South African lottery “unlikely” numbers fraud

New Topic
This topic has been archived, and won't accept reply postings.
 Phil Lyon 02 Dec 2020

So the South African lottery are investigating after many people claimed scam and fraud when the numbers 5,6,7,8,9,10 came out of the machine. 

https://www.bbc.co.uk/news/world-africa-55154525

If this is a suspiciously unlikely permutation, what would be a more likely one?

Post edited at 17:13
 Toerag 02 Dec 2020
In reply to Phil Lyon:

There is no more likely one due to the chance of any number coming out in any position being equal (in theory).

 bigbobbyking 02 Dec 2020
In reply to Phil Lyon:

If you did 'fix' the lottery to produce the numbers you wanted, you would never chose that combination because it looks too neat.

Post edited at 17:59
OP Phil Lyon 02 Dec 2020
In reply to Toerag:

What about 3,19,22,29,37

thats a bit more random isn’t it. 

 FactorXXX 02 Dec 2020
In reply to bigbobbyking:

> If you did 'fix' the lottery to produce the numbers you wanted, you would never chose that combination because it looks too neat.

It would be interesting to know if anyone put a sizeable bet on that particular combination occurring on that particular evening.

 Hooo 02 Dec 2020
In reply to Phil Lyon:

The gist of the story being that your average lottery punter doesn't understand basic probability. No surprise there, as if they did they wouldn't be playing!

 Philip 02 Dec 2020
In reply to Phil Lyon:

> If this is a suspiciously unlikely permutation, what would be a more likely one?

The suspicious nature isn't the simple fact it's sequential. If the balls were fed in sequentially then to extract them with the same sequence looks suspicious. As others say, if you had control of the outcome you would choose something more disordered. In fact you would only have to resort to this kind of fraud if you couldn't control ball selection once mixed (using weights for example) and instead did it through timing by very carefully controlling the movement within the machine.

The real test for fraud would be is there a shift in the number of tickets containing those numbers compared to previous weeks, indicating someone aiming to benefit.

 nniff 02 Dec 2020
In reply to Phil Lyon:

There was a lecturer in probability who would ask his students to flip a coin 100 times and record the results.  He said that they were welcome to cheat and falsify the results if they wished, but he then bet them a small sum that he would be able to spot the ones that cheated.   All he looked for were the results that looked improbable (such as 10 heads in a row) - a cheat would eliminate those because they thought it shouldn't happen.

Similarly, Benford's Law - description courtesy of Wikipedia: "in many naturally occurring collections of numbers, the leading digit is likely to be small. In sets that obey the law, the number 1 appears as the leading significant digit about 30% of the time, while 9 appears as the leading significant digit less than 5% of the time" 

 skog 02 Dec 2020
In reply to Phil Lyon:

> What about 3,19,22,29,37

> thats a bit more random isn’t it. 


Random is a bout the process of generating the numbers, not about the set of numbers themselves.

A random process generating six whole numbers, each from 1 to 37 (or a higher cap) is as likely to produce 5,6,7,8,9,10 as it is to produce 3,19,22,29,37, or any other specific combination.

Every single set of lottery numbers ever drawn was vanishingly unlikely, whether or not it looked like a neat sequence.

(None of this means that the South African lottery is not a scam, of course!)

 Mr Lopez 02 Dec 2020
In reply to Phil Lyon:

I would hate being the guy who played 1, 2, 3, 4, 5, 6. Someone out there is having a really bad day right now...

 ripper 02 Dec 2020
In reply to Mr Lopez:

> I would hate being the guy who played 1, 2, 3, 4, 5, 6. Someone out there is having a really bad day right now...

I remember Derren Brown on TV a few years ago saying hundreds of people choose these very numbers every week, thinking they're being unique - so if you won with 123456 you'd be sharing your jackpot with a lot of other players!

Post edited at 20:25
 mondite 02 Dec 2020
In reply to Phil Lyon:

Depending how it was designed if I was in charge I would probably want a review in case it had been screwed up somehow (eg if they are dropped in in order then not being mixed properly) but I would tend to expect the answer to be pretty quick to be "chance".

Not sure how useful looking at people buying tickets would be to identify if it was a fix. Since you get a shared proportion no point in buying ten tickets with the same number. If I was going for a fix I would want to get the past year or so worth of choices and then find which numbers are chosen least by players(I would think 13 would be a candidate for example but then people might chose it on those grounds).

Post edited at 20:51
 Ridge 02 Dec 2020
In reply to Hooo:

> The gist of the story being that your average lottery punter doesn't understand basic probability. 

Most people don't understand probability.

The number of times I hear "But that's a one in a hundred year event, and the plants only going to operate for another five years, so its incredible that that could happen" from supposedly intelligent people is frightening.

Note to self: Next time turn off web cam and mute microphone before saying "Oh, for f***s sake" out loud.

 wilkie14c 03 Dec 2020
In reply to Mr Lopez:

Using Katie Price as a comparison is t really fair as given her track record, it’s only a matter of time before one of us has to marry her

 petemeads 03 Dec 2020
In reply to Phil Lyon:

I think the scam/fraud callers would have a much stronger case if the numbers came out in order...

Removed User 03 Dec 2020
In reply to ripper:

> I remember Derren Brown on TV a few years ago saying hundreds of people choose these very numbers every week, thinking they're being unique - so if you won with 123456 you'd be sharing your jackpot with a lot of other players!

Yep.

My only trick with doing the lottery is to always pick at least two consecutive numbers. Most people pick a set of numbers spread randomly across the array because they instinctively feel that a random pattern has more chance of winning. If I choose a less popular pattern then my chances of winning stay the same but my chances of being the only winner increase.

mick taylor 03 Dec 2020
In reply to Hooo:

Spot on. I once decided that the only way you could do anything was to choose numbers which are least likely to be picked, thus reducing the chances of having to share the jackpot. So I ruled out the usuals and most numbers below 30 (because ‘dates’ are common choices) and went for 30,35,49,45, 49 and 31. I then realised that if they did come up I would most likely be sharing the jackpot with a few hundred smart arse pseudo mathematicians like me. 

 Oceanrower 03 Dec 2020
In reply to mick taylor:

Not that smart arse. You've got 49 twice...

Post edited at 14:29
 LastBoyScout 03 Dec 2020
In reply to Hooo:

> The gist of the story being that your average lottery punter doesn't understand basic probability. No surprise there, as if they did they wouldn't be playing!

Hasn't stopped people winning some incredible amounts, though!

mick taylor 03 Dec 2020
In reply to Oceanrower:

Bugger - everyone will be doing it now. 

 Hooo 05 Dec 2020
In reply to LastBoyScout:

I think you've just proved my point.

 Wimlands 05 Dec 2020
In reply to Ridge:

Mrs Wilderbeest and I check the Met office site before going out and if it says 10% chance of rain I don’t take a Water proof jacket with me...

the number of times I’ve stood there in the rain saying “well what are the chances of that”


New Topic
This topic has been archived, and won't accept reply postings.
Loading Notifications...