-- only managed to make it half way through the article so apologies if this was covered later - re: the intern game - if the game is confidence in the probability of your answers - couldn't you deliberately get the answers wrong and just bet low confidence in your answers? - this seems like a really stupid comment on my part so I presume I'm missing something important --
You only get more tokens when you're right. So to "win" you have to bet high when you know you'll be right, so you get the most tokens, and bet low when you're not sure so you don't lose too many. Getting it wrong only loses you tokens.
No context around Jane Street, but I'd expect the goal to still be maximising your return / value. The people who are better at assessing probability and risk will have a higher value in aggregate.
Filter out those who made bets too big to avoid outliers.
Presumably the goal is to have a lot of chips at the end (more than you started with?). I would assume if you bet 10 chips, you get 10 for being right, and zero for not. So betting 0 chips all the time probably won't get you the job!
