I'm getting tired of this Kalshi victory lap. Below is a thread as to why. TLDR; forecasters and PMs are playing two very different games, PM performance does not suggest sophistication, and, when accounting for uncertainty, Kalshi is not reliably better. 🧵 1/9
We’ve just made all of our Midterm model data comparisons public @KalshiResearch. You can see them here: kalshi.com/research/forec….View quoted post on X
Today, the NYT claimed prediction markets were 'worse' than election models in the lead up to the midterms. But they ignore a few key pieces of the puzzle 🧵