Show HN: Beating the Optimal Stopping Rule with Game Specific Knowledge

ctsrc.github.io

18 points·codetrotter··8 comments

Yesterday there was this post titled Show HN: Learn When to Quit [0], and in the comments people were talking about how you could perform better in that game because of knowledge about the distribution of the numbers in the game.

In the spirit of the original post I decided to throw a little simulation together and put that claim to the test.

Page source for my simulation is on GitHub [1].

[0]: https://news.ycombinator.com/item?id=21701124#21732695

[1]: https://github.com/ctsrc/simulated-victor-game

I love the fact that you named it Victor Game, but I'm not the one who created it. I got the whole thing from vsauce2, as I explained on https://github.com/victorqribeiro/googol

Glad you like it :) I figured I should give credit where credit is due and also since your game differs a little bit from the original video I think one could say that you are the creator of the game ^-^

I think there's somethings wrong on your get random big number function. For instance you're incrementing "i" twice.

Whoops! Fixed, thanks. Guess I wrote that loop as a while-loop first and then overlooked the increment inside the loop later when I changed it to be a for-loop instead, heh.

Well done ;)

Thank you, and well done you too on the original comment that you wrote which this is an implementation of :)

Flagged as seizure-inducing flashing is dangerous.

Reply on news.ycombinator.com