Will the Movement for Socialism (MAS-IPSP) win the most gubernatorial elections in Bolivia in 2026?

Analysis

Seven of eight agents converge strongly on MAS having very low probability (5-18%), with Agent 1 (50%) being an obvious outlier due to research failure (0 pages read). Excluding Agent 1, the mean drops to ~11.4% and median to ~8%. The consensus is overwhelming and well-evidenced:

Where agents agree (high reliability): - MAS suffered catastrophic collapse in 2025 general elections (75→2 Chamber seats, 0 Senate seats) - Party is deeply fractured between Arce and Morales factions - Multiple MAS-affiliated platforms were disqualified by TSE - No MAS candidate appears to be a frontrunner in any department - Bolivia Brief describes a 'post-MAS era'

Strongest evidence agents (4, 5, 7): - Agent 5 importantly corrected the shared context: MAS actually won only 3/9 governorships in 2021, not 6/9 as stated in base rates. This is critical - it means MAS was already declining even while holding the presidency. - Agent 4 and 7 found that MAS is priced lowest among all 6 parties on Polymarket, and that expert analysis doesn't mention MAS as competitive in any key race. - Agent 6 raised the crucial resolution ambiguity: MAS is running through 'Unidos por los Pueblos' alliance, which may not count as MAS-IPSP wins.

Why market could be wrong (anti-confirmation): For MAS to win at 20%, you'd need to believe opposition fragmentation is severe enough that MAS wins plurality of governorships with small vote shares in multiple departments. But with 5+ opposition parties each priced higher than MAS, and MAS not favored in any single department, this scenario requires MAS to somehow win 3-4 departments where it's not even the frontrunner. The resolution also requires winning 'the most' - meaning more than LIBRE (40%), AP (30%), APB Súmate (26%), PDC (24%), and UNIDAD (22%). The sibling market prices sum to ~162%, implying significant probability of ties or other outcomes, but MAS is clearly the weakest.

Edge calculation: Market at 20%, my estimate 8% = 12 percentage point edge on NO side. This exceeds the 5% threshold significantly. The evidence is specific, verifiable (2025 election results are public record), and there is strong agent consensus. The 'YES=0.20 bucket is toxic' warning applies here - this looks like a classic case where the market is slightly overpriced for a low-probability outcome.

I set probability at 8%, aligning with the three strongest-evidenced agents (4, 5, 7) rather than pulling toward market price, because the evidence is specific, verified, and directionally unanimous.

View on Polymarket