Fibonacci and prime number occurences

[laurontario]

I started looking at how many Fibonacci and how many prime numbers appear in each draw and build a statistic and tonight I was wondering, what if we could look at the combination of these two.

I ran a statistic for my 649 and here are the results:

Fi-Pr Occur. %
0-0 100 4.0%
0-1 307 12.2%
0-2 278 11.0%
0-3 133 5.3%
0-4 25 1.0%
0-5 2 0.1%
1-0 93 3.7%
1-1 301 12.0%
1-2 352 14.0%
1-3 215 8.5%
1-4 68 2.7%
1-5 8 0.3%
2-0 35 1.4%
2-1 118 4.7%
2-2 181 7.2%
2-3 122 4.8%
2-4 46 1.8%
2-5 3 0.1%
3-0 4 0.2%
3-1 14 0.6%
3-2 49 1.9%
3-3 35 1.4%
3-4 14 0.6%
3-5 4 0.2%
4-1 2 0.1%
4-2 4 0.2%
4-3 2 0.1%
4-4 1 0.0%

Would it be possible to implement this in the statistic an also be able to filter based on this?
Of course, one can create a compound filter but it would be much easier if we had something implemented in EL, pretty much the same way we have for segment combinations.

Do you think this is something we could use?

[Jerzy]

laurontario,

There are the following Fibonacci numbers in 6/49 game:

1, 2, 3, 5, 8, 13, 21, 34. That's 8 numbers.

And there are the following Primary numbers:

1, 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47. That's 16 numbers. (Some people think that number 1 is not a primary number).

The numbers 1, 2, 3, 5 and 13 are common to both groups, they are both Fibonacci and Primary numbers. They make 62.5% of Fibonacci numbers and 31,25% of Primary numbers. So you compare very similar groups of numbers.

None of the groups of Fi-Pr numbers you presented above has the occurrence higher than 14.0%. Actually, there are many groups which have a very low occurrence. Removing such a group would result in removing a small number of ticket from the package.

What do you think of it?

Jerzy

[laurontario]

Jerzy,

you are right in what you are saying.
I ran a query on all 14 mils possible combinations to see the total number of occurrences of fibo-prime combinations and - no wonder - the numbers I got were very close to the numbers I sent earlier (which, as I said, came from the Canadian 649).
However, I ran another report to see the number of repetitive occurrences and I noticed that in only a few cases the last fibo-prime combination appeared in the next draw. How can I use this? Well.. let's assume that the last draw I had one Fibo and two primes (1-2), same as in the last draw. In the statistic I can see that this combination appeared twice in a row only 3.3% from the total and has never appeared three times in a row. I could probably eliminate all tickets with this combination. In other words, I could knock off 2,177,175 from 14 mils without eliminating the winning combination.

This is what I hope to achieve with this.

I have already put everything into an MS Access table but it's not very convenient, since I have to filter my package in Access, then export, import in EL... It would be a lot easier if we had it implemented in EL.

Hope my strategy is clearer now.

[Jerzy]

laurontario,

Yes, the occurrences of two different categories of numbers at the same time are less common than the occurrences of each category separately. And as a general rule repeats of two different categories of numbers occur even less commonly. Any category of numbers can be used this way not only primary or Fibonacci numbers.

But if two number groups share some of their numbers, it means that the groups are poorly categorized, and it is likely to decrease the effectivenes of filtering.

Good luck in your research.

Jerzy

[laurontario]

[quote=Jerzy]

But if two number groups share some of their numbers, it means that the groups are poorly categorized, and it is likely to decrease the effectivenes of filtering.

[/quote]
Hi Jerzy,

I understand perfectly, however, a similar statistic is already implemented in EL - Odd/Even+Hi-Low. There are 24 low numbers and half of them are odd, the other half, even. Well, in this case one group share 50% of numbers from the other group too!

As I said, I think it is interesting to see how do these combinations alternate and what is the number of repetitive combinations. I am not saying this is THE solution, only that I am trying to look at these things from another perspective.

As an example: for the last five draws I had 4 numbers from the previous ten draws. This happened in only 2.4% of the total draws. There were some odd cases when this repetitive pattern happened seven times in a row, however, that was in one case only. My predictions was that in the next draw I will not have again four numbers from the previous ten - and this is exactly what happened!

On a similar principle, I can 'predict' that the next draw will have most probably all six numbers from the previous ten (will see Wednesday night if I was right or wrong).

Back to the Fibo-Prime combination, as I said, there were only very few cases when I had a repetition. Based on this I could eliminate easily up to 15% of the total number of combinations and keep the winning ticket (of course, we are dealing with probabilities, so, I cannot say this will happen for sure, only that, in my opinion and based on the previous history, this is what I THINK will happen).

[Jerzy]

laurantio,

On the grounds of your research you are convinced that a filter based on the proportion of numbers coming from two groups- Fibonacci and Primary numbers - would be useful or very useful. And the statistics support your idea.

However, I believe that using two groups of numbers which are entirely different from each other would give better results. These 2 groups of numbers should be preferably user defined and replaceable. I am in favour of using randomly generated numbers. Such groups of numbers can be found in the rows of History Differences in any column. The size of each group should be adjustable, probably not less than 6 numbers in pick 6 game and not more than a half of all numbers in the game.

This way the filter would not be limited to just Fibonacci and Primary numbers, but of course it would be more complicated and more difficult to implement.

Jerzy

[Jerzy]

Stan,

I know I am asking you for new filters too often, but this time I want to support laurantio's idea of the filter based on the proportion of numbers coming from two groups of numbers - Fibonacci and Primary numbers.

We are asking for a possibility of filtering from the statistics table which would show the occurrences of all possible proportions (combinations) of these two groups of numbers (as described by laurantio).

The filter would be more universal if you could make the numbers in these two groups changeable, to be entered by the user.

Could you do it for us, please?

Jerzy

[laurontario]

[quote=Jerzy]
Stan,

I know I am asking you for new filters too often, but this time I want to support laurantio's idea of the filter based on the proportion of numbers coming from two groups of numbers - Fibonacci and Primary numbers.

We are asking for a possibility of filtering from the statistics table which would show the occurrences of all possible proportions (combinations) of these two groups of numbers (as described by laurantio).

The filter would be more universal if you could make the numbers in these two groups changeable, to be entered by the user.

Could you do it for us, please?

Jerzy
[/quote]

Thanks for your support, Jerzy.
If Stan could make this flexible, allowing the user to select two numbers for the combination, even better!

I am currently looking at the LowHi+OddEven combinations - I am curious to see how many repetitive occurrences I have.

[laurontario]

[quote=Jerzy]
Stan,

I know I am asking you for new filters too often, but this time I want to support laurantio's idea of the filter based on the proportion of numbers coming from two groups of numbers - Fibonacci and Primary numbers.

We are asking for a possibility of filtering from the statistics table which would show the occurrences of all possible proportions (combinations) of these two groups of numbers (as described by laurantio).

The filter would be more universal if you could make the numbers in these two groups changeable, to be entered by the user.

Could you do it for us, please?

Jerzy
[/quote]

Thinking about this one... Probably the best idea would be to allow the user to select one or max two numbers and calculate the repetitive occurrences for this. Also, the number of possible combinations should be reasonably small. For example, I wouldn't combine StDev with anything, because the number possible values in a 649 would be very high and not efficient. Even though there is no repetition from one draw to another, if one filters out the combinations for a specific StDev, only a few tickets would be eliminated so... just a waste of time in my opinion.

[Jerzy]

laurantio,

I discussed a possibility of filtering tickets depending on the proportion of numbers that originate in two groups. These two groups should ideally contain randomly selected or randomly generated numbers. The numbers from the first group should not be repeated in the second group. For example:

1st group: 1, 5, 14, 17, 21, 23, 25, 31, 35, 37, 38, 45
2nd group: 2, 4, 7, 10, 13, 19, 22, 29, 32, 39, 40, 41

Each group has 12 random numbers. There are no repeated numbers.

In pick 6 lotto the following combination are possible if the numbers do not repeat in groups:

1st/2nd = 0/0, 0/1, 0/2, 0/3, 0/4, 0/5, 0/6, 1/0, 1/1, 1/2, 1/3, 1/4, 1/5, 2/0, 2/1, 2/2, 2/3, 2/4, 3/0, 3/1, 3/2, 3/3, 4/0, 4/1, 4/2, 5/0, 5/1, 6/0

That's 28 possible settings. If the numbers in groups repeat, there will be even more possible settings, depending on how many numbers are repeated (e.g. 4/4 or 6/6).

For example, the setting 2/4 means that 2 numbers in the ticket are from the 1st group and 4 numbers from the 2nd group.

In pick 6 lotto:

If two groups of numbers have 6 numbers in common, these 6 numbers could appear in one ticket, and the filter setting to deal with such situation would be 6/6.

Sorry, I do not want to discuss the possibility of filtering the tickets depending on the relation between StDev and other parameters. It seems to be too complicated.

Jerzy

[laurontario]

[quote=Jerzy:1204644965]
In pick 6 lotto:

If two groups of numbers have 6 numbers in common, these 6 numbers could appear in one ticket, and the filter setting to deal with such situation would be 6/6.
[/quote]

Yes, but how often would that happen? Not very often I would say...

[quote=Jerzy:1204644965]
Sorry, I do not want to discuss the possibility of filtering the tickets depending on the relation between StDev and other parameters. It seems to be too complicated.
[/quote]

I agree, that is why I said in my last post that I wouldn't combine StDev with anything, however... I have another idea:
We could create and use some StDev groups as follows:

(that is, greater than first value, smaller or equal than last)
0... 3 = 1
3...6 = 2
6...9 = 3
9...12 = 4
12...15 = 5
15...18 = 6
18...21 = 7
21...24 = 8
>24 = 9

Min StDev value in a 649 is 1.87, highest is over 25.

I ran a statistic on my CAD649 - see attached JPEG file. As one can see, the highest occurrence is in the range 12...15, with a max of 8 repetitions in a row (however, after 4 repeats the number drops drastically).

I am still running a report in Access to see how many numbers do we have in each group (from the max possible combinations). This is another way we could use to filter out some tickets.

We could probably use smaller ranges for the middle values (step 2 instead of 3) - this way, the number of occurrences and repeats would be smaller.

[PadawanLotto]

[quote=Jerzy:1204455964]
laurontario,

There are the following Fibonacci numbers in 6/49 game:

1, 2, 3, 5, 8, 13, 21, 34. That's 8 numbers.

And there are the following Primary numbers:

1, 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47. That's 16 numbers. (Some people think that number 1 is not a primary number).

The numbers 1, 2, 3, 5 and 13 are common to both groups, they are both Fibonacci and Primary numbers. They make 62.5% of Fibonacci numbers and 31,25% of Primary numbers. So you compare very similar groups of numbers.

None of the groups of Fi-Pr numbers you presented above has the occurrence higher than 14.0%. Actually, there are many groups which have a very low occurrence. Removing such a group would result in removing a small number of ticket from the package.

What do you think of it?

Jerzy
[/quote]

As discused before somewhere in this fourm, the thing about prime numbers is that in a 6/49 there are 16 prime numbers which is 1/3 fo the total amount of numbers possible and in a 5/39 there are 13 prime numbers which is 1/3 of the total numbers possible. Seeing how all but one of the prime numbers are odd numbers and you are dealing with 1/3 of the total numbers possible you are going to find that most drawings will only produce 1 to 2 prime numbers for a 5/39 and 1 to 3 prime numbers for a 6/49.

If you add the Fibonacci numbers you have a total of 19 numbers for a 6/49 that only have 3 even numbers and producing 1 to 3 numbers most drawings. I'm sure that a good strategy could be devised using these numbers but, I think that it would have to be a trap strategy where you use the same settings every drawing.

[stan]

[quote=Jerzy:1204560746]
Stan,

I know I am asking you for new filters too often, but this time I want to support laurantio's idea of the filter based on the proportion of numbers coming from two groups of numbers - Fibonacci and Primary numbers.

We are asking for a possibility of filtering from the statistics table which would show the occurrences of all possible proportions (combinations) of these two groups of numbers (as described by laurantio).

The filter would be more universal if you could make the numbers in these two groups changeable, to be entered by the user.

Could you do it for us, please?

Jerzy
[/quote]

hmm, i wonder what's keeping joe busy these days:)

can you post or email me some pictures/mockups how the filter should look like?

[Jerzy]

Stan,

I am thinking of two basic versions of two number groups filter. The first version would allow for numbers in common in the groups, the second one would not allow for that. Both versions would allow to enter the number groups according to the user's choice, not just Fibonacci & Primary, but any number groups (eg. in the group I numbers which occurred in the last 10 draws and in the group II the numbers that didn't occur in the last 10 draws).

The examples are for pick 6 lotto.

Version 1

It would look similar to the Odd/Even + Low/High filter. There would be 49 lines in the table with filter settings (from 6 + 6 to 0 + 0) and statistical data. In addition there would be two game panels (like those in the Match Number Filter) to enter numbers by clicking on them.

The game panels should be marked as the 1st and the 2nd.

Initially, the filter would count how many numbers from each panel occurs in each combination in the package. It would do it separately for each panel (first for the 1st panel).

In the next step the filter would compare the results to arrive at the proportion of numbers coming from these two groups which are present in the combination.

And in the next step the filter would "see" whether the proportion of numbers in the combination is consistent with the setting specified by the user.

There would be 49 possible settings (similarity to the Odd/Even +Low /High Filter): 6 + 6, 6 + 5, 6 + 4, 6 + 3, 6 + 2, 6 + 1, 6 + 0, 5 + 6, 5 + 5, 5 + 4, 5 + 3, 5 + 2, 5 + 1, 5 + 0, 4 + 6, 4 + 5, 4 + 4, 4 + 3, 4 + 2, 4 + 1, 4 + 0, 3 + 6, 3 + 5, 3 + 4, 3 + 3, 3 + 2, 3 + 1, 3 + 0, 2 + 6, 2 + 5, 2 + 4, 2 + 3, 2 + 2, 2 + 1, 2 + 0, 1 + 6, 1 + 5, 1 + 4, 1 + 3, 1 + 2, 1 + 1, 1 + 0, 0 + 6, 0 + 5, 0 + 4, 0 + 3, 0 + 2, 0 + 1, 0 + 0

Version 2
The numbers in common not allowed. Because of that there would be only 28 possible settings which could be arranged like segments in the Segment Combination Filter. Selection of settings by ticking chekboxes.

1st/2nd = 0 + 0, 0 + 1, 0 +2, 0 + 3, 0 + 4, 0 + 5, 0 + 6, 1 + 0, 1 + 1, 1 + 2, 1+ 3, 1 + 4, 1 + 5, 2 + 0, 2 + 1, 2 + 2, 2 + 3, 2 + 4, 3 + 0, 3 + 1, 3 +2, 3 + 3, 4 + 0, 4 + 1, 4 + 2, 5 + 0, 5 + 1, 6 + 0

The sum of occurrences in each setting is not higher than 6 for pick 6 lotto. ( in version 1 filter that sum was sometimes higher than 6).

There should be two game panels to enter the numbers, but unlike in version 1 entering the numbers which are already entered in other panel would be impossible. An attempt to do so would generate an error message.

Apart from the mechanism protecting from the numbers in common and lower number of possible settings, processing the numbers would be similar like in the version 1.

Jerzy

[stan]

ok then, i'll look at it
filter should be fairly easy but there's a problem with stats if you want to keep the number groups user-defined