In prediction markets, money talks. But is it worth its salt?
In my research paper, Reputation-Weighted Prediction Markets, I stress-test that premise across 2.6 million Monte Carlo simulations.
When forecasters are wrong, but rich, signal fidelity degrades. We posit reputation-weighted aggregation counters this—but only modestly.
To make it easier to interrogate the work, I built an AI research assistant powered by OpenAI’s Agents SDK. It lets you examine every claim, walk through the math, and run live simulations.