How Wise is the Crowd? Bias and Edge in Prediction Markets

Avaneesh Deleep, John Lee, Jenny Bai, Dhruv Suresh & Harsh Dhawan

Using tick-level order flow, wallet histories, and user commentary across Polymarket and Kalshi, the authors examine who actually generates and exploits pricing inefficiencies in modern prediction markets.

The classic favorite-longshot bias largely disappears once they control for contract lifecycle timing — what looked like that bias in “mention markets” is really a “Yes Bias,” where traders systematically overpay for the affirmative outcome.

“Whales” (the most capitalized traders) aren’t the sharpest; they tend to bleed expected value to small-order traders, trading on ideological conviction and suffering adverse selection.

The loudest participants add no edge — there’s no meaningful correlation between sentiment intensity and informational advantage, so vocal commentary is mostly noise.

Link to full article: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6322678

In:

, , , , ,

Other news

  • How Wise is the Crowd? Bias and Edge in Prediction Markets

    Using tick-level data from Polymarket and Kalshi, this study finds the classic favorite-longshot bias is largely a ‘Yes Bias,’ that the most capitalized traders systematically underperform smaller ones, and that the most vocal participants add no informational edge. It offers a reproducible method for denoising prediction-market probabilities.

  • Reconstructing a Century of U.S. Corporate Bonds

    A new 128-year database of U.S. corporate bonds (1895–2022) — over 100,000 bonds and 7 million observations — uncovers a sizable, statistically significant credit risk premium that shorter modern samples fail to detect. The findings support strategic allocations to corporate credit for long-horizon investors and establish a new benchmark for empirical asset pricing in fixed…

  • 18 Post

    Institute of Advanced Investment Management Article Showcase