Solved: UK Random Lottery Challenge!

[mdmackillop]

Here are the results of the 915 UK lottery draws. Can anyone come up with a "random draw" programme to beat the odds?
MD:creator:

[Jacob Hilderbrand]

I actually made a lotto program. I can send it once I find it. You can't beat the odds. No matter how long the program runs, the percentages of correct picks stay the same. 99% (0-3 correct) iirc.

Doen't matter what numbers you pick, doesn't matter if your picks change or stay the same.

The problem is that you can have a list of a billion previous winners, but that has absolutely no effect on the billion + 1 winner. None at all.

[johnske]

.... You can't beat the odds.....

Doen't matter what numbers you pick, doesn't matter if your picks change or stay the same.

The problem is that you can have a list of a billion previous winners, but that has absolutely no effect on the billion + 1 winner. None at all.Perfectly correct... Have a look at some books on probability MD (the one I have is "A First Course In Probability" By Sheldon Ross (Macmillan Publishing Company) but there are plenty of other good books around...

There is an interesting well-known side problem called "The Gamblers Ruin" problem that all gamblers should be aware of - essentially this says that if two people sit down to gamble continually until one is broke which gambler will win?

Again (paraphrasing) the gambler that almost invariably wins is the one that had the most money to start with - even if 'the most money' was only one dollar more!... :bink:

Edit: PS this is just an extension of the old adages that apply in business, finance, and work....."The rich get richer & the poor get poorer" - and - "Money makes money"

[brettdj]

The problem is that you can have a list of a billion previous winners, but that has absolutely no effect on the billion + 1 winner. None at all.
Needs a small qualifier :)

If the draw is unbiased and independent from previous draws then you can't beat the odds.

[Steiner]

brettdj: let's hope that it is, otherwise I'd like to get my hands on the modification algorithm used...

[mdmackillop]

Oh well,
I'll stick with my other strategy. Buy 5 tickets with the same number, so if the prize is split I get a bigger share.:yes

[Jacob Hilderbrand]

I like going to casinos and watching people play roulette. "Oh there have been 5 blacks in a row, the next one will be red".

If I flip a coin and get 100 heads in a row, the odds of the next toss being heads is still 50/50.

[Jacob Hilderbrand]

Here is the lotto program I made. Losing percentage is around 99.95%.

[brettdj]

I remember a story a couple of years ago about the Mafia corrupting Lotto games, the lotto balls were drawn out by small children with blindfolds.

But the children could see through the blindfold.

[onlyadrafter]

Hello,

as stated the next draw has nothing to do with the previous ones but, if you had selected (from day 1)

7, 35, 40, 43, 44 and 47 = 3 no's 40 times and 4 no's 1 time.
or
4, 16, 25, 35, 43 and 46 = 3 no's 17 times and 4 no's 9 times
or
7, 10, 12, 34, 46 and 48 = 3 no's 14 times, 4 no's 2 times and 5 nos 3 times
or
15, 16, 18, 28, 36 and 59 = 3 no's 15 times, 4 no's 1 time, 5 no's 0 times and 5 and the bonus 2 times.

Should anybody use these numbers, i do expect a small tip.

This weeks numbers will be :dunno

[mdmackillop]

Use or Us?

MD

[janosik90]

I like going to casinos and watching people play roulette. "Oh there have been 5 blacks in a row, the next one will be red".

If I flip a coin and get 100 heads in a row, the odds of the next toss being heads is still 50/50.

Correct, but that does not mean that the probability of 6 blacks in a row is still 50%, right?
If you want to know more, look for Law of Disproportionate Occurences

[Aaron Blood]

I once developed a tool that could actually predict winning lotto numbers!

The only downside was it required a 1-week lag... :mkay

[Ken Puls]

What? And the lotto corporation wouldn't give you that, Aaron? LOL!


Correct, but that does not mean that the probability of 6 blacks in a row is still 50%, right?

I don't remember a lot from my stats classes, but isn't the probability of 6 blacks in a row 1.5625%? (.5^6)

Thinking about it, those would be the odds of any specific 6 colour combination...
-5 blacks then 1 red
-2 blacks, 2 reds, 2 blacks
etc...

[brettdj]

nope :(

no such thing as an even money bet - outside a secondary craps bet - in a casino

roulette wheels have at least one zero (green) outcome, and I believe some wheels have two

so the odds are (18/37)^6 or (18/38)^6

Cheers

Dave

[Ken Puls]

Ah! Didn't know about the green, as I don't go to the casino much. :)

Odds of 6 heads in a row, providing the coin is balanced, it only has two sides, and a different thing on the other side ;) then would be 1.5625%.

[TonyJollans]

Leaving aside the 0 (and 00) the odds of 6 blacks in a row when you have already had 5 blacks in a row are 50:50.

[janosik90]

Leaving aside the 0 (and 00) the odds of 6 blacks in a row when you have already had 5 blacks in a row are 50:50.

The answer is correct.
However, the question is not, that is the point :)

Do the casinos accept the bet like "There will be 6 blacks in a row" ? No.
But does it mean you cannot simulate such a bet, if the situation occurs? No.
You can, if it happens. That?s what the situational betting is about.

[Jacob Hilderbrand]

What happens is, the casino show the last 20 or so numbers that were selected in roulette on a board. So you think there is a pattern, when in fact, each winning number is completely independant from the last.

So the odds of 6 black in a row are slim, but still it was 50/50 for it to be red or black (not counting the two greens).

But just remember, when gambling, the bigger the hook, the more risk. There are some games where you can give youself better odds by playing well (like Blackjack), but others that are just random like Keno or Lotto.

I am still trying to win the Lotto without buying the ticket. I figure I have about the same odds of finding a winning ticket as buying one. About 0.00%.

[bg60]

To all,
Analysis has shown that if you know how to play sensibly with Black Jack, and understand the rules of the game, it is the only game at the Casino where you have as close as possible to a 50/50 ratio, but note that the odds stacked in the Casinos favour. The Casino will utlimately win in the long run, because it makes its money from the large volume of small bets placed.

Remember that in a game of chance, each event is mutually exclusive to any historical (previous) event. In a Casino enviroment, the bet is placed prior to the start of each event, and the bets are not made on a series of events or outcomes. This works in the Casinos favour, as there is no predictability in a game of chance, especially when the Casion has the "bank" and odds on their side.

I challenge anyone to write a program to predict the outcome of the next winning hand, winning lottery results or any other winning results...if you could you would not be here :) Even if its a bet on tossing a coin. I bet you can't !

[Bartek]

Hi,


Remember that in a game of chance, each event is mutually
exclusive to any historical (previous) event.

This is of course incorrect when it comes to card games like blackjack or many games based on mechanic devices.


I challenge anyone to write a program to predict the outcome of the next winning hand, winning lottery results or any other winning results...if you could you would not be here :) Even if its a bet on tossing a coin. I bet you can't !

In this discussion there is some confusion regarding beating the odds. In games of change, not the probability of winning is important, but expected value, i.e., probability of winning times the prize value minus the cost a bet.

Classic 6/49 lotto is in fact a special parimutuel bet, when you are betting against the other players, not the lotto company. Expected value is affected not only by pure probability calculation, but also by the distribution of bets, i.e., combinations chosen by all players. Players do not pick numbers according to uniform distribution, and it creates a serious bias. As a result, in case of a large lotteries like UK National Lottery, there are even bets with positive expected value.

According to your "challenge" - there are programs that constantly beat the odds for horse racing. These are powerful econometric tools used by so called "betting syndicates" - some people earn millions of dollars gambling this way. But of course, these tools require several yers to develop and are not publicly available:

http://www.wired.com/wired/archive/10.03/betting_pr.html