I don't know if individual machines are audited and that the rule applies on the individual slot, and not the whole... but I imagine it does apply at the individual level... the purpose of the law is to protect the consumer from machines that are programmed to pay out less than the advertised odds (also a law - rules of the game, odds of winning, and exact payout amounts, all must be posted for each machine, and all must be in compliance with the payout law).
I have no idea what the guidelines of the gaming commission are but I would assume that they only deal with the statistical payout. For instance, someone playing the roulette table has a one in 37 chance of winning for any given spin (assuming a fair table) and the house payout is 35 to one. This gives the house has a statistical edge over time. I would think that what is regulated in this case is the size of the payout for each win (the house wouldn't be allowed to only pay $30 for a $1 bet that wins). I couldn't imagine that the house would be required to "rig" the wheel to give more wins if for a day or week the percentage of winners dropped below the number predicted by probability.

I would think that the same would apply to slots. The house would be required to have the slots as random as possible and to have the payout for a winning combination no less than a given percentage below the probability of that combination occurring.
 
that's a good question... I don't know if individual machines are audited and that the rule applies on the individual slot, and not the whole... but I imagine it does apply at the individual level... the purpose of the law is to protect the consumer from machines that are programmed to pay out less than the advertised odds (also a law - rules of the game, odds of winning, and exact payout amounts, all must be posted for each machine, and all must be in compliance with the payout law).
I have no idea what the guidelines of the gaming commission are but I would assume that they only deal with the statistical payout. For instance, someone playing the roulette table has a one in 37 chance of winning for any given spin (assuming a fair table) and the house payout is 35 to one. This gives the house has a statistical edge over time. I would think that what is regulated in this case is the size of the payout for each win (the house wouldn't be allowed to only pay $30 for a $1 bet that wins). I couldn't imagine that the house would be required to "rig" the wheel to give more wins if for a day or week the percentage of winners dropped below the number predicted by probability.

I would think that the same would apply to slots. The house would be required to have the slots as random as possible and to have the payout for a winning combination no less than a given percentage below the probability of that combination occurring.

this law applies only to electronic games. roulette is pure random chance (in common use of the term - lets not get into the physics of 'does random exist' here), so the Gambler's Fallacy holds... and there is no regulation designed to ensure a minimum payout... the laws of numbers handle that.

With electronic games, the gaming commission has no capacity to audit the source code, hardware, and firmware of those systems to ensure the manufacturer didn't 'pull a VW' and hard code really really really bad odds into the game, and misrepresent the odds of winning, for the purpose of ripping off the very valuable tourists that the state relies upon. With roulette it is simply a matter of looking at the wheel and weighing the ball. With dice, there is a caliper they use to spin the dice and observe if it is unfairly weighted, etc...
 
In short, yes. They are programmed to ensure a specific payout over time. The laws of numbers relating to the Gambler's Fallacy apply only in a pure random environment. The Gambler's Fallacy is no fallacy at all with respect to electronic slot machines, due to this law.

Sorry if I'm not satusfied... Isnt it enough that the total payout of all machines together have a specified payout? That how its works in sweden.

Gambler's fallacy still applies. The machines are required to have average payout set to whatever the local minimum average payout is. This doesn't mean they have to adjust payout if the machine has a hot streak for the house (is below the minimum average).

It means that the odds have to be set in a way that guarantees the average payout is above 75% in a truly random situation.

It does not mean the house has to pay more if a machine is "cold". This doesn't mean that the house never adjusts payouts to above 100% to fish for interest- no law against paying out MORE money. There is just a minimum payout.

1:1 payout 1/20 odds = 5% payout
2:1 payout 1/20 odds = 5% payout
5:1 payout 3/100 odds = 15% payout
1000:1 payout 1/2000 odds = 50% payout

Total payout is set to 75%, which fulfills the requirements. Although you get a 100% payout in blinky lights and noises, so if that's your thing....
 
Apparently there are teams of slot players who monitor the progressive payout slots and when the jackpot reaches a certain threshold where they deem the jackpot is high enough they will profit, they pile on banks of machines and play like crazy. They even intimidate people to leave machines so they can hog the all the machines and get at the jackpot.
A coworker of mine was on one of those teams back in the 90s. (Not that they intimidated people. The way they hogged the machines was that whenever anybody was waiting for a chance at one of their high-jackpot seats, whoever was in the seat would just keep playing and playing, holding it in however long it took until the guy waiting got bored and wandered away, before handing the seat off to a teammate and running for the bathroom.) After a few years of Vegas trips, one of them calculated how fast they were making money off the casinos and figured out it was sub-minimum wage, so the team broke up.
 
I have no idea what the guidelines of the gaming commission are but I would assume that they only deal with the statistical payout. For instance, someone playing the roulette table has a one in 37 chance of winning for any given spin (assuming a fair table) and the house payout is 35 to one. This gives the house has a statistical edge over time. I would think that what is regulated in this case is the size of the payout for each win (the house wouldn't be allowed to only pay $30 for a $1 bet that wins). I couldn't imagine that the house would be required to "rig" the wheel to give more wins if for a day or week the percentage of winners dropped below the number predicted by probability.

I would think that the same would apply to slots. The house would be required to have the slots as random as possible and to have the payout for a winning combination no less than a given percentage below the probability of that combination occurring.

this law applies only to electronic games. roulette is pure random chance (in common use of the term - lets not get into the physics of 'does random exist' here), so the Gambler's Fallacy holds... and there is no regulation designed to ensure a minimum payout... the laws of numbers handle that.

With electronic games, the gaming commission has no capacity to audit the source code, hardware, and firmware of those systems to ensure the manufacturer didn't 'pull a VW' and hard code really really really bad odds into the game, and misrepresent the odds of winning, for the purpose of ripping off the very valuable tourists that the state relies upon. With roulette it is simply a matter of looking at the wheel and weighing the ball. With dice, there is a caliper they use to spin the dice and observe if it is unfairly weighted, etc...
But they can audit the slots if there is a record kept of plays, wins, and payouts. Just check the records over a month or year to see that they are close to those randomness would indicate. To check (and "correct") hourly or daily would mean that they had no understanding of probability. I assume that the regulators have some understanding but then since they are government agents, who the hell knows.
 
Apparently there are teams of slot players who monitor the progressive payout slots and when the jackpot reaches a certain threshold where they deem the jackpot is high enough they will profit, they pile on banks of machines and play like crazy. They even intimidate people to leave machines so they can hog the all the machines and get at the jackpot.
A coworker of mine was on one of those teams back in the 90s. (Not that they intimidated people. The way they hogged the machines was that whenever anybody was waiting for a chance at one of their high-jackpot seats, whoever was in the seat would just keep playing and playing, holding it in however long it took until the guy waiting got bored and wandered away, before handing the seat off to a teammate and running for the bathroom.) After a few years of Vegas trips, one of them calculated how fast they were making money off the casinos and figured out it was sub-minimum wage, so the team broke up.

Interesting. A lot has changed since the 90s, perhaps the jackpots are bigger now and make it more worthwhile.
 
Sorry if I'm not satusfied... Isnt it enough that the total payout of all machines together have a specified payout? That how its works in sweden.

Gambler's fallacy still applies. The machines are required to have average payout set to whatever the local minimum average payout is. This doesn't mean they have to adjust payout if the machine has a hot streak for the house (is below the minimum average).

It means that the odds have to be set in a way that guarantees the average payout is above 75% in a truly random situation.

It does not mean the house has to pay more if a machine is "cold". This doesn't mean that the house never adjusts payouts to above 100% to fish for interest- no law against paying out MORE money. There is just a minimum payout.

1:1 payout 1/20 odds = 5% payout
2:1 payout 1/20 odds = 5% payout
5:1 payout 3/100 odds = 15% payout
1000:1 payout 1/2000 odds = 50% payout

Total payout is set to 75%, which fulfills the requirements. Although you get a 100% payout in blinky lights and noises, so if that's your thing....

I may not be as familiar with the law as you may be... but to take an average you need some kind of sample size. What is the required sample size? Is it a minute to minute average (there are probably thousands of bets per minute), hour to hour, daily... or annually?

heck, I can program a slot machine to NEVER pay a penny out... and when I am audited, I just say the machine is on a 100 year cycle. rest assured... by the year 2117, it will pay.

100% payout in blinky lights, sounds, AND free alcohol to loosen your grip on your wallet... but I look at my gambling losses as a bar bill ;)
 
Why would a gaming company bother to go through the effort to program a slot machine to run in streaks while simultaneously retaining statistical integrity when a random number generator will do exactly the same thing without any of the extra work?

I'm skeptical. Occam's razor and all that.
 
Why would a gaming company bother to go through the effort to program a slot machine to run in streaks while simultaneously retaining statistical integrity when a random number generator will do exactly the same thing without any of the extra work?

I'm skeptical. Occam's razor and all that.

because it is trivially easy to claim that a machine's odds of winning is X, and then program it to operate with a random number generator that has an actual odd of winning X/1000. The audit does not care about 'hot and cold streaks', if the machne fails to make the advertised payout in a certain amount of time, then there penalties associated with that. The question of the day now is, what is the timeframe of the audit and how is the machine programmed / managed to ensure the advertised payout.
 
Why would a gaming company bother to go through the effort to program a slot machine to run in streaks while simultaneously retaining statistical integrity when a random number generator will do exactly the same thing without any of the extra work?

I'm skeptical. Occam's razor and all that.

because it is trivially easy to claim that a machine's odds of winning is X, and then program it to operate with a random number generator that has an actual odd of winning X/1000. The audit does not care about 'hot and cold streaks', if the machne fails to make the advertised payout in a certain amount of time, then there penalties associated with that. The question of the day now is, what is the timeframe of the audit and how is the machine programmed / managed to ensure the advertised payout.

Any reasonable physical audit will necessarily involve enough runs to make the probability of that happening vanishingly small. A code audit is easier if they use a standard RNG. There's no reason for the casino to bother, and gamblers convince themselves that the transient patterns they observe aren't random all the time. This claim looks like more of the same.
 
Why would a gaming company bother to go through the effort to program a slot machine to run in streaks while simultaneously retaining statistical integrity when a random number generator will do exactly the same thing without any of the extra work?

I'm skeptical. Occam's razor and all that.

If they got caught, they'd lose their gaming license (has to be randomly generated every spin!), and a guaranteed income... so it's not logical to do so.

Here's a online slot website's explanation of why you can't win over time, but if you like the game, go ahead and play (they like money too!):

http://www.vegasslotsonline.com/odds/


Progressive slots are another story, but waiting around for a machine (or connected machines, in which case you need a team) to get above the break even point, and then playing straight through until you hit a Jackpot (could be million of plays for some of them!) just seems like a lot of increasing entropy for nothing.

If you have enough money to do that, you have enough to invest in something that gives you a guaranteed return without sitting in front of blinky lights and noises for hours and hours and hours and hours doing nothing that contributes to the advancement of anything or anyone (unless you are a member of the Misanthrope Society, which believes wasting resources is a good way of being deviously evil towards humanity- if that's the case, just vote Republican (currently... parties switch over time), it's far easier than playing slots).
 
Why would a gaming company bother to go through the effort to program a slot machine to run in streaks while simultaneously retaining statistical integrity when a random number generator will do exactly the same thing without any of the extra work?

I'm skeptical. Occam's razor and all that.

because it is trivially easy to claim that a machine's odds of winning is X, and then program it to operate with a random number generator that has an actual odd of winning X/1000. The audit does not care about 'hot and cold streaks', if the machne fails to make the advertised payout in a certain amount of time, then there penalties associated with that. The question of the day now is, what is the timeframe of the audit and how is the machine programmed / managed to ensure the advertised payout.
Essentially, if you are playing a one dollar slot that has the payout at 90% and after 100,000 spins you haven't lost $10,000 +/- $2,000 or so then you should really complain to the management.
 
because it is trivially easy to claim that a machine's odds of winning is X, and then program it to operate with a random number generator that has an actual odd of winning X/1000. The audit does not care about 'hot and cold streaks', if the machne fails to make the advertised payout in a certain amount of time, then there penalties associated with that. The question of the day now is, what is the timeframe of the audit and how is the machine programmed / managed to ensure the advertised payout.
Essentially, if you are playing a one dollar slot that has the payout at 90% and after 100,000 spins you haven't lost $10,000 +/- $2,000 or so then you should really complain to the management.

Just for fun, I decided to work out the probabilities.

From what I've found online, $1 slots have a house advantage of about 10%, with a single wager payout standard deviation of about $5.

That means that, for the 90% payout slots, after 100,000 spins you should expect to lose $10,000 with standard deviation of around $1581.

To be honest, I was a little surprised at how high the standard deviation is, but, on second thought, high volatility makes the games more fun for the player and the casino can afford to wait and let the law of large numbers do its thing.

40 million people visit Vegas each year, and I'd guess that on average it's $100 in slot pulls each? More?

That would mean the casinos should expect to make $400,000,000 in profit on slots in a year with standard deviation of only $300,000. That's crazy.
 
Essentially, if you are playing a one dollar slot that has the payout at 90% and after 100,000 spins you haven't lost $10,000 +/- $2,000 or so then you should really complain to the management.

Just for fun, I decided to work out the probabilities.

From what I've found online, $1 slots have a house advantage of about 10%, with a single wager payout standard deviation of about $5.

That means that, for the 90% payout slots, after 100,000 spins you should expect to lose $10,000 with standard deviation of around $1581.

To be honest, I was a little surprised at how high the standard deviation is, but, on second thought, high volatility makes the games more fun for the player and the casino can afford to wait and let the law of large numbers do its thing.

40 million people visit Vegas each year, and I'd guess that on average it's $100 in slot pulls each? More?

That would mean the casinos should expect to make $400,000,000 in profit on slots in a year with standard deviation of only $300,000. That's crazy.

When the Treasury Casino opened in Brisbane, I went to have a look around with a mate. He said "I have a system for playing roulette; I can't lose". I said "Take a look around. Plush deep carpets; crystal chandeliers; expensive paintings on the walls; Cheap food and drink; loads of staff everywhere; all of this costs a LOT of money. Do you think that this place would look like this, if there was a system where the punter 'can't lose'?".

The only casino that I would patronise as a gambler would be one where they were about to shut down due to bankruptcy. That's at the minimum indicator that the house doesn't have an edge. Even then, I would suspect that their problems didn't arise because the odds were against them.

While casinos are glitzy, it is clear that Mr Burns had the business model all worked out
C. Montgomery "Monty" Burns said:
Well, I've discovered the perfect business: people swarm in, empty their pockets, and scuttle off.
 
Just for fun, I decided to work out the probabilities.

From what I've found online, $1 slots have a house advantage of about 10%, with a single wager payout standard deviation of about $5.

That means that, for the 90% payout slots, after 100,000 spins you should expect to lose $10,000 with standard deviation of around $1581.

To be honest, I was a little surprised at how high the standard deviation is, but, on second thought, high volatility makes the games more fun for the player and the casino can afford to wait and let the law of large numbers do its thing.

40 million people visit Vegas each year, and I'd guess that on average it's $100 in slot pulls each? More?

That would mean the casinos should expect to make $400,000,000 in profit on slots in a year with standard deviation of only $300,000. That's crazy.

The % hold varies from casino to casino, area to area but overall that sounds about right. And Vegas seems to be growing again. Hotels are adding more rooms and expanding.
 
Back
Top Bottom