Show HN: A Monte Carlo model that called the World Cup finalists 2 weeks ago
1–5 of 5 posts
Re: Show HN: A Monte Carlo model that called the World Cup finalists 2 weeks ago
#2Link to preprint: https://papers.ssrn.com/abstract=7013338
Constructive feedback very much appreciated!
Re: Show HN: A Monte Carlo model that called the World Cup finalists 2 weeks ago
#3Did it confidently predict the same finalists consistently in every iteration? Or did it sometimes give different answers?
Re: Show HN: A Monte Carlo model that called the World Cup finalists 2 weeks ago
#4Did it confidently predict the same finalists consistently in every iteration? Or did it sometimes give different answers?
Thanks for your question. It's Monte Carlo, so individual simulated tournaments produce all sorts of champions, Argentina wins ~28% of the 100k iterations. What's stable is the aggregate: Argentina and Spain consistently come out as the two highest-probability teams, and that barely moves across different seeds.
Re: Show HN: A Monte Carlo model that called the World Cup finalists 2 weeks ago
#5More discussion https://news.ycombinator.com/item?id=48919191 (38 points | day ago | 52 comments)
Luckily the English guy was wrong.