Wolfe Research Report: Polymarket Predictions Outperform Historic Earnings Benchmarks in Key Scenarios

Polymarket users’ earnings predictions demonstrate higher accuracy than historic benchmarks, according to a Wolfe Research report highlighted by Bloomberg on April 16, 2026. When Polymarket bettors predict a company will miss earnings estimates, the firm misses 44% of the time, compared to a historic benchmark of 18% for misses.

Conversely, when Polymarket users are very confident a company will beat earnings estimates, the company meets or exceeds those estimates 90% of the time, surpassing the norm of 81%.

These findings suggest that the crowd-sourced predictions on Polymarket may provide a different signal than traditional sell-side analyst consensus. Yin Luo, who leads quantitative research at Wolfe Research, said the accuracy may stem from the diversity of Polymarket users compared to the sell-side analyst pool that typically drives consensus estimates.

“The accuracy may come from crowdsourcing, with Polymarket users more diverse than the sell-side analyst pool that drives consensus estimates." Separately, a working paper from London Business School and Yale researchers found that earnings markets for companies audited by one particular accounting firm were more accurate than those for companies with other auditors. The academics declined to name the accounting firm involved. Benzinga published an article on April 16, 2026, discussing the role of anonymous Polymarket bettors in predicting earnings, underscoring the growing interest in alternative forecasting methods. These developments highlight evolving approaches to earnings prediction, contrasting traditional analyst-driven estimates with crowd-sourced market data and auditor-related accuracy factors.”