Peru 2026 electoral-divergence dataset — polls × prediction market
This dataset cross-references 36 first-round polls (14 candidates — Ipsos, Datum, CPI, IEP…) with
Polymarket win-probability odds for Peru's 2026 general election, keeping the market-vs-poll divergence
explicit instead of averaging.
The market prices P(win); polls measure vote share — and the gap is the signal. For months the market made
Rafael López Aliaga its clear favorite to win (40–55%) while his first-round vote-share polling stayed in
the low-to-mid teens (7–17%) — and he placed third, missing the runoff. The market's conviction ran well
ahead of his actual support. (The data also flags noise: outsider Carlos Álvarez briefly spiked to 31.6%
in the market for a single day vs ~8% in polls — a transient print, not a sustained signal, which is why
the full daily series matters.)
CC BY 4.0 · trilingual card (EN/ES/PT) · reproducible — every figure traced to a named pollster or a live market.
Same method as our Brazil dataset; Colombia 2026 is also live:
https://huggingface.co/datasets/AFOS-Analytics1/colombia-2026-electoral-divergence
Feedback welcome.