Euromillions

:whoosh:

 

Answer me this then random number experts.

Since the lottery began, the number 38 has been drawn 228 times, and the number 20 has been drawn 162 times, that's a difference of 66 times drawn. So when does this random-ness even out?

 

Randomness doesn't necessarily even out. Something random does not have any form of pattern. In effect its as if each draw is the first ever draw - there is no link to previous numbers, its just pure chance.

Thats assuming the balls and machines are perfectly unbiased, of course.

 

Technically the part of his quote above is true.

If the probability of getting 1,2,3,4,5,6 is p then the probability of not getting 1,2,3,4,5,6 is 1-p.

Since the probability of getting 1,2,3,4,5,6 is very small due to the possible number of permutations in the lottery, 1-p > p.

Therefore, you are more likely to not get 1,2,3,4,5,6 than you are to get 1,2,3,4,5,6.

I know we like to bait simms, but in this case he is right.

Of course King Dev is also right that given a totally fair machine and balls, each combination is as likely as any other combination. The fact that a combination was drawn last week doesn't have any influence on the probability that the combination will occur again in the future.

 

100% correct, only because there are only five numbers in one of those sequences and six in the other !

Mate, i'm a little worried about that comment

 

If each time is it vastly more likely to a degree of something near 0.99999999, than seeing as no balls have memory, then it is a verisimilitude that 1,2,3,4,5,6 will never be drawn as long as the lottery is continued. It would take over 12.2 billion years assumeing a different series is drawn each bi-weekly draw.

The sun will only burn for another 5 billion years, aproximately, thus remembering that the lottery balls have no memory, it is once again a verisimilitude that they will not be drawn.

semantics, and yes sorry, it did sound a bit rapey didn't it. My apologies.

 

What are the odds on the sun burning out in 5 Billion years, worth a punt ?

 

In effect what you are saying is that it is not possible for any combination of any sort to be drawn out.

 

Dead cert isn't it?

Indeed. It's Zeno's arrow paradox all over again.:naughty:

 

Simms is proof positive that a little knowledge is a dangerous thing.

 

also proof that the monkey, keyboard, infinite time, Shakespear theory is true

 

If Simms can't grasp the logic behind the probability of 1-6 coming up on the lottery, I doubt he will get a handle on "To be, or not to be" being cracked out on an Olivetti by a baboon.

 

It's a verisimilitude, it's not true.

 

I know you learnt a new word, but you're still wrong.

 

Don't know what that word means , is it something like very little chance ?

 

It means as close to a truth as you can possibley get.

 

while still being wrong !

 

138million jackpot tonight.

Is the forum feeling lucky?

 

No, we wasted enough income on the last one. I might buy myself a ticket though.

 

dont go for lucky dip. I bought two lines and they both had the same bonus ball numbers. That's a 1 in 36 chance:rant:

 

After this discussion before I thought I'd write a program to see if any sequences come up when randomly drawing 6 balls of 1-49.


I wrote the code in MatLab, assigning 6 random numbers to an array 300,000 times. (The current UK lottery has only had about 1,640 draws)

Code:

Spoiler

Code:

m=300000;
for n=1:m
    a=round(rand*48)+1;
    b=round(rand*48)+1;
    c=round(rand*48)+1;
    d=round(rand*48)+1;
    e=round(rand*48)+1;
    f=round(rand*48)+1;
    while (b==a)
        b=round(rand*48)+1;
    end
    while c==a || c==b
        c=round(rand*48)+1;
    end
    while d==a || d==b || d==c
        d=round(rand*48)+1;
    end
    while e==a || e==b || e==c || e==d
        e=round(rand*48)+1;
    end
    while f==a || f==b || f==c || f==d || f==e
        f=round(rand*48)+1;
    end
    lotto(n,1)=a;
    lotto(n,2)=b;
    lotto(n,3)=c;
    lotto(n,4)=d;
    lotto(n,5)=e;
    lotto(n,6)=f;
    seq(n)=a+b+c+d+e+f;
    aseq(n)=max(lotto(n,:))-min(lotto(n,:));
    
end


minindex = find(seq==min(min(seq)));
%display('lowest sequence = ')
%lotto(minindex,:)

anyindex = find(aseq==min(min(aseq)));
%display('closest sequence = ')
%lotto(anyindex,:)

fprintf('The lowest sequence is: \n %g %g %g %g %g %g\n', lotto(minindex,:))
fprintf('The closest sequence is: \n %g %g %g %g %g %g\n', lotto(anyindex,:))

Low and behold, on draw number 40,370: the numbers 41, 42, 43, 44, 45, 46 came up (not in that order). I was quite surprised as I wasn't expecting a sequence within 300,000 draws but there you go.

View attachment 1487

If this tells us nothing else it as least tells me I have too much time on my hands.

 

An interesting experiment.

You do know that there is not a truly random function in a computer don't you? The closest you can get is to is to trigger on an external event. Randomize Timer is sometimes used for this as it will be down to when the user runs the command.

 

Yeah I did think someone would mention that, but maybe that could represent any un-random elements within the real ball choosing method as used in the actual lottery?

Either way, any unrandom element inherent in the program wouldn't be able to give bias to sequential results. But yeah, you're right.