New 649 filters.

[magusprovernum]

Was looking at the numbers 1-49.

36.7% are prime numbers.
33% are divisible by 3.
18.4% are divisible by 5.

In six numbers drawn, two should be prime; two divisible by 3 and one divisible by five. Look at the following set:

Canada 649, May 05, 2007

01 23 26 28 33 40

Primes: 01, 23
Div3 : 26, 33
Div5 : 40

I am playing the following 14 random numbers in tonight's Ontario49 (May 16):

2 5 9 10 11 24 28 30 35 38 41 45 47 48

When filtered by my decade-one rules, sums 125-175 and the above new rules we get the following 10 sets:

2 9 11 35 38 48
2 9 28 35 41 48
2 9 28 35 47 48
2 9 28 38 41 45
2 9 28 38 45 47
5 9 24 28 38 41
5 9 24 28 38 47
9 10 11 24 28 41
9 10 11 28 47 48
9 10 11 38 47 48

You will notice that the number 30 got filtered out.

Just a new idea to toy with. How many numbers do you play? I have found that you should never play more than 14, because after that the number of new sets you have to play are greater than the difference in odds.

 

[GillesD]

New filters

Was looking at the numbers 1-49.
36.7% are prime numbers.
33% are divisible by 3.
18.4% are divisible by 5.

I have some difficulty with your calculations. Most places you will look for prime numbers under or equal to 49 (such as a search on Google) will give you those:
- 2, 3, 5, 7, 11 13, 17, 19, 23, 29, 31, 37, 41, 43 and 47
with number 1 not being considered a prime number, so this gives 15 prime numbers or 30.6% (15/49).

And if you want to keep the same precision on your three new filters, then there are 16 numbers that can be divided by 3 or 32.7% (16/49).

 

[GillesD]

New filters

And for those interested in actual results compared to the theorical values, after 2,436 draws (or 14,616 numbers drawn) in the Canadian 6/49:

- there were 4535 prime numbers or 31.0% compared to 30.6% expected;
- there were 4713 numbers divisible by 3 or 32.2% compared to 32.7% expected;
- there were 2680 numbers divisible by 5 or 18.4% compared to 18.3% expected.

Fairly near what is espected in all cases.

 

[magusprovernum]

I have some difficulty with your calculations. Most places you will look for prime numbers under or equal to 49 (such as a search on Google) will give you those:
- 2, 3, 5, 7, 11 13, 17, 19, 23, 29, 31, 37, 41, 43 and 47
with number 1 not being considered a prime number, so this gives 15 prime numbers or 30.6% (15/49).

And if you want to keep the same precision on your three new filters, then there are 16 numbers that can be divided by 3 or 32.7% (16/49).

Thank you. Again, most of the time I'm running full speed and make mistakes. In this case it was a two line peice of code. Somehow 21 and 27 crept into the list and I did indeed include 1 as prime... and will continue to do so for my calculations.

As I've stated many times, my math skills are lacking and I do make mistakes quite often. However, my methods of attacking the lottery have always been unique and I am confident that within the next year you will see my face on the OLGC website smiling and holding a cheque!

Can you check to see how many times this pattern has been drawn in your history file? ie, 2 prime - 2 Div 3 - 1 Div 5. It should be roughly 1 in 40. If I make any other errors please do not hesitate to point them out as I do apreciate the help.

 

[GillesD]

Combined data for new filters

Good thing I did save the file I used to calculate my previous posting.

Actually, after 2436 draws, it happened 134 times that in the winning numbers, there were 2 prime numbers, 2 numbers divisible by 3 and 1 number divisible by 5. This amounts, on average, once every 18 draws.

Of course, this does not mean that these filters account for 5 of the six numbers, as both 3 and 5 are prime numbers as well as divisible by 3 or 5. For example, numbers 3, 5, 6, 8, 14 and 16 meet your criteria.

You can inckude whatever numbers you want in your calculations but refering to number 1 as a prime is not right. It would be the same as saying all 25 even numbers (and I include number 13 in this as it is my lucky number). You could say something like "prime numbers plus number 1".

 

[blitzed]

Hello GillesD, ironic that I've been messin with prime numbers for playfield prediction as well for the last few days heh!

did you see the crazy list o primes?
http://en.wikipedia.org/wiki/List_of_prime_numbers

seeya,
blitzed

 

[GillesD]

Prime numbers


Thanks for the reference. I never thought you could look at prime numbers in so many ways.

And as far as lotteries are concerned in relation with prime numbers, looking at the lexographic value of a combination and determining if it is a prime or not could interesting.

I have started looking at the lexographic value of winning combinations up to now and I am down to about 400 that could still be prime numbers.

Anybody knows of a VBA procedure or function that could tell me if a given number is a prime number or not.

 

[blitzed]

Hi GillesD,

perhaps this code will do the trick:
http://www.bluemoosetech.com/microsoft-excel-functions.php?jid=29&title=Microsoft%20Excel%20Prime%20Numbers

later,
blitzed