### Author Topic: NEW SYSTEM: SINGLE DOZEN  (Read 31605 times)

0 Members and 4 Guests are viewing this topic.

#### Reyth

• Global Moderator
• Hero Member
• Posts: 3779
• Thanked: 1197 times
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #420 on: April 16, 2017, 08:52:46 AM »
Most successful roulette players I think bet 0 from time to time!

The other one (cough, cough) I think is doing something crazy like some form of "min bet". O_o

#### kav

• www.Roulette30.com
• Hero Member
• Posts: 1908
• Thanked: 885 times
• Gender:
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #421 on: April 16, 2017, 10:10:33 AM »
Kav, less than 1 win in a cycle is the condition for betting 0. A cycle for the dozen bet being 3 spins, for DS it's 6 spins, etc.

Bayes I believe this a great idea.
In fact I believe it is a great type of trigger build in a progression. It is something I would certainly like to try when betting dozens continuously.
But I don't think it works well in combination with an independent trigger like the one of Palestis. It is like the one trigger tells you "bet now" and you saying "no I wont, because I lost too many attacks recently"

Maybe we should start a "Bayes' dozens" Topic.
« Last Edit: April 16, 2017, 10:13:02 AM by kav »

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #422 on: April 16, 2017, 11:17:52 AM »
Kav, it's interesting to calculate the chances of getting a "whipsaw" when skipping bets in this way. This is the curse of any attempt to use "virtual wins" as a trigger. For a dozen, we know that the chance of at least 1 hit in 3 spins is 69.1% (single zero). Suppose we follow the rule of no bet when you don't get at least one hit in the last 3 bets (a cycle).

The following could happen, in which case it's equivalent (in terms of loss) to getting 12 losses (individual single bets) in a row. Assuming flat bets to keep things simple, we get:

W Trigger
L  -3 u
W Trigger
L -6 u
W Trigger
L -9
W Trigger
L -12

Now let's calculate the probablity of this sequence, given the probability of a win = 69.1% and probability of loss = 100 - 69.1% = 30.9%

There were 4 losses and 4 wins so the probability is (0.691)4 x (0.309)4 = 0.21%

Now let's calculate the probability of 12 straight losses in a row for a dozen bet:

Probability = (1-12/37)12 = 0.91%

So although both probabilities are small, and both results are financially speaking the same (and 12 bets were made in both cases), the 12 straight losses are 0.91/0.21 = 4.33 times more likely than the same number of losses when following the "only bet after a win" rule. Nice!
« Last Edit: April 16, 2017, 11:37:40 AM by Bayes »

The following users thanked this post: Reyth, TERMINATOR

#### Reyth

• Global Moderator
• Hero Member
• Posts: 3779
• Thanked: 1197 times
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #423 on: April 16, 2017, 12:28:36 PM »
Holy crap!

The following users thanked this post: Bayes

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #424 on: April 16, 2017, 01:26:32 PM »
Reyth, that was my reaction too. I kept thinking I'd made a mistake so I checked and double-checked, but the calculation is good as far as I can see.

Most system players who play around with bet selections would probably feel intuitively (there's that word again!) that there is some value to them - some more than others. But it's nice to see a mathematical confirmation; it means they can't all be dismissed as "wishful thinking", or an illusion caused by the fact that it takes longer to lose merely because you're waiting for triggers.

The following users thanked this post: Reyth

#### kav

• www.Roulette30.com
• Hero Member
• Posts: 1908
• Thanked: 885 times
• Gender:
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #425 on: April 16, 2017, 01:48:42 PM »
Bayes,

I don't understand what your first equation calculates. IMO we don't care about the probability of the win, we just wait for it to happen. When, how soon or late, it happens is not important. There is no point to take into account the probability of win. The trigger is a point of reference, its probability doesn't affect the other probabilities. If the trigger was snow weather, would you take into account the probability of snowing?
« Last Edit: April 16, 2017, 01:51:25 PM by kav »

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #426 on: April 16, 2017, 02:48:57 PM »
Kav, the point was to show that 12 losses in a row can happen when you're using the rule about only betting when the last 3 spins show at least one win, and that can only happen when the particular sequence wLwL... etc. occurs. Note that if there are multiple losses between the wins such wLLwLLLw etc, it doesn't matter, because you won't be betting after the initial L.

To calculate the probability of that sequence, you can't ignore the wins, you have to get the probability of the sequence as it actually would occur, otherwise a comparison with the "straight" sequence of LLLL wouldn't be valid. See what I mean?

Suppose you decide to bet Red only after it hits. It would be the same thing. What would be the chance of getting 5 losses in a row if you're following this rule? You would have to calculate the probability of RBRBRBRBRB, because this is the alternating pattern which must appear in order that you lose 5 times in a row, when following the rule that you only bet on red after a red.

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #427 on: April 16, 2017, 03:06:21 PM »
If the trigger was snow weather, would you take into account the probability of snowing?

It depends on whether you're looking for a "sequence" of snow. Suppose I want to go skiing for a couple of days, and my "trigger" is a day of snow. Since I'm going for 2 days I need to know the probability of 3 days of snow (the trigger day, plus the remaining 2 days). I can then look at the weather records and see how many times 1 day of snow will be followed by a further 2 days of snow.

The following users thanked this post: Reyth

#### TERMINATOR

• Search YouTube for MANDELA EFFECT
• Mature Member
• Posts: 263
• Thanked: 321 times
• Gender:
• MANDELA EFFECT
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #428 on: April 16, 2017, 05:18:45 PM »
Bayes wrote:
Quote
So although both probabilities are small, and both results are financially speaking the same (and 12 bets were made in both cases), the 12 straight losses are 0.91/0.21 = 4.33 times more likely than the same number of losses when following the "only bet after a win" rule. Nice!

Interesting Bayes. So, in other words, are you saying that by using Virtual Bets (after a loss, wait for a win and then bet), it will be over 4 times more likely to avoid a losing streak of 12 losses than playing without virtual bets?

This knowledge could definitely help protect our bankroll!

My longest losing streak was 20 individual losses in a row!!!!

TERM, so that equates to 6 losses in row, counting 3 individual losses as one "attack" loss.

Yes, 6 LEVEL losses in a row. And I finally won my 21st wager on the LAST bet of the 7th level. Then, I won my next bet on the second spin after that.

So, if I was using Virtual Bets, instead of 20 losses in a row, I would have had 4 losses in a row (the initial 3 losses, then after the virtual win I would have bet and lost the first bet but won the second).

Quote
Since the bet selection seems to be in a state of flux (I believe palestis is also testing some other variation), I'm going to hold off recoding  the tracker until we get some firm results one way or the other. In the meantime I will focus on the progression, and add it to the tracker.

Sounds good. I think we could eventually incorporate both your virtual losses AND the XYY trigger into the bet selection eventually.

I was tentatively thinking of it this way: Your virtual losses will ALWAYS be in effect throughout the betting selection. But as a FURTHER safeguard, to do something like this:

Once we LOSE level 1, BEFORE we begin level 2, we simply look for an XYY trigger (no matter what 3 spins came before it) before betting level 2. Then if we lose level 2, THAT is when we look for the XYY special trigger (only play XYY if the 3 spins prior to it had no "X").
« Last Edit: April 19, 2017, 06:05:49 AM by TERMINATOR »

The following users thanked this post: Reyth

#### MrPerfect.

• Hero Member
• Posts: 1156
• Thanked: 657 times
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #429 on: April 16, 2017, 08:12:58 PM »
Nothing will happen wait you for something or do not wait. Result on long run will be same. Variance is limited by limiting degrees of freedom. In this particular case nothing is limited.   If in doubt, try to explain to the ball this theorium of yours and make sure it understood and remembers everything.
There is one way only to protect a bankroll playing this way: keep your losses on the paper.

#### palestis

• Veteran Member
• Posts: 642
• Thanked: 454 times
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #430 on: April 16, 2017, 10:35:46 PM »
Once we LOSE level 1, BEFORE we begin level 2, we simply look for an XYY trigger (no matter what 3 spins came before it) before betting level 2. Then if we lose level 2, THAT is when we look for the XYY special trigger (only play XYY if the 3 spins prior to it had no "X").
There is a good reason y the XYY trigger is so powerful if there is no X 3 spins prior to XYY.
It means that the X only showed up once in 6 spins. And if it doesn't show up in the next 3 spins that will be played, it means that only one appearance of dozen X in 9 spins will have to happen in order to lose the  level.
In the 2nd level the next XYY trigger has to also be limited to one appearance in 9 spins in order to lose. And that X will be a different dozen.
One dozen may disappear for over 20 spins. But a variety of dozens appearing only once in 9 spins, I find it extremely rare.
It's worth testing this trigger for a long time to confirm its better performance. My tests show the same results.
If long term tests confirm  the initial tests,   then it's the best trigger for recovery from losses when using all triggers.
But at the same time it can be used as the only trigger if you bet with high chips.
2-3  wins with \$100 chip and you are done.

« Last Edit: April 17, 2017, 02:38:37 AM by palestis »

The following users thanked this post: Reyth, TERMINATOR

#### TERMINATOR

• Search YouTube for MANDELA EFFECT
• Mature Member
• Posts: 263
• Thanked: 321 times
• Gender:
• MANDELA EFFECT
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #431 on: April 17, 2017, 05:02:53 AM »
Just an update. I just finished another 4000+ spins, 20 games, and the XYY Special Trigger results are still holding at 3.3 wins for every loss. Way above the expectation of 2.3 wins for every loss.

As a side note, my first 20 games, as I already mentioned, my individual wins were 86 spins ABOVE expectation. Which means I ran really good. THESE 20 games, my individual wins were 85 spins BELOW expectation. Which means I ran really bad.

Yet, the XYY trigger results remained relatively unchanged!
« Last Edit: April 18, 2017, 04:51:50 AM by TERMINATOR »

The following users thanked this post: Reyth, ShadowBlue

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #432 on: April 17, 2017, 08:25:16 AM »
TERM, do you have any stats on your results (individual wins/losses) for the XYY trigger (where no X in the previous 3)?

BTW, I know I said I'd post details of the progression, but I'm short on time today, and since it's an adaptation for roulette of a staking plan for sports betting, you can read about it here:

http://thestakingmachine.com/pro.php

The example given is for UK horse racing, where "point" means "unit", and substitute "spin" for "race". The target value given in the example is far too high, but this is a parameter which can be varied to whatever you like, as can the divisor.

The only way my version differs is the introduction of the "bet zero" rule when the last 3 outcomes don't include a win, otherwise the principles are exactly the same. Any questions on it let me know.
« Last Edit: April 17, 2017, 08:48:54 AM by kav »

#### kav

• www.Roulette30.com
• Hero Member
• Posts: 1908
• Thanked: 885 times
• Gender:
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #433 on: April 17, 2017, 08:49:15 AM »
Bayes,
The link doesn't work

#### Bayes

• Veteran Member
• Posts: 688
• Thanked: 556 times
• roulettician.com
##### Re: NEW SYSTEM: SINGLE DOZEN
« Reply #434 on: April 17, 2017, 09:16:24 AM »
That's odd, it works fine in my browser. Do a google search for "The Staking Machine", then click on the "Pro" link in the list of "Back Staking Plans" in the top right corner. The Pro plan is the 7th link down.

The following users thanked this post: Reyth