I mean, a small chance is much larger than an actually zero chance. I see no harm in it if you're not spending a ton of money, and like most of us, only ever touch this stuff when it gets absurd. It's a couple bucks for a fun fantasy, basically.
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Nobody won Mega Millions so expected jackpot now $970 million
- Thread starter Vanillalite
- Start date
I mean, a small chance is much larger than an actually zero chance. I see no harm in it if you're not spending a ton of money, and like most of us, only ever touch this stuff when it gets absurd. It's a couple bucks for a fun fantasy, basically.
Okay, ken_matthews
I thought of another way to try and help you understand during my commute.
You have been conflating the properties of the integers 1-70 (their order and differences) with the properties of the game.
I have been trying to explain that the properties of these numbers have no bearing on the chances of winning the game. You've come back with math and code that, still, has not made your point, though you believe it has.
Consider this:
The integers from 1-70 are just convenient labels for the balls. Nothing more.
The formula for combinations, nCr = (n!)/(r!(n-r)!), counts how many ways r elements can be selected, without order, from a set of n elements. That's elements, in general. Not necessarily numbers.
So when you compute 70C5, you're thinking of choosing five numbers, yes. But that formula works for counting and choosing anything.
Instead of numbers, let's print each of the 70 balls with its own unique snowflake. We're still choosing 5 of 70. It's the exact same game as Mega Millions, just different labels. Each play has the exact same probability of winning, 1/12103014.
Or, let's print each of the 70 balls with the portraits of 70 different people. Still the same game. Same probability of winning.
Or 70 unique portraits of one person. Same game.
Or pictures of 70 different musical instruments. Same game.
Or depictions of the unique sinusoidal sound wave of each of those instruments. Same game.
Or 70 different colors. Same game.
Or photographs of 70 different high schools from across the USA. Same game.
Or cover art from 70 different comic books. Same game.
Or photographs of the crushed and compacted cubes from 70 different junk cars. Same game.
Or the thumbprints of 70 different people. Same game.
Or the retinal scans of 70 different people. Same game.
Or the genomes of 70 different species. You'd have to print really really small, or have huge balls, but same game.
All that matters is that we can tell 70 balls apart from each other. That's it. How we choose to label them has absolutely no effect on the chances of winning the game. Any labels will do, even if they lack the properties (such as order) of the integers 1--70. The lottery uses numbers out of convenience. It's easier to read and print numbers than to compare sine waves, photographs, snowflakes, etc. But the fact that those numbers have an order on the number line does not affect the odds of winning the game.
The peaks in your graph come from the order of the numbers. They do not indicate anything of the game, or increased odds of winning it. If the balls were printed with 70 unique thumbprints, you wouldn't be able to order them as Ball 1, Ball 2, etc like with the numbers, or discuss whether they are adjacent or spread out, or have your graph with the balls and their peaks.
But it would be the same game.
I thought of another way to try and help you understand during my commute.
You have been conflating the properties of the integers 1-70 (their order and differences) with the properties of the game.
I have been trying to explain that the properties of these numbers have no bearing on the chances of winning the game. You've come back with math and code that, still, has not made your point, though you believe it has.
Consider this:
The integers from 1-70 are just convenient labels for the balls. Nothing more.
The formula for combinations, nCr = (n!)/(r!(n-r)!), counts how many ways r elements can be selected, without order, from a set of n elements. That's elements, in general. Not necessarily numbers.
So when you compute 70C5, you're thinking of choosing five numbers, yes. But that formula works for counting and choosing anything.
Instead of numbers, let's print each of the 70 balls with its own unique snowflake. We're still choosing 5 of 70. It's the exact same game as Mega Millions, just different labels. Each play has the exact same probability of winning, 1/12103014.
Or, let's print each of the 70 balls with the portraits of 70 different people. Still the same game. Same probability of winning.
Or 70 unique portraits of one person. Same game.
Or pictures of 70 different musical instruments. Same game.
Or depictions of the unique sinusoidal sound wave of each of those instruments. Same game.
Or 70 different colors. Same game.
Or photographs of 70 different high schools from across the USA. Same game.
Or cover art from 70 different comic books. Same game.
Or photographs of the crushed and compacted cubes from 70 different junk cars. Same game.
Or the thumbprints of 70 different people. Same game.
Or the retinal scans of 70 different people. Same game.
Or the genomes of 70 different species. You'd have to print really really small, or have huge balls, but same game.
All that matters is that we can tell 70 balls apart from each other. That's it. How we choose to label them has absolutely no effect on the chances of winning the game. Any labels will do, even if they lack the properties (such as order) of the integers 1--70. The lottery uses numbers out of convenience. It's easier to read and print numbers than to compare sine waves, photographs, snowflakes, etc. But the fact that those numbers have an order on the number line does not affect the odds of winning the game.
The peaks in your graph come from the order of the numbers. They do not indicate anything of the game, or increased odds of winning it. If the balls were printed with 70 unique thumbprints, you wouldn't be able to order them as Ball 1, Ball 2, etc like with the numbers, or discuss whether they are adjacent or spread out, or have your graph with the balls and their peaks.
But it would be the same game.
Last edited:
How much are people putting in? I was thinking of getting 5 tickets with the megaplier, so $15 total. I figure it's still within the realm of a nice 6-pack or 2 of beer or cider so instead of buying any booze this week I'll get this instead.
I always do 10 bucks. The last time it was huge I did 50 bucks and felt guilty that I could've put it in stocks. 😂
what's a megaplier? Should I get that as well? This is the first time I'm ever gonna play this, lol.
Double your winnings for a buck a ticket. Not the jackpot obviously but If you win a million it's 2. Or something like that, I never do it myself.
THErest, oh sure, of course I may be misunderstanding something, but it seems like I'm talking about A and you're talking about B. I agree with you on B 100%, but I am still interested in A, which is different than B. Also, I have always been talking about the relationship between the numbers. See one of my earlier posts from Tuesday.
Sometime this weekend, I will take another stab at explaining, and I will code up the two tests and run the experiment. If you care to, we can continue the conversation then.
Sometime this weekend, I will take another stab at explaining, and I will code up the two tests and run the experiment. If you care to, we can continue the conversation then.
To use your metaphor, you linked A and B when you said you could increase the odds of winning by choosing numbers near the peaks.
I've been trying to show you that it does not and will not work that way. The peaks are caused exclusively by A and do not have any relevance to B.
A being numbers on the number line and B being the probability of winning the game.
I'm gonna have to put a like, page 15 limit at this rate LOL at least everyone hoping is fun :)
Do you think you have a higher chance of winning if you let the computer quick pick, or would your chances go up if you pick the numbers yourself?
Any chance it hits 1B before tonight? Usually when a jackpot becomes newsworthy is when you get people coming out of the woodwork to buy stupid amounts of tickets.
I've wondered this too. When you hear about someone who wins, they typically mention how it was some string of numbers that had significance or whatever, but I'm always too lazy to actually fill out a card and just ask for quick pick. My sneaking suspicion/conspiratorial fantasy is that the generated numbers are designed never to win, lol
Computers I think are better at generating random numbers than humans. I also think relying on dates puts you in a smaller number pool as well as likely splitting the jackpot because people use dates.
Lucky numbers also drive you to keep playing I think. I know some heavy players who do daily plays and they always notice the one day they don't get to play their numbers coming out or something they thought about playing coming or an old number they haven't played for months coming. It just keeps you buying.
I don't see how it doesn't.
I've got 3 tickets. So long...SUCKERS!!!
If I win, let's assume the payout ends up being about $600 million.
- My parents, my sister and her husband, my mother-in-law, and bro-in-law are all getting $2 million ($6 million total)
- I'm meeting with a financial planner, and my wife and I are going to live the rest of our lives off of $20 million.
- That leaves $574 million to change the world.
But think of the dumb things you can waste with that money.
A pool full of swedish fish
A recreation of Les Mserables with puppets
A statue of baseball pitcher Rollie Fingers made of chicken fingers
ken_matthews
You seem to respond more to experimental probability, although the lottery is simple enough to work out theoretically. Maybe it will help if I scale down the game so we can actually see all possible outcomes. Let's try!
Instead of 70 balls numbered 1-70, we'll have just 5 balls labeled 1-5, and we will choose just two of those, order does not matter.
Essentially the same game but with different parameters so we can list all possible outcomes easily.
5C2 gives us 10 possible outcomes to this drawing. So any one play has a 1/10 probability of winning. Or 1:9 odds, if you prefer.
Indeed, here are all the possible outcomes (again, we don't actually care what order the two numbers are in):
1 and 2
1 and 3
1 and 4
1 and 5
2 and 3
2 and 4
2 and 5
3 and 4
3 and 5
4 and 5
If you made a "frequency" table based just on this list of all possible outcomes, and you decided that you would designate the least of each pair of numbers as "Ball 1" and the greatest of each pair as "Ball 2" (even though order didn't matter) you would see something like this:
If you ran a simulation of 100,000 trials and made a table, the curves would resemble these as well, with the same peaks.
By your reasoning in previous posts, you would claim that playing 1 and 5 would have better odds of winning the game than other choices. Because those are the peaks.
But we've already listed all possible outcomes to the game. 1 and 5 has the same probability as any other play. 10% chance of winning.
The only reason these peaks occur is because we, first of all, decided to label our balls with numbers instead of something unordered like snowflakes or thumbprints, and, secondly, we chose to classify the balls in order from least to greatest label. Thus, if (and only if) 1 appears as a winning number, it is guaranteed to be labeled as "Ball 1", the least of the two. And if 2 should appear, then it has a 3/4 chance of being labeled as "Ball 1". And if 3 should appear, it has a 1/2 chance of being labeled as "Ball 1". And so on. If 5 appears, it is impossible for it to be labeled as "Ball 1" and guaranteed that it is "Ball 2".
None of that changes the chances of winning. Changing how the balls are labeled has no effect whatsoever on the probability of winning. The numbers are only there so we can tell the 70 balls apart. No other reason.
You seem to respond more to experimental probability, although the lottery is simple enough to work out theoretically. Maybe it will help if I scale down the game so we can actually see all possible outcomes. Let's try!
Instead of 70 balls numbered 1-70, we'll have just 5 balls labeled 1-5, and we will choose just two of those, order does not matter.
Essentially the same game but with different parameters so we can list all possible outcomes easily.
5C2 gives us 10 possible outcomes to this drawing. So any one play has a 1/10 probability of winning. Or 1:9 odds, if you prefer.
Indeed, here are all the possible outcomes (again, we don't actually care what order the two numbers are in):
1 and 2
1 and 3
1 and 4
1 and 5
2 and 3
2 and 4
2 and 5
3 and 4
3 and 5
4 and 5
If you made a "frequency" table based just on this list of all possible outcomes, and you decided that you would designate the least of each pair of numbers as "Ball 1" and the greatest of each pair as "Ball 2" (even though order didn't matter) you would see something like this:
If you ran a simulation of 100,000 trials and made a table, the curves would resemble these as well, with the same peaks.
By your reasoning in previous posts, you would claim that playing 1 and 5 would have better odds of winning the game than other choices. Because those are the peaks.
But we've already listed all possible outcomes to the game. 1 and 5 has the same probability as any other play. 10% chance of winning.
The only reason these peaks occur is because we, first of all, decided to label our balls with numbers instead of something unordered like snowflakes or thumbprints, and, secondly, we chose to classify the balls in order from least to greatest label. Thus, if (and only if) 1 appears as a winning number, it is guaranteed to be labeled as "Ball 1", the least of the two. And if 2 should appear, then it has a 3/4 chance of being labeled as "Ball 1". And if 3 should appear, it has a 1/2 chance of being labeled as "Ball 1". And so on. If 5 appears, it is impossible for it to be labeled as "Ball 1" and guaranteed that it is "Ball 2".
None of that changes the chances of winning. Changing how the balls are labeled has no effect whatsoever on the probability of winning. The numbers are only there so we can tell the 70 balls apart. No other reason.
Don't forget about dear old Yasuke, that'll clear my student loans lmao
I like the cut of your jib, matey. 😎
I picked up 5 tix (10 bux) with rando numbers. I very rarely play the lotto; maybe once every couple of years, if the prize is ludicrous enough?
Good luck! 👍
One ticket, $2. It's a waste of money, statistically, so I'd prefer to not waste more than the minimum if I'm gonna play at all.
:) I really want to play but in my state, you have to buy with cash and it doesn't seem worth the risk to go out to catch COVID when we don't even go grocery shopping. Good luck!
I wouldn't even begin to know where to start to prep myself for receiving that much money. I wouldn't be able to trust anyone (family, friends, lawyers, financial planners... anyone).
Also, if someone wins are they able to keep their identity anonymous?
Also, if someone wins are they able to keep their identity anonymous?
You can only remain anonymous in Delaware, Kansas, Maryland, North Dakota, Ohio and South Carolina.
It's not like the money is gonna be sitting in your house, someone killing you (outside of a spouse or something) isn't going to net them anything. Main issue would be everyone you've ever known coming out of the woodwork and people trying to steal your identity, that's main threat of winning so much money.
If I win, everyone who posted in this thread before the numbers are called will get a $10 Wendy's gift card.
