4 ms·
Best solution proposed here! Put your money where your mouth is Taleb and Silver! Btw, I read 3 of Taleb's books, 1 from Nate Silver. Both are pretty good. Tal
by starchild_3001 8y ago
Best solution proposed here! Put your money where your mouth is Taleb and Silver!
Btw, I read 3 of Taleb's books, 1 from Nate Silver. Both are pretty good. Taleb is the more original, but also more often-objectionable.
- dagw 8y agoBest solution proposed here So what should the bet be? As far as I can tell they're not disagreeing on anything that can easily be reduced to a bet
- lixtra 8y agoNate calls the chances like 80:20 Clinton:Trump. Taleb takes the side: 100,000 USD that Trump wins. If Trump wins, Nate pays 500,000 to Taleb. If Clinton wins, than Taleb pays 100,000 USD to Nate. Do this for each election.
- dagw 8y agoBut that doesn't seem to be the nature of their disagreement. Taleb's main complaint seems to be about how the swings in the prediction probabilities far out from the event indicates a fundamental flaw in how the underlying model uses probability. That being said it would be a fun little exercise to see who much money would be made if they bet had $1 on every race 538 predicted in 2016 and 2018 using the model odds.
- mhb 8y agoIt doesn't need to be a simple bet. It could incorporate the volatility which Taleb might believe shouldn't be present in an unflawed model.
- lixtra 8y agoIndeed after skimming through the paper[1] (I find it hard to grasp without a solid background) it seems that Taleb wants to play this game every day before the election and expects to come out ahead on election day when all the bets are settled. [1] https://arxiv.org/pdf/1703.06351.pdf https://arxiv.org/pdf/1703.06351.pdf