Structural Analysis
AI-generatedPredictIt is the worst-calibrated exchange in the research — only 78% outcome-correctness on Kalshi vs. 67% on Polymarket, and PredictIt is historically even softer on non-US elections where retail participation skews heavily toward name-recognition bets. Zema is a longshot in a multi-outcome Brazilian presidential market, and the favorite-longshot bias (Snowberg & Wolfers) means retail traders systematically overprice low-probability outcomes like this — contracts trading as deep longshots lose over 60% of capital on average per the Burgi et al. findings.
ResolutionThis is a multi-outcome market on PredictIt, which means prices across all candidates must sum to roughly 100 cents, but in practice they often don't — creating phantom value that evaporates at settlement. The resolution criteria link is vague enough that a runoff scenario (Brazil's two-round system) creates real ambiguity: does 'win' mean winning the first round outright, or winning the presidency after a runoff, and traders who haven't confirmed this distinction are exposed.
Resolution criteria available at: https://www.predictit.org/markets/detail/8544/Who-will-win-the-2026-Brazilian-presidential-election
CalibrationThe research shows that long-dated political markets compress probability toward 50%, meaning longshots get systematically overpriced relative to their true odds — traders should assume the true probability of a Zema win is meaningfully lower than the market price implies, not higher. PredictIt's 67% outcome-correctness rate (the lowest of the major exchanges) means mispricing persists longer here, but it cuts both ways: the market is more likely to be wrong, and correcting that wrongness requires a catalyst that retail flow may never provide.
RisksThe related market — 'Will Zema finish second in the first round?' — is anti-correlated with this one, so traders holding both are not diversifying, they're creating a structural contradiction in their book. Prices are described as 'tightly aligned' across Polymarket and PredictIt, which means the cross-exchange arbitrage window that peaked in final weeks before resolution (Clinton & Huang) is already mostly closed, removing a key edge most traders assume is available.