Can the National Lottery be fixed???

[lost thought]

The odds of winning the lottery are overwhelming, infact the lottery sucks more balls than guinevere on a rollover weekend. They always seem to donate the charity money they collect to some obscure pointless charity like "plaster casts for hedgehogs" or "save the privets foundation" its just bad news all around.

I believe the charity Tiggywinkles would like some "plaster casts for hedghogs" not sure about the care and rehabilation of privit hedges though.
http://www.sttiggywinkles.org.uk/
The World's Busiest Wildlife Hospital
Caring for sick and injured hedgehogs, badgers, wild birds, foxes, even reptiles and amphibians

[Matt]

Absolutely! I have for many years believed something regarding the way the Lottery is drawn is in some way manipulated.
Here are my questions.
1. Why does the machines close so long before the draw takes place, you can place a bet right up to the last minute with a Horse/Dog race?

If I remmeber correctly it's half an hour before the draw that the terminals stop issuing tickets. This may have somethign to do the the cooling off period where you can return national lottery tickets for a full refund within, it think, it's twenty minutes. Something I'm sure is hardly ever used but which was incorporated into the charter for the national lottery to appease our self appointed moral guardians who were fearful that our nation would be swept up by an irrational lottery fever in the wake of this huge state sponsored payout.

A second reason is that all transaction must complete processing before the draw. Camelot's lottery terminals form the largest private network in the UK, capable of processing 400,000 trancasction per minute but at launch tickets were being sold at a far greater rate peaking just before the draw. Caching the transactions and processing them after the draw would introduce the possibility for fraud should a terminal operator attempt to reverse engineer access to the system.

2.Back around 1992 117 people won the lottery getting all 6 numbers, did something go wrong on that day? I mean would so many people do it if you were expected to share 3 million with over a hundred people?

Well the first ever draw wasn't until 1994 so that date is certainly wrong. Occassionally the Jackpot is shared between a large number of people. This can be for three reasons.

The most obvious is syndicates. Without further details to go on those 117 people may represenst just one winning ticket and be nothing extraordinary at all.

The second is that you'd expect to see some level of variation in the number of winning tickets. These days an average of 32.5 million tickets are sold for a regular saturday draw. Less for a wednesday, more for a rollover. Much much more when the lottery was in it's first flush of youth. The more people who play the more jackpot winners you'd expect. The more draws there have been the more chance there is of an anomolous result cropping up. Even so, If after ten years of sales at current levels getting more than ten tickets to sharing the jackpot would be anomolous given that everyboyd picks their numbers independantly. As such this second can only be a contributory factor.

Finally is the false assumption above, that everybody makes their pick independantly, in fact there appeasr to be a great deal of grouping. Many people pick patterns on their coupon, many pick birthdays. The result of this is that when popular combinations come up the prize is shared between far more people than you'd otherwise expect, there are also far more rollovers than you'd expect, when less popular combinations are drawn. I don't rememebr your 117 people sharing a jackpot but I do remember two triple rollovers in a row. The introduction of the lucky dip was to help combat this.

3. Out of all the lines placed each week, what are the odds of only 1 person week in week out pick the same set of numbers? I think if your going to do it, you have better chances if you chose a lucky dip, if you happen to have chosen a set of numbers that say 10 or more people have chosen, your never going to win, it wouldn't be good practice to have so many winners now would it!

So, I believe real people win real money, but I also believe that machine is rigged so that it generates 1 or 2 winners!
And yes if your wondering, yes i do the lottery, just in case!!

According to my calculations, on a typical saturday draw with 32.5 million tickets sold, if the picks on those tickets were all allocated randomly:
1% of the possible combinations would each appear on 7 or more tickets,
2% of the possible combinations would each appear on 6 tickets
6% of the possible combinations would each appear on 5 tickets
12% of the possible combinations would each appear on 4 tickets
20% of the possible combinations would each appear on 3 tickets
26% of the possible combinations would each appear on 2 tickets
23% of the possible combinations would each appear on 1 tickets only
and
10% of the possible combinations would not have been picked by anybody

Of course only the lucky dips are allocated randomly. The rest are picked by people choosing numbers with some siginificance to them. As discussed earlier this should make no difference to their chance of winning but is likely to decrease their payout should they win, as others are likely to be picking the same numbers for similar reasons.

I agreee that such grouping is a problem for the organisers. They're selling hope. The hope of life changing amounts of money. Jackpots shared between large numbers of people devalue their product. The average Jackpot share is £2.3 million. Just enough to buy a nice house and an annuity. Give up work and live comfortably for the rest of your life. Occassionaly however that share is much lower.

Take for example the draw for Saturday 11th October 2008. The numbers drawn were 20, 21, 23, 24, 27, 28. Three pairs of consecutive numbers closely grouped. I can easily imagine that the people picking this combination thought to themselves, quite rightly, that they were just as likely to come up, as any other. Where they went wrong was in thinking that everybody else would think that consecutive pairs would be unlikley to come up and that they would therefore be less likely to share the jackpot if such a richly patterned sequence should be drawn. In actual fact there were 11 jackpot winners for that rollover jackpot, each netting just £713,924

About 35 million tickets were sold that week only 11 people had the same idea but that was enough to reduce the payout significantly.

This exception proves that the machine isn't rigged to avoid such occurences, or if it is rigged, it isn't rigged very well.

[Pebble]

Matt, I think it is even more random than that.

If 32.5 million tickets are sold per Saturday, on average 2 winning tickets should be sold (1:14m chance), if even 1% of tickets are within 1 number of also being winning combinations, then over 100 weeks, the confidence interval around that number of 2 is nearly 19. So anywhere between 0 and 21 winners may occur.

[Matt]

Matt, I think it is even more random than that.

If 32.5 million tickets are sold per Saturday, on average 2 winning tickets should be sold (1:14m chance), if even 1% of tickets are within 1 number of also being winning combinations, then over 100 weeks, the confidence interval around that number of 2 is nearly 19. So anywhere between 0 and 21 winners may occur.

I must confess I'm unfaliar with the technique you've used to arrive at this conclusion. I'm using the BINOMDIST function in excel. With such extreme numbers it appears that rounding errors predominate. It's giving me the cummulative probability that there'll be 16 or less winning tickets in any one week as 1.000000001

=BINOMDIST(16,32500000,1/13983816,1)

I'm getting the same error when looking at 21 or less winning tickets in any one week.

Lower down the scale I've no reason to believe excel has crapped out.

It's giving the cumulative probability that ten or less tickets are winning in any one week as 0.99996764. Raised to the power 100 this gives the probability of 100 draws in a row of ten or less winninng tickets as 0.996769192 or 99.7% As such I can be pretty confident that 21 winning tickets would be extraordinarily unlikley.

Of course this is all assuming that the picks were allocated randomly. As we've allready discussed this is not that case and certain combinations are much more highly favoured than others. This vastly increases the possibility of these large group wins. This is evidenced by the fact that we only have to go back a couple of months to find a jackpot shared by 11 people.

[Pebble]

Thanks Matt,

Have used BINOMDIST, was unaware of this function, and results look very reasonable, however my version of Excel seems to yield slightly different results. Also will not handle numbers of trials in the 100s of millions.

My mistake was that in order to use confidence interval caclulations I had to make an assumption about the number of tickets within one number of being correct - I obviously over estimated.

Whether BINOMDIST is accurate at these levels I am also unclear on, but it yields less than a 10% chance of having no jackpot which seems about right.

[lost thought]

I was unable to get a ticket for sat as all the machines where down in my area, I could have been a winner. Poo

[FarSideOfTheMoon]

I was unable to get a ticket for sat as all the machines where down in my area, I could have been a winner. Poo

Yeah, but you would have won a million less than usual. So it's probably a good thing you didn't win.

[lost thought]

Yeah, but you would have won a million less than usual. So it's probably a good thing you didn't win.

A million or three I could learn to live with a measly 1 Million Quid... Boo Hoo

[davym]

Last saturday my partner bought her usal tkt for the lottery a fixed line of numbers and a lucky dip on the dream number.
She got the first number on the dream number, so she should win £2.

This saturday today 28/3/09 she takes the tkt into the same shop to get her £2 and redo the lottery.

The shop assistant takes the tkt to check and says sorry no winner because its for tonights draw, the tkt was printed a week ahead okay so its valid for tonights draw.

On getting home she reads the tkt again !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

In the bottom third of the ticket where the adverts are placeed it says.


Lotto rollover this wednesday estimated £8.5 million jackpot.


?????????????? How did it know wednesday was going to be a rollover on the saturday before??????????

This is a genuine post.

Any ideas?

[Trinoc]

Any chance you could scan the ticket and post the image?

[skbuncks]

Last saturday my partner bought her usal tkt for the lottery a fixed line of numbers and a lucky dip on the dream number.
She got the first number on the dream number, so she should win £2.

This saturday today 28/3/09 she takes the tkt into the same shop to get her £2 and redo the lottery.

The shop assistant takes the tkt to check and says sorry no winner because its for tonights draw, the tkt was printed a week ahead okay so its valid for tonights draw.

On getting home she reads the tkt again !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

In the bottom third of the ticket where the adverts are placeed it says.


Lotto rollover this wednesday estimated £8.5 million jackpot.


?????????????? How did it know wednesday was going to be a rollover on the saturday before??????????

This is a genuine post.

Any ideas?

Because the wednesday lottery is separate from the saturday lottery. Wednesdays roll over to wednesdays and saturdays roll over to saturday.

i.e. she buys ticket on saturday 21st. The Advert at the bottom is relating to the draw on wed 25th, there were no winners on wed 18th so it rolled over to the 5th. No mystery here

skb

[skbuncks]

Because the wednesday lottery is separate from the saturday lottery. Wednesdays roll over to wednesdays and saturdays roll over to saturday.

i.e. she buys ticket on saturday 21st. The Advert at the bottom is relating to the draw on wed 25th, there were no winners on wed 18th so it rolled over to the 25th. No mystery here

skb

Added a 2 (in bolding)

ETA: http://www.ukskeptics.com/forum/show...66&postcount=8

[NelMac]

Since the start of the lottery I have often wondered why the delay exists. The fact that there IS a delay points to the fact there is some mathematical computation going on. The great bit is this, as the power of cpu's has increased over the years I've even noticed a slight reduction in the delay. Also, on evenings of high ticket sales the delay is WITHOUT DOUBT, LONGER. The fact the program is aired alongside that crap they put with it, singers/performers competitions, all introduce VARIABLES into the equation that allows them to chose WHEN to hold the draw. Out the 30 to 60 million tickets bought with 1:49*6^6 combinations per line per ticket, it would be reasonable to expect an array of 64bit cpu's to take 15-20 minutes to compute a selection that was already PRE-DEFINED. Now, here's the bit that gets me. Well, if you had a 'cameloot' representative that toured the country on a weekly basis, visiting retailers with terminals at random, then, armed with the retailer number and time of ticket purchase it would be EASY to go back to HQ, fire up the system and look at the numbers selected by a certain individual. This is how most of the winners tend to be about 50 years old or over. These can be compared to see what other selections have won and if it is a 'viable' selection. By doing this, they ensure the cash isn't getting into the wrong hands. The chances are, when anyone they wouldn't like winning wins, it was actually GENUINE luck. To get the machines to select the balls is the EASY bit!! Get a SYRINGE and inject the selection with any type of liquid!!! If you used mercury though, you would definetly increase the probability to around 80-100% !!! On the days where the pots are split these are also taken into account, e.g, they MUST have SOME general public winning SOME amount of money in order to keep the incentive seem 'real'. UNDOUBTEDLY a rip off.

[Trinoc]

I don't doubt it could be fixed if there was a good enough reason, but what would be the point? The amount of money given out in prizes is fixed and closely monitored, so the only point would be to divert winnings to particular people. I don't think it would take long for someone to figure out if prizes were going disproportionately to people close to Camelot.

The only areas where I can see a fiddle might help Camelot would be to make the lottery seem more appealing, e.g. by choosing numbers which give higher individual jackpots, or which result in more roll-overs.

You'd still have to explain how a machine using balls being mixed up in full view of the public can be made to come up with a desired result, though.

[Matt]

Since the start of the lottery I have often wondered why the delay exists. The fact that there IS a delay points to the fact there is some mathematical computation going on. The great bit is this, as the power of cpu's has increased over the years I've even noticed a slight reduction in the delay. Also, on evenings of high ticket sales the delay is WITHOUT DOUBT, LONGER. The fact the program is aired alongside that crap they put with it, singers/performers competitions, all introduce VARIABLES into the equation that allows them to chose WHEN to hold the draw. Out the 30 to 60 million tickets bought with 1:49*6^6 combinations per line per ticket, it would be reasonable to expect an array of 64bit cpu's to take 15-20 minutes to compute a selection that was already PRE-DEFINED. Now, here's the bit that gets me. Well, if you had a 'cameloot' representative that toured the country on a weekly basis, visiting retailers with terminals at random, then, armed with the retailer number and time of ticket purchase it would be EASY to go back to HQ, fire up the system and look at the numbers selected by a certain individual. This is how most of the winners tend to be about 50 years old or over. These can be compared to see what other selections have won and if it is a 'viable' selection. By doing this, they ensure the cash isn't getting into the wrong hands. The chances are, when anyone they wouldn't like winning wins, it was actually GENUINE luck. To get the machines to select the balls is the EASY bit!! Get a SYRINGE and inject the selection with any type of liquid!!! If you used mercury though, you would definetly increase the probability to around 80-100% !!! On the days where the pots are split these are also taken into account, e.g, they MUST have SOME general public winning SOME amount of money in order to keep the incentive seem 'real'. UNDOUBTEDLY a rip off.

Not 1:49*6^6, that would be 1 in 2,286,144

Actually 1:49!/(6!*(49-6)!) whihc is 1 is 13,983,816

If the balls are tampered with then the independant inspector from Oflot and the meber of the public are shills in on it. Wider conspiracies have less credibility.

The delay will have more to do with netwropking speeds than processing speeds and though faster networking speeds are now just as available as faster CPU's upgrading a nationwide fleet of terminals is a far greater task that upgrading the centralised processing system.

Have you any numbers of age distribution of lottery players vs age distribution of lottery winners to back up you claims. If most lottery winners are over 50 then it's the first I've heard of it.

You'll note that the most infamous case of a lottery winner being "The Wrong Type of person," King of the Chavs - Michael Carrol, there were no other winners for this rollover Jackpot. This is clearly not explained by him being lucky enough to have picked the same numbers as a selected winner.

If the lottery is being manipulated for these purposes then it's not being done very well.