1. Welcome to the #1 Gambling Community with the best minds across the entire gambling spectrum. REGISTER NOW!
  2. Have a gambling question?

    Post it here and our gambling experts will answer it!
    Dismiss Notice
  3. Discussions in this section are assumed to be EV- as they are outside of the Advantage Play section. For EV+ discussions, please visit the Advantage Play section.
    Dismiss Notice

Roulette Thread for the math "guys" about random.

Discussion in 'Roulette Forum' started by TurboGenius, Jul 19, 2015.

  1. TurboGenius

    TurboGenius Well-Known Member Founding Member

    Joined:
    Jan 25, 2015
    Likes:
    1,794
    Occupation:
    Self proclaimed Theoretical Philosopher
    Location:
    Near Atlantic City New Jersey
    I don't want to deviate from the other thread but here is something nice that 'random'
    produces when it comes to math.
    Roll off 38 spins (on a 38 number wheel of course)

    How many numbers show up 0 times ?
    How many numbers show up 1 time ?
    How many numbers show up 2 times ?
    How many numbers show up 3 times ? etc.
    Go ahead, try it and write down the results and keep that info for a moment.....you'll see why.
    =======================================================================
    Here is one set of 38 spins I just ran off at random.
    55abc648.jpg

    How many numbers appeared 0 times ? 14

    How many numbers appeared 1 time ? 14 (not relevant to this thread)
    - - - - - - - - - - - - - - - - - - - - - - - -
    How many numbers appeared 2 OR MORE times ? 10
    How many numbers appeared 3 OR MORE times ? 4
    4 or more times ? 0
    5 or more times ? 0
    6 or more times ? 0
    Now add these up to get 14 !
    (10+4+0+0+0) This is almost (yes, almost) always the exact same number as
    The amount of number that didn't appear during the 38 spins....

    Here's a big run of hundreds of spins in 38 spin cycles. Remember I used x "or more" times for the
    values.
    55abc8a2.jpg

    So look at the 2x,3x,4x,5x columns and add them up - they almost always equal the same number
    of numbers that don't show during the cycle.

    10+2+1 = 13 (same as the # of 0 shows for that session)
    10+3 = 13 (same)
    12+4 = 16 (same) etc
    The only time it didn't was where a number showed 5x and that might have been a error
    on my part now that I look at it.
    This is 532 spins anyway, but certainly produces the result.

    For the law of third people and non-believers in patterns from random :
    Compare the 0x and 1x columns for basic data on how the amount of numbers that don't show compare to the amount that "do" show (1 or more times) and you'll see the pattern. Or graph it if you like.
    Here's the averages from this set of 532 (not that much changes when you study more cycles of spins)
    Average # of numbers that don't appear : 12.42 out of 38 possible (or basically 1/3 of the 38)
    Average # of numbers that appear at least once : 24.5 / 38 (or basically 2/3 of the 38)
    (You can also 1/3 and 2/3 these results further for numbers that appear 2 or more times, 3 or more times, etc).

    (edit) There is a error in the above chart in the 1x column - far right - the 17 should be 26
    it should appear as : (to be correct)
    1x = 25.25.22.25.25.25.24.23.28.25.23.26.23.24 1 or more shows
    (changes nothing where the 2 or more = 0 explanation, it was my error)
     
  2. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    Turbo,

    Regarding the law of the third, it's basically just binomial distribution. There are more ways for the numbers to be distributed than there are for each number to hit only once. (However, if you were to run enough trials, then you would find that it can happen.) On average, the numbers are distributed as probability expects them to be distributed.

    On a random wheel, the law of the third doesn't provide any information that you can use in order to predict the next number(s). A number doesn't know whether it's "due" to hit or if it should continue sleeping. It's merely an observation of what has hit. It's about as useful as the "law of 1 in 38" at predicting what should hit in the next series of spins.

    The law of the third/binomial distribution could be used to measure the fitness of the wheel in a crude kind of way, but the chi square test is a more efficient test for determining how likely the results were from a random wheel verses a biased wheel.

    -Dr. Sir Anyone Anyone
     
  3. TurboGenius

    TurboGenius Well-Known Member Founding Member

    Joined:
    Jan 25, 2015
    Likes:
    1,794
    Occupation:
    Self proclaimed Theoretical Philosopher
    Location:
    Near Atlantic City New Jersey
    So.. we know that each number won't show in 38 spins, or that a single number won't show 38 times in a row...
    (yes, it's theoretically possible "if you were to run enough trials, then you would find that it can happen" is true) - But we both know that won't happen. If a number were to appear just 6 times in a row the world might end (lol). 38 times in a row is quite impossible - enough so that it doesn't have enough importance to plan for it happening.

    But it does - It's been shown above and in the other thread.
    I did spend about 15 years "hooked on sleepers", I can admit to that - and still have great love for them. I can rattle off a few systems that work amazingly well on them but lately I'm more interested in patterns.
    So instead of looking at spin #2 and saying it has nothing to do with what happened in spin #1 - think of it as a set of 38 spins... That's the "whole". Knowing exactly where the ball lands on the next spin is impossible and would require some kind of superpowers (like that other guy says that I pretend to have because he won't think about what I'm saying one bit).. That's not the case at all.
    I CAN however tell you that when I sit down and begin charting - I see #3 appear.... it's no longer in
    the column of "numbers that won't show"..
    So you can laugh and say "so what, that spin is over" which is entirely true.. but -
    Instead of having a possible 38 numbers to try to work with, now there's one less in the "not going to show"
    column. Now that number is put in the "Has shown 1 or more times". (grins evily).
    And the law of third (or whatever name it has today) will back up that there "are" going to be repeats.
    The #3 that just appeared is now certainly in the potential list of numbers that could repeat (because there are just as many repeats as there are numbers that never show up (as shown above).
    So If I can use past spins to point to #3 now and move it from "not going to show" to "has shown 1 or more times"
    (you see where I'm going don't you lol).
    The house edge is small.
     
  4. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    Turbo,

    After 38 spins, which do you feel is more likely to hit?

    1. A number that has not hit? What do you feel is the new probability of the number hitting after the observation?
    2. A number that has hit once? What do you feel is the new probability of the number hitting after the observation?
    3. A number that has hit twice? What do you feel is the new probability of the number hitting after the observation?
    4. A number that has hit three times? What do you feel is the new probability of the number hitting after the observation?
    5. A number that has hit four times? What do you feel is the new probability of the number hitting after the observation?
    6. A number that has hit five times? What do you feel is the new probability of the number hitting after the observation?
    7. A number that has hit six times? (Yes, I've seen this.) What do you feel is the new probability of the number hitting after the observation?
    8. A number that has hit seven times? (Yes, I've seen this as well.) What do you feel is the new probability of the number hitting after the observation?
     
  5. TurboGenius

    TurboGenius Well-Known Member Founding Member

    Joined:
    Jan 25, 2015
    Likes:
    1,794
    Occupation:
    Self proclaimed Theoretical Philosopher
    Location:
    Near Atlantic City New Jersey
    Any number, regardless of how many times in a row it showed up will have a 1 in 38 chance of showing up in the next spin.
    It will also have a 2 in 3 chance of showing up in the next 38 spins regardless of what it did in the last cycle....
    and if it does show up, it will have a chance of repeating in that cycle of spins, as compared to the numbers that have not shown up in that cycle we are looking at now (after your 38 spins mentioned above).

    About sleepers - After 38 spins, I would be more interested in the 1/3 of the numbers that didn't show up during your set of spins mentioned above.

    About what I'm talking about - if a number shows up 2,3,4,5,6,7 times ? I will have been playing it and would care less about the next session of 38 spins - I would be at the desk changing those chips into a cashiers check.

    In the other thread I said everyone would agree - because we will, in the end.
    Even looking at it from the point of view of AP and not "silly system players", somehow I know we'll be betting on the same numbers anyway and winning together.
     
  6. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    Of the 1/3 of the numbers that remain, what do you believe the probability is that they will hit on the next spin?

    How much do you believe the probability has changed after your observation?
     
  7. TurboGenius

    TurboGenius Well-Known Member Founding Member

    Joined:
    Jan 25, 2015
    Likes:
    1,794
    Occupation:
    Self proclaimed Theoretical Philosopher
    Location:
    Near Atlantic City New Jersey
    Of the 1/3 that remain. I am almost damn near 100% positive with no doubt that after the next cycle of 38 spins - the exact same numbers won't all still be sleepers.

    See you are asking about "the next spin" which can't be predicted. I am talking about a set of spins seen as a group and played as a group - which follow the lovely way random spits out numbers. Those number won't be the same ones after the next 38 spins. I can therefore reason that playing those sleepers gives me an advantage. (and I don't need much advantage over an opponent who's edge over me is only 5.25%.

    Is it fallacy to play sleepers like that ? So they say it is....
    What are the odds of the exact same numbers being the exact same sleepers after the next 38 spin cycle is complete ?
    Grins.
    At some point you see there is low enough risk of something extremely rare happening, and you take advantage of it - it's not fallacy - just smart bet selection.
    If it were true that lightning randomly hits the ground, and "never strikes the same place twice" - I would be sure to stand where it had already hit. (and a 'probability guy' would surely tell me... "you know there's a 1 in 434 million chance you'll get hit standing there" and he'd probably be right lol)
     

  8. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    You're saying that the 1/3 of the numbers that remain have a better chance of hitting on the very next spin and the next series of spins. So the question remains, how much do you believe the probability of each of these numbers has changed? In your opinion, what is the expected hit frequency of the unhit numbers? 1/38? 1/37? 1/36? 1/35? Stronger?
     
  9. albalaha

    albalaha Active Member Founding Member

    Joined:
    Dec 29, 2014
    Likes:
    122
    Occupation:
    player
    Location:
    India
    Can we play 14 sleeper numbers to get at least 1 or 2 hit in next 38 spins, that can fetch some earning for us? Undoubtedly, they have much higher probability to get 1 or more hits than none but how will it help in playing with real casino conditions(table limits and bankroll)?
    I have simulated this, way back. It looks reasonable but not good to win, by itself. It is then, only as good, as playing anything randomly.
     
  10. Rona

    Rona Active Member

    Joined:
    Jul 8, 2015
    Likes:
    91
    Of the 12 numbers that remain, 1/3, that is 4, will remain un-hit after the next 38 spins.
    After the next 38 spins 1/3 of the 4, that is 1+ will remain un-hit.
     
  11. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    There's no benefit to betting the numbers that have not hit (cold numbers). Furthermore, it's a really bad idea to play such numbers on the live wheel. In the event that the wheel is biased, chasing the cold numbers is one way by which you could actually lose at a rate that could theoretically exceed the house edge. In other words, some of the coldest numbers may not be hitting because they could be negatively biased (defective). After having tested millions of spins on live wheels, I've found that some numbers can drop well below the long term expectation.

    Furthermore, betting on the hottest numbers could be beneficial for the same reason. There's a better chance that one or some of numbers that are hitting the most, could be performing better than expectation due to bias.

    In the long run, on the live wheel, it's always better to be on just a few of the hottest numbers, and to avoid the coldest numbers all together, in the off chance that the wheel is significantly biased.

    -Dr. Sir Anyone Anyone
     
  12. TurboGenius

    TurboGenius Well-Known Member Founding Member

    Joined:
    Jan 25, 2015
    Likes:
    1,794
    Occupation:
    Self proclaimed Theoretical Philosopher
    Location:
    Near Atlantic City New Jersey
    That would be a rare event - to happen to be at a table with a bias wheel.
    Aside from that, I agree that playing numbers that have been showing makes more sense than the cold or sleepers.
    However - If I have 13(ish) numbers that did not show in 38 spins, I know that one of them are going to show (at least) in the next cycle. Which means trying to choose from 38 numbers is now 13(ish) numbers. It's simple.
     
    GaryG likes this.
  13. Dr. Sir Anyone Anyone

    Dr. Sir Anyone Anyone Well-Known Member Lineage to Founders

    Joined:
    Feb 18, 2015
    Likes:
    931
    Occupation:
    Shoe Cobbler
    Location:
    Merica
    Every wheel is biased to some small degree.
     

Share This Page