Slots, random or.............
-
- VP Veteran
- Posts: 670
- Joined: Fri Feb 17, 2012 10:13 pm
Slots, random or.............
I'm not much of a slot player anymore, never really understood how a slot machine worked. My question is are slot machines random or rigged?
-
- Video Poker Master
- Posts: 3298
- Joined: Tue Jul 26, 2011 12:49 am
Most laws say it's random. Paybacks are tight enough to ensure the house wins big regardless of individual results where there is no real good reason to have a non-random result. Unlike VP, the entire math model to set the payback is completely in the hands of the manufacturer. High variance, low variance, 85% payback, 90% payback...all can be widely adjusted.
Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated.
Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known.
Deconstructing Jackpot Party
Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated.
Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known.
Deconstructing Jackpot Party
-
- VP Veteran
- Posts: 670
- Joined: Fri Feb 17, 2012 10:13 pm
Most laws say it's random. Paybacks are tight enough to ensure the house wins big regardless of individual results where there is no real good reason to have a non-random result. Unlike VP, the entire math model to set the payback is completely in the hands of the manufacturer. High variance, low variance, 85% payback, 90% payback...all can be widely adjusted.
Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated.
Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known.
Deconstructing Jackpot Party
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.
Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated.
Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known.
Deconstructing Jackpot Party
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.
-
- Forum Newbie
- Posts: 5
- Joined: Sun Aug 23, 2015 5:40 pm
Most if not all the slots have multiple pay tables that are selected either by (in really old machines) changing out a memory chip or (as in all modern machines) downloading a new pay table over the network. In the latter case state laws decree that the pay table can only be changed if the machine is idle (no money in it).However what confuses people is that, in actuality, slots are a double lookup. When you hit the button the RNG picks a number and that indexes into a table of payouts to determine what you will win (or not). It is the population of this table that defines what the return of the machine is. Then the RNG picks another number and that's used to determine which of 'n' combinations that have been set to pay that amount will be displayed. The "show" has absolutely nothing to do with the return.Variance is not the return or EV. Variance is the relative ratio of how big the rare payout is. An easy way to think of it is to picture a game where you bet $1 to throw two dice and what you get back is based upon the number you throw. Say that there are three tables that offer this, but each table has a different payout.Table 1: lose on 2,3,11,12; even money 4,5,6,8,9,10; 2:1 for 7Table 2: 4:1 payout for 2, 3:1 on 11, 2:1 for 4 or 10, even money for 5,7,9; lose on 3,6,8,12.Table 3: 18:1 payout for 2 or 12, lose otherwise.Each game has a 100% expected return, but the variance (square of the standard deviation) is quite different. For table 1 the variance is 0.418, for Table 2 the variance is 1.818 and for Table 3 the variance is 53.018. 9/6 Jacks has a variance of about 19; DDB a variance of about 42. Slots can have variance numbers in the hundreds.
-
- VP Veteran
- Posts: 670
- Joined: Fri Feb 17, 2012 10:13 pm
Most if not all the slots have multiple pay tables that are selected either by (in really old machines) changing out a memory chip or (as in all modern machines) downloading a new pay table over the network. In the latter case state laws decree that the pay table can only be changed if the machine is idle (no money in it).
However what confuses people is that, in actuality, slots are a double lookup. When you hit the button the RNG picks a number and that indexes into a table of payouts to determine what you will win (or not). It is the population of this table that defines what the return of the machine is. Then the RNG picks another number and that's used to determine which of 'n' combinations that have been set to pay that amount will be displayed. The "show" has absolutely nothing to do with the return.
Variance is not the return or EV. Variance is the relative ratio of how big the rare payout is. An easy way to think of it is to picture a game where you bet $1 to throw two dice and what you get back is based upon the number you throw. Say that there are three tables that offer this, but each table has a different payout.
Table 1: lose on 2,3,11,12; even money 4,5,6,8,9,10; 2:1 for 7
Table 2: 4:1 payout for 2, 3:1 on 11, 2:1 for 4 or 10, even money for 5,7,9; lose on 3,6,8,12.
Table 3: 18:1 payout for 2 or 12, lose otherwise.
Each game has a 100% expected return, but the variance (square of the standard deviation) is quite different. For table 1 the variance is 0.418, for Table 2 the variance is 1.818 and for Table 3 the variance is 53.018.
9/6 Jacks has a variance of about 19; DDB a variance of about 42. Slots can have variance numbers in the hundreds.
I used to play a slots, didn't know the first thing about how they worked. I used to enjoy playing them but when things in this area got tight the slots got real mean. At one time there was a casino in this area that required all games be considered games of skill. When you played slots there you would push the button then select which reels you wanted to hold and spin again, the games were basicly the same as at any other casino and the results were no different as far as I could tell.
However what confuses people is that, in actuality, slots are a double lookup. When you hit the button the RNG picks a number and that indexes into a table of payouts to determine what you will win (or not). It is the population of this table that defines what the return of the machine is. Then the RNG picks another number and that's used to determine which of 'n' combinations that have been set to pay that amount will be displayed. The "show" has absolutely nothing to do with the return.
Variance is not the return or EV. Variance is the relative ratio of how big the rare payout is. An easy way to think of it is to picture a game where you bet $1 to throw two dice and what you get back is based upon the number you throw. Say that there are three tables that offer this, but each table has a different payout.
Table 1: lose on 2,3,11,12; even money 4,5,6,8,9,10; 2:1 for 7
Table 2: 4:1 payout for 2, 3:1 on 11, 2:1 for 4 or 10, even money for 5,7,9; lose on 3,6,8,12.
Table 3: 18:1 payout for 2 or 12, lose otherwise.
Each game has a 100% expected return, but the variance (square of the standard deviation) is quite different. For table 1 the variance is 0.418, for Table 2 the variance is 1.818 and for Table 3 the variance is 53.018.
9/6 Jacks has a variance of about 19; DDB a variance of about 42. Slots can have variance numbers in the hundreds.
I used to play a slots, didn't know the first thing about how they worked. I used to enjoy playing them but when things in this area got tight the slots got real mean. At one time there was a casino in this area that required all games be considered games of skill. When you played slots there you would push the button then select which reels you wanted to hold and spin again, the games were basicly the same as at any other casino and the results were no different as far as I could tell.
-
- Video Poker Master
- Posts: 3298
- Joined: Tue Jul 26, 2011 12:49 am
[QUOTE=Vman96]Most laws say it's random. Paybacks are tight enough to ensure the house wins big regardless of individual results where there is no real good reason to have a non-random result. Unlike VP, the entire math model to set the payback is completely in the hands of the manufacturer. High variance, low variance, 85% payback, 90% payback...all can be widely adjusted. Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated. Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known. Deconstructing Jackpot Party
Â
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.[/QUOTE]
They wouldn't have to necessarily raise the variance to decrease the return, for example, you can lower the frequency of the higher paybacks getting hit, then it will both lower return and variance. However, if I ran a casino, I would try to avoid offering a low payback game with low variance (like 6/5 JoB in the VP world). It will be likely be boring while people lose heavily. An exception to this is if the game is really flashy and interactive while playing (e.g. Sinbad slot machines).
Â
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.[/QUOTE]
They wouldn't have to necessarily raise the variance to decrease the return, for example, you can lower the frequency of the higher paybacks getting hit, then it will both lower return and variance. However, if I ran a casino, I would try to avoid offering a low payback game with low variance (like 6/5 JoB in the VP world). It will be likely be boring while people lose heavily. An exception to this is if the game is really flashy and interactive while playing (e.g. Sinbad slot machines).
-
- VP Veteran
- Posts: 670
- Joined: Fri Feb 17, 2012 10:13 pm
[QUOTE=ko king] [QUOTE=Vman96]Most laws say it's random. Paybacks are tight enough to ensure the house wins big regardless of individual results where there is no real good reason to have a non-random result. Unlike VP, the entire math model to set the payback is completely in the hands of the manufacturer. High variance, low variance, 85% payback, 90% payback...all can be widely adjusted. Generally slots get numbers from an RNG, and each number is mapped to a reel symbol, and the result is evaluated. Here is a decent example of a older video slot and how one can calculate the payback if the reel symbol pattern and bonus games rules are completely known. Deconstructing Jackpot Party
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.[/QUOTE]
They wouldn't have to necessarily raise the variance to decrease the return, for example, you can lower the frequency of the higher paybacks getting hit, then it will both lower return and variance. However, if I ran a casino, I would try to avoid offering a low payback game with low variance (like 6/5 JoB in the VP world). It will be likely be boring while people lose heavily. An exception to this is if the game is really flashy and interactive while playing (e.g. Sinbad slot machines).[/QUOTE]
Well I guess that's where's my confusion starts and ends, "They wouldn't have to necessarily raise the variance". The possibilty that they could simply raise the variance on slots and still remain within the regulations, slots by law must be random, they're run by an RNG, video poker must be random by law and it is also ran by an RNG, whether you do good or bad playing depends on the level of variance you experience. So what separates slots from video poker when it comes to variance levels and the way they are delivered to players as far as the regulations go?
So slots by law must be random, the only difference is the level of variance, if a casino chooses to raise their take on a slot machines all they have to do is raise the level of variance, right or wrong? The slot reels don't have to change, nor the pay for say double or triple bars, in oher words the slot would look exactly the same way.[/QUOTE]
They wouldn't have to necessarily raise the variance to decrease the return, for example, you can lower the frequency of the higher paybacks getting hit, then it will both lower return and variance. However, if I ran a casino, I would try to avoid offering a low payback game with low variance (like 6/5 JoB in the VP world). It will be likely be boring while people lose heavily. An exception to this is if the game is really flashy and interactive while playing (e.g. Sinbad slot machines).[/QUOTE]
Well I guess that's where's my confusion starts and ends, "They wouldn't have to necessarily raise the variance". The possibilty that they could simply raise the variance on slots and still remain within the regulations, slots by law must be random, they're run by an RNG, video poker must be random by law and it is also ran by an RNG, whether you do good or bad playing depends on the level of variance you experience. So what separates slots from video poker when it comes to variance levels and the way they are delivered to players as far as the regulations go?
-
- Video Poker Master
- Posts: 3298
- Joined: Tue Jul 26, 2011 12:49 am
Variance has nothing to do with how good or bad you do longterm. It affects on how your how extreme your session result is.
Lets say a casino change machine is a game. The return of it is 100.00% with the variance of zero.
Let's also say that 9/6 JoB with a royal of 4880 coins is also a game available in the casino. This also has a return of nearly 100%, technically 100.00001895% with a variance of 30.16. But for this argument, let's say it's exactly 100% return.
And lets say you play each "machine" with $50 for $5 a try 20 times each or until you go broke. Barring a malfunction, the change machine will always give you your $50 back. Meaning you win nothing long term.
The VP game however, will sometime take your $50, sometimes let you win near $4880 with a royal, and many other results in between for your 20 hand (or less) sessions. But after millions of sessions of this game, the amount you've won overall should approach zero just like the change machine. But the higher the variance is, the more your results can deviate from the expected long-term result.
Legally what sets VP and slot machines apart in most states is that the probability of the cards in a video poker game must behave like a random, standard poker deck. So variance and return may only be changed by changing the paytable values.
Let's look at two examples, 9/6 Jacks or Better and 9/6 Bonus Deluxe. The two games have nearly the same return, 99.54% vs. 99.64%, respectively, but they have two significantly different variances, 19.51 vs 32.13, respectively.
Why are the variances so different? JoB pays 2 for 1 with two pair, but only 25 for 1 on quads while Bonus Deluxe pays only 1 for 1 on two pair, but 80 for 1 on quads.
Now with slots, there is no law mandating the probability of slot symbols appearing because the slot manufacturer may create as many symbols on a video slot machine as they want. The only thing the law asks is the probability of each symbol being mapped to an RNG number is constant. So if there is 500 reel positions for reel 1, then each position should have a 1 in 500 chance of appearing.
So now slot manufacturers can manipulate return and variance by both changing payouts AND changing symbol probabilities by adding/removing symbols.
Lets say a casino change machine is a game. The return of it is 100.00% with the variance of zero.
Let's also say that 9/6 JoB with a royal of 4880 coins is also a game available in the casino. This also has a return of nearly 100%, technically 100.00001895% with a variance of 30.16. But for this argument, let's say it's exactly 100% return.
And lets say you play each "machine" with $50 for $5 a try 20 times each or until you go broke. Barring a malfunction, the change machine will always give you your $50 back. Meaning you win nothing long term.
The VP game however, will sometime take your $50, sometimes let you win near $4880 with a royal, and many other results in between for your 20 hand (or less) sessions. But after millions of sessions of this game, the amount you've won overall should approach zero just like the change machine. But the higher the variance is, the more your results can deviate from the expected long-term result.
Legally what sets VP and slot machines apart in most states is that the probability of the cards in a video poker game must behave like a random, standard poker deck. So variance and return may only be changed by changing the paytable values.
Let's look at two examples, 9/6 Jacks or Better and 9/6 Bonus Deluxe. The two games have nearly the same return, 99.54% vs. 99.64%, respectively, but they have two significantly different variances, 19.51 vs 32.13, respectively.
Why are the variances so different? JoB pays 2 for 1 with two pair, but only 25 for 1 on quads while Bonus Deluxe pays only 1 for 1 on two pair, but 80 for 1 on quads.
Now with slots, there is no law mandating the probability of slot symbols appearing because the slot manufacturer may create as many symbols on a video slot machine as they want. The only thing the law asks is the probability of each symbol being mapped to an RNG number is constant. So if there is 500 reel positions for reel 1, then each position should have a 1 in 500 chance of appearing.
So now slot manufacturers can manipulate return and variance by both changing payouts AND changing symbol probabilities by adding/removing symbols.