5 ms·

> The math explanation could fill an article itself. If you are interested in how exactly he did it, here is a more detailed description: https://math.berkele

by lqet 6y ago

> The math explanation could fill an article itself. If you are interested in how exactly he did it, here is a more detailed description: https://math.berkeley.edu/~kpmann/ellenberg.pdf https://math.berkeley.edu/~kpmann/ellenberg.pdf

dmurray 6y ago
The main strategy is the tremendously complicated scheme of "buy as many tickets as you can physically key in, any time the prize pool goes above $x." Everyone who played the Michigan or Massachussets lottery that week was "hacking" the lottery the exact same way! That's not quite the optimum solution - what if you accidentally played the same numbers a million times, instead of a million different sets of numbers? So you should make sure you spread out your plays. However, I believe spreading them randomly is "good enough": the interesting maths only improves your edge a tiny amount.
- at-fates-hands 6y ago
  >> the interesting maths only improves your edge a tiny amount. Its only anecdotal evidence, but this would support your idea. I used to work for a mom & pop company. The owner was really wealthy. He would only play once the pot got to an insane amount (I think it was over $5 or $10 million) and then on his way into the office, would stop at 5-8 places, buying anywhere from $1-$3K in tickets a shot. I still vividly remember huge stacks of lottery tickets sitting on his huge desk in his office. He never did win the big pots, but did win a $1K or $2k amounts by matching a few of the numbers. But very clearly, he spent WAY more then he ever won.
  - themodelplumber 6y ago
    > But very clearly, he spent WAY more then he ever won. Which, in a recreational context, one could argue has more potential upside than a lot of other recreational activities (at least it's on the table, unlike other activities), and a pretty high baseline thrill as well.
    - umvi 6y ago
      You could argue that, but I think it's still a pretty weak argument. Arguments in favor of gambling as recreation are similar to ones made in favor of drugs as recreation where you basically just burn money in exchange for chemicals that make your brain feel good, but with few other benefits. Whereas other forms of recreation (IMO) give you more in return, such as increased dexterity, increased knowledge, strengthened human relationships, an interesting story to ponder and discuss, etc. I'm not saying people who gamble or use drugs are bad, I'm just saying the value of a given recreational activity can be measured along more axes than "dopamine yield"
      - circlefavshape 6y ago
        I don't know anything about gambling, but recreational drugs use can certainly strengthen human relationships and generate interesting (or at least amusing) stories
        krustyburger 6y ago
        I would not deny this, but it’s worth stating the obvious: most recreational drugs can also have the effect of destroying human relationships and generating stories that are highly tragic.
        advrs 6y ago
        [citation needed]
        isthatsoup 6y ago
        google "overdose deaths"
      - barkingcat 6y ago
        There isn't a rational way to value "thrill" - each person has a different thing that gets them the thrill. For you, maybe increased dexterity or knowledge is good, but for another person, that's downright boring. One other way to "value" thrill is whether it is legal or illegal, whether it damages other people or just a single individual. In terms of pleasure derived, you can't really say "this is useless entertainment" vs "this entertainment makes you smarter too so it's better"
  - treeman79 6y ago
    Done correctly, it’s investing at an awesome interest rate. Buying a bond that has a 10x return with a 50% default rate is still a great investment.
    - sandworm101 6y ago
      >> 50% default rate If only 50% of tickets are defaulting, there is something very very wrong with that lottery. This is more like a bond that pays a 100x return but has a 99.9% default rate.
      - treeman79 6y ago
        Yes.
  - MaximumYComb 6y ago
    That's the beauty of the lottery from the article. I've done the math on my local games and it tips into your favour after the jackpots get high enough. The issue with my local lotteries is that the EV comes partly from the main jackpot, which is a very low occurence outcome. The article has the EV come from the smaller prizes which can be easily won.
knuthsat 6y ago
It sounds like they were solving the lottery problem. [0] From my own research, the sets of tickets that give you the best chances still have negative expectation of reward. [0]: https://en.wikipedia.org/wiki/Lottery_mathematics#Minimum_number_of_tickets_for_a_match https://en.wikipedia.org/wiki/Lottery_mathematics#Minimum_nu...
based2 6y ago
https://en.wikipedia.org/wiki/Charles_Marie_de_La_Condamine https://en.wikipedia.org/wiki/Charles_Marie_de_La_Condamine
ab3rC1te 6y ago
I would say something but I'll pass. Working on breaking the lottery at the moment. Could give away trade secrets to make everyone rich.

by lqet 6y ago

dmurray 6y ago

The main strategy is the tremendously complicated scheme of "buy as many tickets as you can physically key in, any time the prize pool goes above $x." Everyone who played the Michigan or Massachussets lottery that week was "hacking" the lottery the exact same way! That's not quite the optimum solution - what if you accidentally played the same numbers a million times, instead of a million different sets of numbers? So you should make sure you spread out your plays. However, I believe spreading them randomly is "good enough": the interesting maths only improves your edge a tiny amount.

at-fates-hands 6y ago
>> the interesting maths only improves your edge a tiny amount. Its only anecdotal evidence, but this would support your idea. I used to work for a mom & pop company. The owner was really wealthy. He would only play once the pot got to an insane amount (I think it was over $5 or $10 million) and then on his way into the office, would stop at 5-8 places, buying anywhere from $1-$3K in tickets a shot. I still vividly remember huge stacks of lottery tickets sitting on his huge desk in his office. He never did win the big pots, but did win a $1K or $2k amounts by matching a few of the numbers. But very clearly, he spent WAY more then he ever won.
- themodelplumber 6y ago
  > But very clearly, he spent WAY more then he ever won. Which, in a recreational context, one could argue has more potential upside than a lot of other recreational activities (at least it's on the table, unlike other activities), and a pretty high baseline thrill as well.
  - umvi 6y ago
    You could argue that, but I think it's still a pretty weak argument. Arguments in favor of gambling as recreation are similar to ones made in favor of drugs as recreation where you basically just burn money in exchange for chemicals that make your brain feel good, but with few other benefits. Whereas other forms of recreation (IMO) give you more in return, such as increased dexterity, increased knowledge, strengthened human relationships, an interesting story to ponder and discuss, etc. I'm not saying people who gamble or use drugs are bad, I'm just saying the value of a given recreational activity can be measured along more axes than "dopamine yield"
    - circlefavshape 6y ago
      I don't know anything about gambling, but recreational drugs use can certainly strengthen human relationships and generate interesting (or at least amusing) stories
      - krustyburger 6y ago
        I would not deny this, but it’s worth stating the obvious: most recreational drugs can also have the effect of destroying human relationships and generating stories that are highly tragic.
        advrs 6y ago
        [citation needed]
        isthatsoup 6y ago
        google "overdose deaths"
    - barkingcat 6y ago
      There isn't a rational way to value "thrill" - each person has a different thing that gets them the thrill. For you, maybe increased dexterity or knowledge is good, but for another person, that's downright boring. One other way to "value" thrill is whether it is legal or illegal, whether it damages other people or just a single individual. In terms of pleasure derived, you can't really say "this is useless entertainment" vs "this entertainment makes you smarter too so it's better"
- treeman79 6y ago
  Done correctly, it’s investing at an awesome interest rate. Buying a bond that has a 10x return with a 50% default rate is still a great investment.
  - sandworm101 6y ago
    >> 50% default rate If only 50% of tickets are defaulting, there is something very very wrong with that lottery. This is more like a bond that pays a 100x return but has a 99.9% default rate.
    - treeman79 6y ago
      Yes.
- MaximumYComb 6y ago
  That's the beauty of the lottery from the article. I've done the math on my local games and it tips into your favour after the jackpots get high enough. The issue with my local lotteries is that the EV comes partly from the main jackpot, which is a very low occurence outcome. The article has the EV come from the smaller prizes which can be easily won.

knuthsat 6y ago

It sounds like they were solving the lottery problem. [0] From my own research, the sets of tickets that give you the best chances still have negative expectation of reward. [0]: https://en.wikipedia.org/wiki/Lottery_mathematics#Minimum_number_of_tickets_for_a_match https://en.wikipedia.org/wiki/Lottery_mathematics#Minimum_nu...

based2 6y ago

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

ab3rC1te 6y ago

I would say something but I'll pass. Working on breaking the lottery at the moment. Could give away trade secrets to make everyone rich.