Page 1 of 4
One for the maths experts...
Posted: Mon Sep 23, 2013 11:30 am
by jimbo
Random question this but it's been bugging me, and I consider myself quite good at maths
The national lottery: pick a number, say 12, you have the same odds of number 12 coming out as any other number in the machine, right? So realistically you could choose any set of numbers you want, your odds of winning don't increase or decrease. But realistically NOBODY would choose 1-2-3-4-5-6-7, because somehow it feels even less likely that not only will your numbers come up, they'll be in a perfect sequence. Hold on, we've just established that the odds of any number coming up is the same as any other number coming up, so what am I missing??
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 12:15 pm
by David
The odds of 6 numbers coming up in a perfect sequence is much higher than any 6 random numbers. However, it is not a requirement to win, so it has no influence on the odds of winning. It is just in your mind that you have added an extra parameter. It would be unusual to see 6 sequential numbers win, just as it might to see just even or odd numbers but that's irrelevant to winning.
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 2:01 pm
by jimbo
thanks for the reply, appreciate it

Re: One for the maths experts...
Posted: Mon Sep 23, 2013 2:03 pm
by sendmyusername
We need a like button...
This is the sort of questions that fill my head when i'm on nightshift.

Re: One for the maths experts...
Posted: Mon Sep 23, 2013 2:17 pm
by BiggestNizzy
I'm sure the Israeli lottery had the same numbers 2 weeks in a row.
edited to add it was 3 weeks apart
http://www.telegraph.co.uk/news/worldne ... efore.html
Humans see patterns where there are none. It's the same thing as pictures of Jesus on a dog's bum / toast.
Good thing to try is toss a coin 10 times and write down the results you will see a pattern in the randomness.
Also if your in a room of 30 people 2 of them will probably share the same birthday.
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 2:30 pm
by robin
In fact there is a good reason to choose a set of consecutive numbers ... because people shy away from them (thinking they are somehow less random than any other set of numbers) ... so if you choose them and they come up you're less likely to share the prize money with anyone else.
Similarly you should avoid all numbers below 32 because people often use their family birthdays as part of their number choice and these increase the use of the numbers 1-12 and 1-31 ...
So were I to play (which I don't) I would choose 32,33,...,37.
Finally, there is bias in every system, no matter how well we attempt to design it out. So you should perform an analysis of all the numbers that have ever come up (grouped by set of balls/machine/both/neither). Any number that comes up more frequently should be a better bet in case there is a bias towards that number for some reason.
BTW, the probability of your numbers coming up (6 numbers from a set of 49) is:
6/49 * 5/48 * 4/47 * 3/46 * 2/45 * 1/44 = 720 / 10068347520 or 1 in 13,983,816.
Cheers,
Robin
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 3:06 pm
by jason
When the lottery began, I remember reading somewhere that hundreds (even thousands) of tickets eqch week were for 1,2,3,4,5,6. Presumably all those folk thinking nobody else would have a similar unique' tactic
You are, apparently, less likely to share your jackpot if you avoid birthdays. So pick at least some numbers >31.
[doesn't do the lottery faction]
<edit to add> oops, just spotted Robin made the same birthdays comment

Re: One for the maths experts...
Posted: Mon Sep 23, 2013 3:32 pm
by Corranga
Hmm,
BUT I am likely to make the assumption that picking 6 consecutive numbers is likely to be strayed from (as said by Robin and Jason).
I am also likely to make the assumption that people will generally choose birthdays
Then, I'd assume that people will often go for the boundaries, so 49 is out too.
so 32, 33... would in theory be my picks too.
So, I then realise that, since I'm not unique, that people will in fact possibly skew towards choosing numbers that they think others won't, which means that there is just as much chance of them choosing 32, 33.... as me, meaning I should just choose a lucky dip.
Safest bet, if I win, great. If I loose, I don't feel like I have to play the same numbers ever again. If my numbers come up in a week when I don't play, I'd never know as I wouldn't have memorised them..
As a final observation, I don't play the lottery either, so there seems to be a definite bias, and possibly a link between those that do not play the lottery drawing conclusions based upon birthdays and consecutive numbers, meaning that they are right back in there as a good choice to become a sole winner

Re: One for the maths experts...
Posted: Mon Sep 23, 2013 5:54 pm
by robin
There was a weird thing early on in the lottery where the jackpot was won by two people, but one of them had 100's of tickets for the same numbers ... what a sh*t for the guy that just bought the one ticket, then thinking you had won the jackpot, only to discover you'll only get 1/250th of it or something!
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 6:10 pm
by tut
Any combination of six numbers coming up is the same as any other six numbers. Each one picked out is random. 123456 has not come up yet but has as much chance as any other combination.
The consideration is maximising the odds of what numbers others will choose so that you cut down the chances of sharing, ie:- some numbers will not fit in with a birthday date so they are probably chosen less.
tut
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 6:20 pm
by C7Steve
robin wrote:There was a weird thing early on in the lottery where the jackpot was won by two people, but one of them had 100's of tickets for the same numbers ... what a sh*t for the guy that just bought the one ticket, then thinking you had won the jackpot, only to discover you'll only get 1/250th of it or something!
I have heard of a guy that put on his works syndicate lottery line every week. He used to buy two tickets with the same numbers on so he was guaranteed half the winnings if successful. Don't suppose he would be very popular if the jackpot came in and his fellow workers found out about it afterwards.
Steve.
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 10:11 pm
by jason
robin wrote:There was a weird thing early on in the lottery where the jackpot was won by two people, but one of them had 100's of tickets for the same numbers ... what a sh*t for the guy that just bought the one ticket, then thinking you had won the jackpot, only to discover you'll only get 1/250th of it or something!
I always thought that would be the ultimate act against your worst enemy, of you had the cash.
I vaguely remember one jackpot early on being shard by about one thousand tickets. Or hundreds at least. But I don't remember that incident connected to your story, guess it was the same event.
Google hasn't helped me find record of it.
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 10:13 pm
by jason
C7Steve wrote:robin wrote:I have heard of a guy that put on his works syndicate lottery line every week. He used to buy two tickets with the same numbers on so he was guaranteed half the winnings if successful. Don't suppose he would be very popular if the jackpot came in and his fellow workers found out about it afterwards.
That's a pretty low thing to do!
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 10:23 pm
by jason
A spot of Googling says the biggest ticket share was 133 winners, each getting a mere £122,500 share of the £16m jackpot back in 1995. That's the event I'd vaguely remembered.
Apparently approx 10,000 people enter 1,2,3,4,5,6 each week. A simple Google would tell them of their folly!
Re: One for the maths experts...
Posted: Mon Sep 23, 2013 10:39 pm
by BiggestNizzy
A good few years ago my brother got 5 numbers and thought we was in for a £100k or something , he got £250 quid.