1. Welcome to the #1 Gambling Community with the best minds across the entire gambling spectrum. REGISTER NOW!
  2. Join our $5,000 Cash Giveaway!

    Win Cash by Posting and Inviting New Members!
  3. Have a gambling question?

    Post it here and our gambling experts will answer it!
    Dismiss Notice
  4. 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 sub-forum.
    Dismiss Notice

Roulette Using sets to Create an HG

Discussion in 'Roulette Forum' started by delectus, Feb 5, 2019.

  1. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Jerome, briefly I mentioned previously in the thread "Want an HG? Here's One"
    that using 90 sets would be ideal in creating an HG. At this stage I am not
    suggesting it's an absolute certainty, but if carefully worked out it could well
    continue for thousands of continuous spins and show a profit at the end of
    an agreed target.

    You mentioned the possibility of 40,000 spins, that's for 4 numbers flat bets.
    It's difficult because as I said the bets are in two parts. When two or more
    sets are equal then flat betting will take place at 10c for each of the 4
    numbers in a set, a 40c bet every spin. As soon as a set takes the lead
    from the other 89 sets then, the bets change to $1 bets. There will be a
    progression applied to each of the four numbers, but I have yet to work
    it all out, that is one of the difficult parts of this project.

    I am slowly working through a thousand spins, which will hopefully
    resolve any problems, especially the betting process. I very much like
    the idea the $1 and 40c process, which I think will be a big advantage.

    The large number of sets should also be advantageous, as 2 or more sets
    should always be close to the leading set and therefore close to taking
    the lead when the leading run ends.

    I will provide more details having worked through the 1000 spins
     
  2. Jerome

    Jerome Active Member

    Joined:
    Mar 15, 2018
    Likes:
    83
    Occupation:
    Self-Employed
    Location:
    England
    delectus, in the other thread I wrote :
    This isn't quite right as I discovered when I wrote a simulation. It's not that the theory is wrong but I made a mistake with the degree of certainty, which is not 99% but more like 96%. This means that the threshold should be higher than 40,000 bets if you want to be more certain that your result couldn't have occurred by chance. Anyway I've attached the simulation program here so you can play around with it. It's not restricted to 4 numbers (you can input any number of numbers). It runs from a windows command prompt, just input the number of numbers and the session length, press ENTER and it will simulate 100 sessions each of the specified length, betting numbers picked randomly. The output lists the final bank for each of the 100 sessions and tells you how many of them ended in profit (marked with a "P"). eg -

    [​IMG]

    [​IMG]

    You can do the same thing multiple times without having to re-run the program. Try increasing the length of a session beyond 40,000 until there are no sessions in profit; this will be the threshold beyond which it's almost impossible that a bet selection without an edge will win.

    Incidentally, running the program a few times using different amounts of numbers and session lengths will demonstrate that for a random selection (or one without an edge), betting few numbers will always give you a much higher percentage of winning sessions compared to betting many numbers (theory tells you this will happen but it's nice to see an actual demo of it). For example if you're betting 30 numbers with a session length of only 5000, there will likely be NO sessions in profit, but betting 2 numbers over a much longer session (say 20,000) will result in about 15 profitable sessions.

    Something else you can learn from the program is how easy it can be to jump to the conclusion that because you've tested your system over many spins and it shows a profit, it must be a winner. The simulation shows that even with a completely random bet selection (and not even a progression), a significant number of sessions will result in profit, especially if you're only betting on a few numbers.

    Regarding your system based on sets, probably better if I don't comment until I know more about how it works. ;)
     

    Attached Files:

    Last edited: Feb 6, 2019
  3. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Jerome, thanks very much for the simulation program, will get round to it soon.

    My early assumption of what's required for this project were only partially
    correct. I assumed incorrectly, that a leading set would appear soon after
    a previous hit! The fact is that it could take 200 to 250 spins between hits.
    Ordinarily this would be disastrous, but the 10c bets reduces substantially
    the need to apply the $1 bets.

    At present I am using the same progression I use for the ' A Date with
    Probability' thread. At one point there was a step 96, that's a $8 bet,
    profit $7. The largest so far steps 119, that's a $16 bet, profit $22. As
    I write the progression shows steps 82:88:02:01 for the four numbers
    of a set. I am at spin 329, so just about a third of the way to 1000
    spins.,

    The 90 sets produces what I call 'clusters' (borrowed from eggs :))
    of hits, example set 06:07:26:32 = 07,26,32,06,06,07,06. If this is
    a theme of the 90 sets, that would be great.

    I am also considering using just one number for the 10c bet and will
    work out how the number will be chosen. As you have shown it is
    possible to make winning sessions with just one number.

    Most probably I will have to work through the 1000 spins 2 or 3 times,
    to be absolutely certain, that I can't make any further improvements.

    The biggest concern is the progression, perhaps I will think of
    something as I work through the spins.
     
  4. Sharptracker

    Sharptracker Active Member

    Joined:
    Apr 12, 2018
    Likes:
    139
    Location:
    Belgium
    @Jerome

    As i told you i don't have your skills about Mathematics but i had asked a similar test from a german forum, long time ago, where some people had real skills also in mathematics and the fork i had differ from yours...

    Actually it's a bit like i ask what is the extreme right position possible limit of the Gauss curve and it concerns only spins played of course (no observed spins), in order to claim that we got something with "x" system. (i don't have the numbers in my head, i just remember)

    6 numbers = 60k spins flat bet (i have a doubt about that one, but i'm pretty sure that i remember correctly about the two others)
    3 numbers = 120K spins flat bet
    1 number = 250K spins flat bet

    Could it possible that it differs from the fact that you play all consecutives spins randomly & with nothing built to play a certain % of spins?
     
    Last edited: Feb 7, 2019
  5. Jerome

    Jerome Active Member

    Joined:
    Mar 15, 2018
    Likes:
    83
    Occupation:
    Self-Employed
    Location:
    England
    Sharp, if you're betting randomly skipping spins shouldn't make any difference, and I would argue that a specific bet selection makes no difference either (others here would disagree though. :D).

    Regarding the limits you posted, I guess it depends on what your criteria is. The standard is usually 5% but 1% gives you more certainty, meaning that on average 99 sessions out of 100 will finish losers. I ran the simulation with your numbers and 60k seems too high for 6 numbers (should be more like 40,000), but the others seem about right (assuming 99%). Download the program for yourself and give it a try. ;)
     
    Last edited: Feb 7, 2019
  6. Sharptracker

    Sharptracker Active Member

    Joined:
    Apr 12, 2018
    Likes:
    139
    Location:
    Belgium
    Hopefully :), i just wanted to make it sure that my memory failed to remember the number for 6 spins.

    Thanks for the program, this is beast knowledge you got imo!... I'm curious about progressions, how can you determine it? do you have to build it progression by progressions? I'm curious also about the numbers of spins!
     
  7. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    I'm at spin 375 and the step progression for the four set numbers stands
    at steps 10:01:03:35. Number 11 hit at step 106, $11 bet, profit $17.
    Number 19 hit at step 122, $17 bet, profit $8.

    Previously I mentioned about clusters of hits (got the eggs wrong last
    time that's clutches not clusters oh the English language :confused:) and the
    last two sets produced 29,29,32,29 and 11,26,19. The last set
    11,19,26,36 is currently 54 numbers long. 12,14,27,34 is only 28
    numbers long. Unlikely to become a leading set any time soon.

    I'm hoping that the above hits will be a repeat cycle maximum of
    300 to 400 spins.

    I will work at the 10c spin over the weekend.
     

  8. Jerome

    Jerome Active Member

    Joined:
    Mar 15, 2018
    Likes:
    83
    Occupation:
    Self-Employed
    Location:
    England
    Not sure what you mean? There's no progression used in the sim; flat bets only. Actually it would be interesting to put one in and see how much difference it makes to the number of winning sessions.
     
  9. Sharptracker

    Sharptracker Active Member

    Joined:
    Apr 12, 2018
    Likes:
    139
    Location:
    Belgium
    Yeah, i have seen that you were running it flat bet. But i was wondering if some progressions could perform better than others and could put the limit further or if them are also all equal. Also it would be very interesting to see how long you could resist also than if you were flatbetting.

    Thanks!
     
  10. Sharptracker

    Sharptracker Active Member

    Joined:
    Apr 12, 2018
    Likes:
    139
    Location:
    Belgium
    @Jerome

    Up, in case you haven't seen it. (If you need me to reformulate, don't hesitate) I really would like to know (curious) about that.
     
  11. Jerome

    Jerome Active Member

    Joined:
    Mar 15, 2018
    Likes:
    83
    Occupation:
    Self-Employed
    Location:
    England
    I'm working on it and will post in another thread so as not to derail this one.
     
  12. Sharptracker

    Sharptracker Active Member

    Joined:
    Apr 12, 2018
    Likes:
    139
    Location:
    Belgium
    Great! Thanks Jerome
     
  13. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Progress on this project is slow but sure. I got to nearly 400 spins
    and decide the way I presented it on my work sheets was
    unsatisfactory so back to spin 1. The good thing is I am able to
    correct mistakes, which is easy to do, having so many rows of
    numbers.

    I have decided on the way to place a 10c bet, when 2,3 or 4 sets
    are equal and will explain at spin 1000 to make sure it works
    satisfactorily.

    I have never created such long sets before, so I am finding it
    very interesting.
     
  14. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Jerome, I haven't posted here for a while, not because I have given up on the contrary
    I am very enthusiastic about it. It is a far bigger project than I first imagined.
    What else would you expect from an HG lol. Certainly a great deal of work
    with 90 rows of numbers and errors never far away. I have however created
    checks that should eliminate hopefully all errors.

    I have gone back 3 times to the beginning, having reached 3 to 400
    spins each time, hence the slow progress. I am content now that it should work
    well. The obvious question is why should it work and reach 40,000 + spins and
    show a healthy profit! All your theories will demand a negative result. As far as
    I am aware no one has produced a project using 90 sets with a number of
    advantages, so it will be interesting to see how it progresses.

    I will show next post a sample of spins. I keep getting logged out after a while
    so annoying.
     
    Last edited: Mar 2, 2019

  15. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Here are the first 58 spins from my workbook, Bold numbers are $ bets

    20
    03
    02
    23
    36
    14
    17
    32
    35
    0
    17
    10
    33
    37
    05
    29
    11
    13
    11
    <......hit
    18
    35
    31
    20

    19
    07
    16
    10
    19
    14
    07
    23
    25
    24
    26
    17
    30
    08
    10
    32

    12
    01
    06
    23
    32
    30
    06 <.....hit
    28
    29
    11
    35
    12
    26
    35
    26 <....

    26 <....hit
    18
    20
    04
     
  16. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    At spin 500 I will provide more details and show the current profit.

    In order to progress with the 10 cent spins a digit count takes place.
    If we start with 20 (the 1st spin) 2+0 =2 therefore looking for number
    2 to repeat in the next 2 spins. 02 produces another 2 spins at 2.
    Now 36 therefore 3 + 6 = 9, so looking for 36 to appear during the
    next 9 spins and so on. Unfortunately the 26 was $ bet so no win,
    but was still part of the count and the next 26 was a 10c win
     
  17. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Jerome, I have now reached 500 spins and very pleased with the results. The 90 sets
    have produced leading sets, with a cluster of hits, many with repeat numbers. In
    addition 2 or 3 sets are close to a leading set in order to take over after a certain
    number of spins. If the leading sets are equal at any time, then the spin at 10c kicks
    in until there is again a clear leading set. I previously explained (see post 16) how
    the 10c spin operates and it should almost pay for itself, but is not expected to show
    a profit. It will however keep the progressions lower than would normally be possible
    when using the $ bets. So far there are 166 10c spins at a loss of $5.80 and 3 hits.

    The progression for the $ bets is very long at 196 steps, with a potential loss of
    $4987. This is the worse case scenario and as this project has a number of
    advantages, I don't expect to be close to the 196 steps. There may be a possibility
    of introducing a safety net, but that will be discussed much later.
     
    Last edited: Mar 11, 2019
  18. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Every roulette project must have a tricky problem and this project is no exception.
    I have recorded the steps of the progressions in two different ways as follows:

    Usual continuation

    11 19 26 36 set
    09 32 42 30
    11 13 27 29 set
    10 33 43 31
    11 34 44 32
    12 35 45 33 hit
    13 36 46 01

    Switch

    11 19 26 36 set
    84 35 30 09
    11 13 27 29 set
    85 36 31 10
    86 37 32 11
    87 38 33 12 hit
    01 34 39 88


    The 'switch' will be wanting repeat numbers

    With the 'normal continuation' the progression reached 158 steps
    and with the 'switch' 122 steps. At this stage I favour the switch
    method to take advantage of a cluster of hits produced by a
    leading set. Ideally I would not want the steps of the progression
    to go above 130 steps, with a maximum of 140 steps.
     
    Last edited: Mar 11, 2019
  19. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    At spin 500 the profit is as follows:

    Usual continuation

    The first column is the steps 2nd $ profit 3rd accumulated profit

    002 34
    012 24 058
    027 09 067
    035 01 068
    029 07 075
    042 23 098
    012 24 122
    047 13 135
    070 17 152
    050 07 159
    119 22 181
    007 29 210
    025 11 221
    019 17 238
    023 13 251
    011 25 276
    106 17 293
    158 31 324
    004 32 356
    029 07 363
    010 26 389
    038 31 420
    005 31 451
    039 29 480
    043 21 501
    091 10 511
    012 24 535
    005 31 566
    042 23 589
    033 03 592
    052 03 595
    046 15 610
    003 33 643
    015 21 664
    002 34 698
     
  20. delectus

    delectus Member

    Joined:
    Jul 2, 2018
    Likes:
    18
    Location:
    uk
    Switch

    The first column is the steps 2nd $ profit 3rd accumulated profit

    002 34
    012 24 058
    027 09 067
    035 01 068
    012 24 092
    030 06 098
    054 34 132
    048 11 143
    045 17 160
    060 16 176
    112 16 192
    122 08 200
    025 11 211
    093 31 242
    023 13 255
    041 25 280
    010 26 306
    029 07 313
    052 03 316
    008 28 344
    014 22 366
    048 11 377
    074 01 378
    039 29 407
    014 22 429
    009 27 456
    019 17 475
    002 34 507
    033 03 510
    012 24 534
    045 17 551
    011 25 576
    100 07 583
    001 35 618
    049 09 627
     

Share This Page