Baccarat Swing Trading Baccarat

We can cluster groups into patterns with a certain length and create dice with six sides.
Some of the groups will hit and finally, at some point, one group will fall asleep.
Here we would bet that it continues to sleep.

Is the same with football where you make a swing trade that there will be no goals during a certain time interval during the match and where you profit 10% or make a around 40% loss so the win ratio is 1 in 4.
The difference is that you can get some value using pre-match data, for example during the current season there might been less than 5% scoring during the 10 to 20 min mark.
You can use match reading using in-play stats to see if the game is hot and active or slow and passive and if the latter is true you would continue with the swing trade and enter at the 10-minute mark.
If or during the interval 10-20 min mark the game speeds up with shots on target and other indicators you would exit with a smaller profit and if not then you would complete the whole 10% win target.

We don't have that kind of value parameter using random bits.
But there are some statistics we can use to increase our likelihood.

We can measure sequential sleeping frequencies for each sleeping scenario.
We can ask ourselves what frequencies is the average when a group falls to sleep.
This would indicate that we would not continue riding that dispersion for any longer than that.
And we would define a win target with reasoning.
Then we could be measuring at what moment we have a 50, 65, 75, or 95% likelihood of a hit.

So forget about finding cycles with groups and hope for the best when one group fall asleep.
You need to measure the frequencies from different angles to get hit ratio and win targets and combine those with what % of frequencies you will enter and exit.
Here you will discover if you need a smooth progression or can flat betting.

Cheers

 

So if we get the chance of one sleeping event we could just play against one repeat of the same event and have the same math and probability for a hit or no hit.
For example, John Legend tested this approach with eight groups.
So the odds of one in eight were the same no matter how much or how little one even slept.
In the end, you get singles versus series with runs and gaps.

Cheers

 

So we know that EC games produce majority of outcomes in series of 1 and 2 while series 3+ are less frequent. Can we develop a "trigger" to predict when the series of 3+ will arrive? I've been taking a break from live play and today was looking at such a possibility.

Cheers

 

There is the obvious play to bet against a 3+ because it only occurs 1 in 4 “events”. Davis had a system that tracked these events and gave a “trigger” as to when a 3 or 4+ could happen based on the number of 1s and 2s that occurred. My guess is that if that ratio got to over 1:4 then it would be time to begin betting them.

Maintaining two separate bankrolls would be ideal to play both sides of the coin. Your normal play would be to bet against the 3 or the 4 and then you have a separate play for when the 3 and 4 are “due”. Is this kind of what you were thinking @Jimske ? Well the play of following a 2 into a 3 that is.

 

Hello - I am looking at runs and gaps.
Let's say it takes a certain amount of smaller series to show before a larger series show.

For example, a series of two, three and four versus a series of five.
Then sometimes the smaller series will hit many times before the series of five show.

So the run and gaps between each hit of a series of five create frequencies or distances between each occurrence of a series of five.
With raw explanation, we can say we get small runs, medium runs and large runs.

If we swing trading we are interested in medium and large runs.
So using the file I posted in another topic for skips and strikes we can get different percentage values for the occurrence of hit frequencies.

That way we can decide and reflect on what would be a good entering and exit point.
And experiment with different staking plans.

I am about to dive deep into the 6-point divisor plan and with Lanky's description of using and applying safety breaks with new recoup divisors - I have some reflection and thinking process that this way you get as close to alternative staking towards kelly criteria without winning It all back and that you can tackle large samples of negative variance.

Cheers

 

This is how it looks like with gaps - what you learn by manually testing is to see how losses behave and that is one important lesson.
Will you get 5 2 5 2 5 4 5 3 5 2 5 2 5 zig-zag when the smaller series has 75% probability?
Here is another anomaly to look at 555255525255 where fives shop with no repeat present with the 75% probability!
Only testing will reveal the truth and form the rules to consider for optimal play.
Here you want to measure the distances and the frequencies % for appearance.

Code:

2
2
2
2
5    gap 4
2
2
3
2
5    gap 4
2
2
4
2
2
5    gap 5
2
5    gap 1
2
2
5    gap 2
2
3
3
2
2
4
2
5    gap 7
3
2
3
2
2
2
3
2
3
4
2
2
3
4
2
3
2
2
2
3
2
2
2
2
3
2
5    gap 26
2
2
3
2
2
5    gap 5
4
2
4
2
3
2
2
4
2
4
3
3
2
2
2
4
3
2
2
3
4
2
3
4
2
4
3
4
4
2
2
4
5    gap 32
3
3
4
3
3
3
3
2
2
4
3
2
5    gap 12
3
4
2
2
2
2
4
5    gap 7
2
4
5    gap 2
2
2
4
2
3
2
2
2
2
2
2
3
4
2
2
2
2
3
2
2
3
2
2
2
5    gap 24

- - - 1046 spins black rng

 

Here are some sequences using the odds one in three, where we ignore singles and concentrate on a series of two, three and four.
So you would take the framework from skip and streak and measure the cut-point value for the median.
Maybe the majority of sequences have at least two hits as medium value, then you know what to expect.

LWWLWLL RUNS 3
LWLWLWLL RUNS 3
WWLWLWWLL RUNS 5
LWW RUNS 2
WLL RUNS 1
LWWWWWWLWLL RUNS 7
WWWWLWLWWWLL RUNS 8
WWWLWLL RUNS 4
WWWLL RUNS 3
LWLWLL RUNS 2
WLL RUNS 1
WLWLWWLL RUNS 4
WWLL RUNS 2
WWWLWWWLWWLL RUNS 8
LWWLWLWLL RUNS 4
WLL RUNS 1

This is only the Runs and if we would attack once after one fictive win we would get...
WLWWWLLWWWWLWLLLWWWWLL

Now this is a sequential way of looking at how long the medium value is for such a wave.
In this case, the events are two, three versus four and should be the same with one, two versus three.

Here I know from experience that you can get no wins and one win and no wins chopping for a long period - so there is some need for observations on how to tackle a black swan as I mentioned above.

Now let's take the same sequences and look at the loss performance.

LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL LL RUNS 2
LL RUNS 1
LL LL RUNS 2
LL RUNS 1
LL LL LL RUNS 3
LL RUNS 1
LL LL LL RUNS 3
LL RUNS 1
LL RUNS 1

Did you notice that we get singles loses with groups of two the majority of times?
And that we only got single series of loses with several in a row.
Did you notice that we did not get a series of loses chopping with several series of loses after each other, you should make notice of this way of observations.

What to look at
Well if you have measuring at least 1000 triggers or sequences.
Then you should have the median value and let say its is two of winning runs in each sequence.
Now that means you can look at a fictive win after such run of loses.
If not then a losing sequence from hell would look like this with the example above.

W W W W W W LL W LL W LL W LL W

My suggestion is not only to focus on winnings and winnings strike ratio.
Also focus on loses and how they performed with or without fictive wins.
And look at series as single losses and losses as series.
Maybe you find that the true meaning behind hit and run is to bank on single groups of loses and try to avoid groups of series of loses.

Closer to the true meaning of local attractors.

Cheers

 

Here are another 300 TRNGs that result in 4 losses once when concentrating and focusing on losses and not winning bets.
So if we take both samples with 600 TRNGs we get the following results with John's baccarat method.

W W W W W W LL W LL W LL W LL W W - sample 1
W W W W LL W W LL LL W W W W LL W W - sample 2

Sample 1 based upon TRNGs and 2, 3 versus 4

LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL LL RUNS 2
LL RUNS 1
LL LL RUNS 2
LL RUNS 1
LL LL LL RUNS 3
LL RUNS 1
LL LL LL RUNS 3
LL RUNS 1
LL RUNS 1

Sample 2 based upon TRNGs and 1, 2 versus 3

LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL LL LL RUNS 3
LL RUNS 1
LL RUNS 2
LL LL LL LL LL RUNS 5
LL LL RUNS 2
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL RUNS 1
LL LL RUNS 2
LL RUNS 1
LL RUNS 1

 

One more detail i want to test this with two attempts as above but with the odds 1 in 4.

BBB W
BBP W
BPB LW
BPP LL

25% versus 75% and not 33% versus 66%
Maybe there is a slight difference in the long term.

Cheers

 

Hi Patrik,

What would you presume would be the average number of bets per shoe with John’s method using your analysis?

 

Hello, I am testing 1000 triggers or sequences and will focus on both runs and gaps from different angles.
I don't have the median value and will not share the results when I have it.
Because I want others to do their homework, I just describe how to measure what you can measure and what kind of observation to do.

But let's say the median value is 3 hits within each sequence or two, then you have a solid approach betting once after each losing group.
And bet against a series of losses to chop.
That would - I assume - result in strikes with ten and above as winning bets before you hit four losses regularly.
Then you have the foundation of success, but to know that for sure each one needs to crack some random bits.

But my main reason for this is the horse race for the UK and Irland where I have a selection process for Lay The Favorite
I have 10 years of results/stats from horse races with Lay The Favorite - but I don't use that data until I test 1000 triggers or sequences using TRNGs as my first sample and after that backtesting with a real true sample size to get an estimation for how many years I can bet the horse races with success and see expectations of drawdowns and winnings each month.

My aim is to bet once for one group with two attempts and bet against a series of groups of losses.
Using a 6-point divisor plan with Lanky's approach for Saftey Breaks (new divisor for reducing or recouping losses when a bad drawdown)

My goal is to develop a method that can stand for several years of betting with around 25 triggers each month.
And we talk bout profits from 1 to 5 units each month.

Cheers

 

I have a method with long shots in horse racing. I love it.

 

For baccarat, you would risk five steps to win 18.5 units for around 365 trails.

Simulated one year of betting on horses using the method above - did not use a 6-point divisor - just experiment with five bets 1 1 1,5 2 2,5
I used a hybrid between ongoing march and pause situations.
18,5 Units during one year of betting is a very nice overall profit.

And I would end up on the second winning level with Johns's method.

 

I think I will test this extended fibo with 10 years of data.
I am pretty sure it will not bust and I am more curious about the monthly profit and potential drawdowns.

1
1
1,5
2
2,5
3
4
5
6,5
8
10,5
13
17
21
27,5
34

 

I could also test one extended version using six attempts cycles and measuring the Runs & Gaps

Then the progression levels would look like this

1 1 1,5 2 2,5 3 3,5
2 2 3 4 5 6 7
And one more ...

That would give us cycles with a 1,5% likelihood to fail and 98,5% to succeed.
But we don't know the value for the median for the Runs and the length of strikes for each sequence before hitting three losing groups.
For each level.
We need to find the median value of the gaps and measure the % of the likelihood of hitting our targets with a four-level progression.

I think or feel that this is interesting data to explore.
So a winning level until a bust then recoup to zero and start over.
That was the idea with a 6-point divisor, but maybe the Ching A Ling staking will perform on the same level.

Cheers

 

111222333555888131313212121... use this

 

I like this. And you reset once you get into profit?