• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Lotto: Statistics question

No, not "another" deal.

Deal with the questions people ask, and stop vilifying them based on who they are.

To make it crystal clear: Stop vilifying people, period.

Think you can do that?

Deal?
Who asks the question is important,

If Kumar asks why water can’t water have a memory. Should I treat that question the same as if Rolfe asks it ?

Perhaps you have a good reason for going native, I am just disappointed.
 
You're just plain wrong.

Here's another example.

Do the math!

Let's say the random # gen the customer uses to pick has a glitch in which it never chooses the number 1-18 from 1-36. It never picks half the numbers. The random # gen that picks the winning number does NOT have this same glitch.

Yes. Still, the number generator with the glitch will produce a valid ticket with 6 different numbers.

A person using the glitched random # gen has now decreased their odds of winning by 50% versus someone who didn't use the glitched random # gen.

No!

Suppose we both played the lottery. I use the generator with the glitch, you use a flawless generator. Let us further assume we're both playing tow lines!

Rasmus-A: 23 - 34 - 36 - 33 - 30 - 31
Rasmus-B: 32 - 24 - 25 - 26 - 19 - 27

OnlyTellsTruths-A: 29 - 20 - 8 - 2 - 24 - 28
OnlyTellsTruths-B: 36 - 30 - 34 - 25 - 20 - 21

Obviously the odds of the winning number containing the numbers 1-18 is 50%,

That is far from obvious to me:

The odds for every single number to be <19 are 50%. So the chances that at least one of the 6 numbers is <19 is very large. (Something like 0.5^6? against?)

and the person who used the glitched random # gen cannnot have those numbers.

Please look at the numbers that I imagine we would play. (Note that one of your lines doesn't contain numbers <19, which is unlikely but possible!)

What chances of winning do yoiu think do these individual lines have?

Why would any of these lines be more likely to come up in the actual draw - which we assume is fair and truly random?
 
Choosing a particular number sequence from a set= always the same odds
Choosing a particular number sequence from a set that has been removed of any number of allowed sequences = odds change versus others who aren't likewise limited


Who needs to re-read what?
 
You're just plain wrong.

Here's another example.

Let's say the random # gen the customer uses to pick has a glitch in which it never chooses the number 1-18 from 1-36. It never picks half the numbers. The random # gen that picks the winning number does NOT have this same glitch.

A person using the glitched random # gen has now decreased their odds of winning by 50% versus someone who didn't use the glitched random # gen. Obviously the odds of the winning number containing the numbers 1-18 is 50%, and the person who used the glitched random # gen cannnot have those numbers.

Perhaps you are looking at it from the opposite pov (where the machine that picks the winning number is glitched and not the one the customer uses).
I'll do the math for you.

A) The winning number has a 1/36 chance to be any integer from 1 to 36.

B) The customer's number has a 1/18 chance to be any integer from 19-36.

C) P(winning number in 1-18) = 0.5 and P(winning number in 19-36) = 0.5

D) P(Win! | winning number in 1-18) = 0

E) P(Win! | winning number in 19-36) = P(winning number is 19)*P(customer's number is 19) + P(winning number is 20)*P(customer's number is 20) + ... = 1/36*1/18 + 1/36*1/18 + ... eighteen times = 18*(1/36*1/18) = 1/36.

Now suppose another customer picks entirely at random. We'll call him smart.

F) The smart's number has a 1/36 chance to be any integer from 1 to 36.

G) P(Smart Win!) = P(winning number is 1)*P(smart's number is 1) + P(winning number is 2)*P(smart's number is 2) + ... = 1/36*1/36 + 1/36*1/36 + ... thirty six times = 36*(1/36*1/36) = 1/36

So the customer and the smart both win with probability 1/36.

You can now believe we're correct and possibly study some probability to find out what part of your beliefs are wrong or point to some error I made or believe you're correct despite the evidence. They're all fine choices. :D
 
Choosing a particular number sequence from a set= always the same odds
Choosing a particular number sequence from a set that has been removed of any number of allowed sequences = odds change versus others who aren't likewise limited


Who needs to re-read what?

If I try to pick the outcome of a flipped coin by always predicting heads, will the odds of me being correct be less than someone who uses a random number generator to pick between heads and tails?
 
Last edited:
Who asks the question is important,

If Kumar asks why water can’t water have a memory. Should I treat that question the same as if Rolfe asks it ?

Yes.

That's exactly what you should do. Because how can it be important who asks the question? Should we dismiss questions, based on who asks them?

If Sylvia Browne asks a question about black holes, should we dismiss her question out of hand, merely because she is the lowest form of life that could ever be imagined?

Who the hell put you in charge of who can ask what questions?

Perhaps you have a good reason for going native, I am just disappointed.

I have no idea what you mean by "going native". Please explain.
 
Yes. Any individual game has the same chances of winning - no matter how the numbers were arrived at or what they look like.

Agreed.

Why do you think this is so?

I am not even sure that the biased generator would win less often than an un-biased one. But in the real world, the biased generator plays agaisnt highly based real people. (They are the ones playing birthdays, avoiding the 13 and numbers on the border of the playing field, etc.)

Also, you would need to know the relation of generated numbers to manually picked numbers.

The biased generator doesn't have as large a field to pick from is why I think so.

Assume the lotto's winning numbers are a uniform distribution over all possible winning combinations. The biased generator is a uniform distribution over a SUBSET of the possible winning combinations. Therefore, the odds that a winning number came from the biased generator are lower than the odds that the winning number came from all possible combinations - that probability is 1 :)
 
Rasmus said:
That is far from obvious to me:

The odds for every single number to be <19 are 50%. So the chances that at least one of the 6 numbers is <19 is very large. (Something like 0.5^6? against?)

Yes, in the original example where 7 numbers are picked, but if just 1 number is picked the odds are clearly 50%.

Say you have 3 people and 3 dice (normal 6 sided). The second two people try to match what the first person rolls, they each get 10 tries. Say you are one of these second people and your dice is broken (gum on it) so that it always displays 3.
The other guy gets 10 rolls to try and match, but you don't.

Perhaps everyone is forgetting about multiple tickets being bought?
 
Last edited:
Yes, in the original example where 7 numbers are picked, but if just 1 number is picked the odds are clearly 50%.

Say you are have 3 people and 3 dice (normal 6 sided). The second two people try to match what the first person rolls, they each get 10 tries. Say you are one of these second people and your dice is broken (gum on it) so that it always displays 3.
The other guy gets 10 rolls to try and match, but you don't.

Perhaps everyone is forgetting about multiple tickets being bought?

Multiple tickets being bought has nothing to do with it. The odds of a given combination matching the lotto's winning combination does not change based on how you get the given combination, as most have pointed out.
 
But that is not the lotto. I thought we were discussing whether or not the odds change based on how the number is selected.

The odds that the second player will match more often is greater than the player with the dice that only rolls a 3. I tried to explain this in post 127, second paragraph.
 
Well if it's my fault that I'm answering a seperate question let me be the first to apologize for when I said others arguments were wrong (they were for a seperate question), happens all the time.
 
Last edited:
Say you have 3 people and 3 dice (normal 6 sided). The second two people try to match what the first person rolls, they each get 10 tries. Say you are one of these second people and your dice is broken (gum on it) so that it always displays 3.
The other guy gets 10 rolls to try and match, but you don't.

OK, I think I see what you're saying. You're saying that if I buy 10 biased random lottery tickets, and someone else buys 10 non-biased random lottery tickets, they have a much better chance of winning because their tickets will cover more of the possible numbers than mine will.

I admit, I can't answer this one. Anyone who's better at statistics than I am?
 
OnlyTellsthetruth - i'm sorry, you're simply wrong. Changing to your dice example is unhelpful because you have not specified what constitutes a "match," whether order is important, etc etc.

let's keep to the lotto example, but a much more simplified case. There are 5 balls. 1,2,3,4,5. The machine will draw 2 at random. If they match the two balls you've chosen, you win.

You have (5 choose 2) = 10 different combinations possible.

1 2
1 3
1 4
1 5
2 3
2 4
2 5
3 4
3 5
4 5

Therefore you have a 1/10 chance of winning.

Now in order to choose which numbers to pick,

player 1 uses a glitched generator which always picks 1 and 2

player 2 use a non glitched generator which can pick any 10 with equal likelihood

So

On the day of the big draw,

player 1 has 1,2
player 2 has (say) 4,5

When the draw is made they both have 1/10 chance of winning.

If the draw was made again

player 1 having 1,2
player 2 having (say) 2,5

they would both still have 1/10 chance of winning.

No mattter how many times the draw was made, both player 1 and player 2 would have the same odds of winning the lotto.
 
Last edited:
To truly "do the math" on this, one would have to take into account not only a certain group buying multiple tickets from the biased machines, but also them doing so on multiple drawings/days.

The control group (that doesn't use biased machines) would have increased odds of winning overall.
 
Last edited:
Yes.

That's exactly what you should do. Because how can it be important who asks the question? Should we dismiss questions, based on who asks them?
I am not dismissing the question. I am questioning why you are asking it.

If Sylvia Browne asks a question about black holes, should we dismiss her question out of hand, merely because she is the lowest form of life that could ever be imagined?
We are not talking about Sylvia Browne asking questions about black holes. I am questioning why Phil Platt is asking if Planet X will hit us.

Who the hell put you in charge of who can ask what questions?
the hell put you in charge of what question I can ask? :)
I have no idea what you mean by "going native". Please explain.
Ok let me give an example. See your quote about Sylvia Browne above. If next week you started asking lots of questions along the lines of; what is wrong with Sylvia Browne? Why can’t we speak to the other side? What reasons are there for calling her the lowest form of life ? Would I be weird to question why you have changed your tack.

Same here. I can’t understand why you had a good understanding of maths and now it has deserted you. Are you playing devils advocate or has my memory tricked me ?
 

Back
Top Bottom