Paper: arXiv 2611.00338

Authors: Raphael Coelho

Abstract

We rebuild the public count of Brazil’s 2026 presidential runoff minute by minute from the electoral court’s section-level records and compare it with every trade on Kalshi and Polymarket. A projection published live on the night gave the eventual winner, Lula, a probability of 0.998 from 19:15, with 1.6% of sections counted. The markets reached 0.9 at 20:24, with 46% counted. Read as the projection’s central estimate with a wider uncertainty, their prices imply a standard deviation of the final margin about four times the live projection’s, and about twice that of a section-level projection calibrated on the 2018 runoff, the earlier night with the largest errors. Valued at these projections, takers’ purchases of Lula after the count resumed had an expected profit of 4 to 18 cents per Kalshi contract after fees; valued at a state-level extrapolation of the count, it was zero. During a two-hour suspension of the count the markets gave Lula 0.25 to 0.44, below every reading we compute afterwards from the sections then public, none of which was available live. On the first-round night, when the headline count showed the eventual leader ahead from its first file, the same markets were faster than our projection.

Complexity vs Empirical Score

  • Math Complexity: 4.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper demonstrates strong empirical rigor through detailed data reconstruction and comparison with prediction markets. While the methodology is sound, the mathematical complexity is moderate, focusing more on statistical application than advanced theoretical derivations. The novelty lies in the granular, minute-by-minute analysis of election market efficiency against real-time public information.

Research Flowchart

  flowchart TD
    A[Research Goal: Compare Prediction Markets vs. Live Projection for Brazil's 2026 Election] --> B{Key Methodology: Rebuild Public Count & Compare with Market Trades};
    B --> C[Data/Inputs: Electoral Court Section-Level Records, Kalshi & Polymarket Trade Data];
    C --> D{Computational Processes: Minute-by-Minute Count, Probability Projections, Expected Profit Calculation};
    D --> E[Key Finding 1: Live Projection (0.998 prob. for Lula at 1.6% count) Outperformed Markets (0.9 prob. at 46% count) in Runoff];
    D --> F[Key Finding 2: Markets' Implied Std Dev of Final Margin 4x Live Projection's, 2x 2018-calibrated Projection's];
    D --> G[Key Finding 3: Takers' Purchases of Lula after count resumed had expected profit (4-18 cents Kalshi) based on projections, zero based on state-level extrapolation];
    D --> H[Key Finding 4: Markets were slower than projection during runoff suspension (0.25-0.44 for Lula vs. higher computed readings), but faster in first round];