Lottery Calculating Trouble

[Raider5678]

If they come round asking for your money then tell them to take a hike. You won it fair and square according to the terms and conditions on the tickets right?

Technically.

But it still scares the shit out of you.

[Lord Antares]

It shouldn't. They are not mafia.

If you bet on sports event and win, they are required to pay up, regardless of whether they might think you lucked out or used a system of sorts. All chances of using a system is their fault, as I said. If they were smart, their numbers would be random.

 

That's it. Don't be concerned over this.

[Phi for All]

Yeah well when two police officers and someone from the pa lottery knocks on your door inquiring about how often you win the lottery and saying you may have to pay back the money, your parents don't pat you on the back.

 

Did this happen already?

[Raider5678]

 

Did this happen already?

To me particularly no. I did not come up with this system on my own, I had help.

But the other guy got in legal trouble of some sort, and I figured I should definitely stop before the same happened to me.

So I'm treading careful ground at the moment. I'm not really sure what he's in trouble for, but I know it had to do with the lottery.

[CharonY]

To me particularly no. I did not come up with this system on my own, I had help.

But the other guy got in legal trouble of some sort, and I figured I should definitely stop before the same happened to me.

So I'm treading careful ground at the moment. I'm not really sure what he's in trouble for, but I know it had to do with the lottery.

 

As others have said, exploiting a system cannot be illegal, as it is the lottery's fault. They may have terms but I doubt that they could work something like that into it. The usual limitations are more things like age, for example.

[Raider5678]

 

As others have said, exploiting a system cannot be illegal, as it is the lottery's fault. They may have terms but I doubt that they could work something like that into it. The usual limitations are more things like age, for example.

But if I have my parents buy the tickets it's fine right?

[Prometheus]

Well mine is more on percentages.

If such and such number hasn't been picked for N amount of draws, it's probability to be picked is P.

For example, numbers in the 40s have a 21% chance to be picked 4 draws after the last 4 has been picked. If it isn't picked then, then the probability goes down to 6% chance to be picked. But once you get 11 draws after a 40s number has been picked, the chances of a 40s number being picked is 85%.

The whole system is based off of stuff like this, and it works.

So the lottery isn't random.

 

So basically you pick a number between 01 - 99 and look for a perfect match in the draw?

 

After the draw is the chosen number then put back in the pot for the next draw or is it excluded for some time period?

 

If the former then the draws are independent: any/all proceeding draws do not effect the next draw.

 

If the latter then you are simply exploiting the design within the rules of the game.

 

Either way, the lottery is still random, just different 'kinds' of random.

[Raider5678]

 

So basically you pick a number between 01 - 99 and look for a perfect match in the draw?

 

After the draw is the chosen number then put back in the pot for the next draw or is it excluded for some time period?

 

If the former then the draws are independent: any/all proceeding draws do not effect the next draw.

 

If the latter then you are simply exploiting the design within the rules of the game.

 

Either way, the lottery is still random, just different 'kinds' of random.

Computer generated random I believe.

So it's not perfectly random.

Considering 22 was picked 5 times in the last 100 drawings.

[CharonY]

But if I have my parents buy the tickets it's fine right?

 

Well. if your parents are playing I do not see any issue with that. Also based on your description it does not really seem that you were exploiting anything. At the end of the day you were only guessing right, so I fail to see how that can lead to any legal issues.

[Raider5678]

 

Well. if your parents are playing I do not see any issue with that. Also based on your description it does not really seem that you were exploiting anything. At the end of the day you were only guessing right, so I fail to see how that can lead to any legal issues.

I'll look into it further.

For now I made enough.

[John Cuthber]

Even if it was completely random, I can still find the most probable right?

If 99 hasn't been picked for the last 3000 times, it has a large chance of being picked then 1 if it had been picked 150 times in a row before that.

 

" I can still find the most probable right?"

I doubt it.

"If 99 hasn't been picked for the last 3000 times, it has a large chance of being picked "

No.

How would it "know" that it had been picked a lot?

I rather suspect that the "success" of your "method" relies heavily on confirmation bias.

Don't worry, the lottery companies love to have people like you lipreading stories that they system can be "gamed" in some way.

 

You may take some comfort from the fact that your mistake is so common that it has its own name

https://en.wikipedia.org/wiki/Gambler%27s_fallacy

Computer generated random I believe.

So it's not perfectly random.

Considering 22 was picked 5 times in the last 100 drawings.

How many "balls"- i.e. what's the range of numbers?

[Delta1212]

Computer generated random I believe.

So it's not perfectly random.

Considering 22 was picked 5 times in the last 100 drawings.

How many numbers are there available to be picked, and how many numbers per ticket?

[Raider5678]

" I can still find the most probable right?"

I doubt it.

"If 99 hasn't been picked for the last 3000 times, it has a large chance of being picked "

No.

How would it "know" that it had been picked a lot?

I rather suspect that the "success" of your "method" relies heavily on confirmation bias.

Don't worry, the lottery companies love to have people like you lipreading stories that they system can be "gamed" in some way.

 

You may take some comfort from the fact that your mistake is so common that it has its own name

https://en.wikipedia.org/wiki/Gambler%27s_fallacy

How many "balls"- i.e. what's the range of numbers?

100 different numbers ranging from 00-99.

1 number per ticket.

What I do was buy 25 tickets with what I "calculated"(apparently mistaken in someway according to gamblers fallacy) would be the most likely numbers. Increasing my win rate from what should be 25% to 90.90%(10/11 times I win.)

Maybe I am just really lucky. But if I win far more often then I should be, I'm going to assume I'm doing something right.

And I don't think they can be gamed for too large sums of money. The larger lotteries really are too random to analyze. Analyzing a two digit lottery is easier then analyzing one with millions of outcomes.

[Lord Antares]

(apparently mistaken in someway according to gamblers fallacy)

 

I thought you admitted that you were mistaken in the mathematics of this. I thought you understood that the odds are always the same, no matter what gets picked and when.

 

@John - We are considering the possibility that this a flawed and stupid lottery system, as it is small and local. The only way this would work is if the numbers weren't random, but somehow biased. Then this method would be sound. We agreed upon this; you should read the rest of the thread if you haven't.

[John Cuthber]

If the lottery is rigged the gambler's fallacy usually makes things worse.

Let's say I rig the lottery by simply not having a ball with 11 on it.

Raider will say "look- number 11 hasn't been drawn for ages- it must be more likely to come up soon- so I will bet on it"
He buys tickets with 11 on.

 

I sit here, laugh, + take his money.

 

It's possible to run a crooked lottery without getting caught, but it's not worth the effort. You can make a fortune from an honest one and you don't have to worry about the law catching up with you.

Lots of people will slavishly give you their money in exchange for a dream.

Some dream that they will win (and a vanishingly small number are right).

Some dream that they have "beaten the system" and an even smaller number are right.

[Lord Antares]

No. Not rigged. Biased. That means the numbers are fair but the draw is not random. Only in this case is the system very effective.

He says he has won 10 of 11 of the ones he has played. I suppose he could keep doing it - reducing the suspicious that it was just dumb luck.

[Raider5678]

I don't think it's rigged so much as not truly random.

Besides, betting on numbers that haven't been picked in a long time is a bad strategy because it usually takes 10-15 days before it will pay off.......long after you lost a lot of money.

[John Cuthber]

In what way is the draw biased?

[Raider5678]

In what way is the draw biased?

I believe that the computer system used to pick the numbers is flawed. Getting 100% true randomness from a computer is extremely hard, and I dare say impossible. Every tiny manifestation of a pattern can be exploited.

I'm not sure quite why it's biased, but the distribution of numbers is way to clustered to be truly random. I mean, one to ten coincidences is probably just a coincidence. But when the numbers follow a pattern based off of the said coincidences, I come to believe it's a pattern.

Now maybe I'm flawed, and this lottery actually is random.

In which case, it's safe to say one day my method will simply come crashing down around me.

But at this point I've made a massive surplus, that will take weeks to lose. And I'll stop to reconsider the method long before I lose all the money I made. Besides, i haven't saved 100% of the money, so no matter what I've made a profit already off of what should be a non profitable source.

 

Also, I would like to point out something with the fallacy described above. More of a question.

 

If I have 3 dice. There are 216 possible combinations.

Of all those combinations, there are only 6 that are 3 consecutive numbers.(lets pick 5) 1/216 chance to get 5 three times consecutively.

While, a combination containing 2 of the same number and then a random number (lets pick 3) would have a chance 3 times higher. Because there are 3 combinations that will result in two (2s) and one (3). 2-2-3, 2-3-2, 3-2-2.

Meaning the chances of getting 2-2-3, in any order, is three times higher then the chances to get 3 in a row.

Meaning, 1/72 chance.

[John Cuthber]

It's difficult to get mathematically "good" random numbers.

But Excel is good enough for it to be practically impossible for any human to spot patterns.

https://support.microsoft.com/en-us/help/828795/description-of-the-rand-function-in-excel

"Tiny" deviations from true randomness are not easy to exploit- certainly not at the level of making a regular profit.

 

I think you have fallen victim to the remarkable ability of humans to spot patterns.

 

It's so good, we even spot patterns that are not there

https://en.wikipedia.org/wiki/Martian_canal

 

Do you have a record of the numbers drawn in the lottery?

It's fairly easy to test for bias.

[Raider5678]

It's difficult to get mathematically "good" random numbers.

But Excel is good enough for it to be practically impossible for any human to spot patterns.

https://support.microsoft.com/en-us/help/828795/description-of-the-rand-function-in-excel

"Tiny" deviations from true randomness are not easy to exploit- certainly not at the level of making a regular profit.

 

I think you have fallen victim to the remarkable ability of humans to spot patterns.

 

It's so good, we even spot patterns that are not there

https://en.wikipedia.org/wiki/Martian_canal

 

Do you have a record of the numbers drawn in the lottery?

It's fairly easy to test for bias.

I collected the records for the last 3 years, but have only tested the most recent 93 into trying to see a pattern.

 

https://www.palottery.state.pa.us/Draw-Games/Winning-Numbers-History.aspx

Select "pick 2 day" and hit search.

 

 

If you hit sort by numbers, you can see that in just 93 picks, a large portion of numbers show up 2-5 times.

I think you have fallen victim to the remarkable ability of humans to spot patterns.

And even I believe this is the most likely scenario.

But I didn't immediately invest money to buy tickets.

We simply wrote on a piece of paper what our different methods were predicting, and when a particular method consistently became correct at a much higher rate then the others we built off of that to make a effective one.

It seems to be a pattern. Not a perfect pattern. We still lost. But in over all profit, we could make money.

Edited April 6, 2017 by Raider5678

[John Cuthber]

It's late in the day .
I will try to remember to look into it later.

[Raider5678]

Here's the last 93 numbers in a bulk data set. Frequency of them showing up, total times they started with a particular number, and total times they ended with a particular number.

 

Ignore this. The charts won't load and it won't let me upload screen shots of them.

Edited April 6, 2017 by Raider5678

[Delta1212]

I believe that the computer system used to pick the numbers is flawed. Getting 100% true randomness from a computer is extremely hard, and I dare say impossible. Every tiny manifestation of a pattern can be exploited.

I'm not sure quite why it's biased, but the distribution of numbers is way to clustered to be truly random. I mean, one to ten coincidences is probably just a coincidence. But when the numbers follow a pattern based off of the said coincidences, I come to believe it's a pattern.

Now maybe I'm flawed, and this lottery actually is random.

In which case, it's safe to say one day my method will simply come crashing down around me.

But at this point I've made a massive surplus, that will take weeks to lose. And I'll stop to reconsider the method long before I lose all the money I made. Besides, i haven't saved 100% of the money, so no matter what I've made a profit already off of what should be a non profitable source.

 

Also, I would like to point out something with the fallacy described above. More of a question.

 

If I have 3 dice. There are 216 possible combinations.

Of all those combinations, there are only 6 that are 3 consecutive numbers.(lets pick 5) 1/216 chance to get 5 three times consecutively.

While, a combination containing 2 of the same number and then a random number (lets pick 3) would have a chance 3 times higher. Because there are 3 combinations that will result in two (2s) and one (3). 2-2-3, 2-3-2, 3-2-2.

Meaning the chances of getting 2-2-3, in any order, is three times higher then the chances to get 3 in a row.

Meaning, 1/72 chance.

The odds of rolling 3 then 3 then 3 are exactly the same as the odds of rolling 3 then 3 then 2. So even though rolling two 3s and a 2 in some combination is more likely at the beginning than rolling three 3s, once you start rolling and excluding possibilities, the odds change. If you roll a 2 on your first throw, you don't maintain the original odds of getting three 3s (which drop to 0) and neither do you retain the original odds of rolling three 3s if you at first roll a 3.

 

Once you get to the final draw, the odds of getting the exact pattern of previous numbers that you got is exactly the same for all numbers. You have exactly the same chances of rolling 33333333 as rolling 45826478. Which means that, once you've rolled up to that point, your odds of getting 333333333 and 458264783 are exactly the same.

 

Also, keep in mind the Birthday Paradox, when talking about numbers repeating. Despite there being 365 possible birthdays, you only need 23 people in a room to get a 50% of at least two people in the group sharing a birthday, and the odds increase to over 99% chance of someone charging a birthday at only 70 people.

 

With 100 possible number instead of 365, a series of 22 draws gives you a 92% chance that at least one number will repeat even if there is no bias.

 

There may very well be bias in this lottery, but you still need to learn to distinguish what is and is not an actual sign of bias. We as humans have a tendency to see patterns that aren't there because we improperly weight the importance of certain things when it comes to calculating probabilities and it can make it look like there are patterns or expected results that are not actually present. Something that seems like a very strong indication of a pattern can easily just be noise or even precosely the expected outcome of a random system, so it's important to learn how to distinguish between real signs of a pattern and ghosts cooked up in the wiring of your brain.

 

Otherwise you may stumble across a situation where there really is a pattern and go chasing it down the wrong rabbit hole because you don't know what the actual pattern you're exploiting really is.

[Raider5678]

The odds of rolling 3 then 3 then 3 are exactly the same as the odds of rolling 3 then 3 then 2. So even though rolling two 3s and a 2 in some combination is more likely at the beginning than rolling three 3s, once you start rolling and excluding possibilities, the odds change. If you roll a 2 on your first throw, you don't maintain the original odds of getting three 3s (which drop to 0) and neither do you retain the original odds of rolling three 3s if you at first roll a 3.

 

Once you get to the final draw, the odds of getting the exact pattern of previous numbers that you got is exactly the same for all numbers. You have exactly the same chances of rolling 33333333 as rolling 45826478. Which means that, once you've rolled up to that point, your odds of getting 333333333 and 458264783 are exactly the same.

 

Also, keep in mind the Birthday Paradox, when talking about numbers repeating. Despite there being 365 possible birthdays, you only need 23 people in a room to get a 50% of at least two people in the group sharing a birthday, and the odds increase to over 99% chance of someone charging a birthday at only 70 people.

 

With 100 possible number instead of 365, a series of 22 draws gives you a 92% chance that at least one number will repeat even if there is no bias.

 

There may very well be bias in this lottery, but you still need to learn to distinguish what is and is not an actual sign of bias. We as humans have a tendency to see patterns that aren't there because we improperly weight the importance of certain things when it comes to calculating probabilities and it can make it look like there are patterns or expected results that are not actually present. Something that seems like a very strong indication of a pattern can easily just be noise or even precosely the expected outcome of a random system, so it's important to learn how to distinguish between real signs of a pattern and ghosts cooked up in the wiring of your brain.

 

Otherwise you may stumble across a situation where there really is a pattern and go chasing it down the wrong rabbit hole because you don't know what the actual pattern you're exploiting really is.

I understand the ghost patterns already, that the human brain finds patterns that aren't patterns at all. Multiple people have pointed this out already, but thank you for your input.