Priced In

EP01 · October 18, 2026

64% to 85% in One Hour: Was Brazil's Vote Priced In?

Within an hour of the polls closing in Brazil, the market's estimate of a Flávio Bolsonaro presidency went from 64% to 85%. We check what the papers said against what prediction markets did, story by story, and say where we can't call it yet.

Bought it

Brazil: Flávio Bolsonaro tops round one

Polymarket, who wins the presidency: Flávio Bolsonaro 64% at 20:00 UTC to 85% at 21:00 UTC on Oct 4; Kalshi 65% to 87%. Both platforms at 86% on Oct 8. Runoff Oct 25.

Line chart of the market estimate of a Flávio Bolsonaro presidency jumping from 64% to 85% as Brazilian polls closed on October 4, 2026

Sources: AP via France 24 · Al Jazeera · BBC News

Priced in

Spain: Sánchez calls a snap election for November 29

Polymarket, next prime minister: 15% to 23% late on Oct 4 UTC, after first reports and before the official announcement, then 20%. The market on the election call itself is too thin for a verdict (under $250,000 notional volume).

Line chart of the Polymarket estimate for Spain's next prime minister rising the night before the official election announcement

Sources: Reuters via France 24 · Al Jazeera · The Guardian

Ignored

US Senate: Trump rallies in Ohio

Polymarket, Democratic control of the Senate: 64.5% on Oct 4, touched 67.5%, 63.5% on Oct 8. Never moved more than 3 points. Kalshi 62%. Neither story is directly about Senate control, so read with care.

Line chart of the Polymarket estimate of Democratic Senate control staying within a three-point range from October 4 to 8, 2026

Sources: Al Jazeera · CNBC

Priced in

The Fed: a pause in October

Polymarket, no change in October: 32% on Sep 29 to 83.5% on Oct 8. 44 of those 52 points came before the jobs report (Oct 2, 12:30 UTC). Kalshi 83%.

Line chart of the Polymarket estimate of a Fed pause in October rising 44 of 52 points before the September jobs report

Sources: Bureau of Labor Statistics · CNBC

Open

The AI task force

No contract on the task force. Context only: best AI model market, Polymarket about $65,000 a week (thin), notional. The newsroom uses AI tools from several companies in this market.

Line chart of the Polymarket best-AI-model market around the task force announcement

Sources: BBC News

Open

Can you trust this market? Indiana data centers

Polymarket, Indiana data center moratorium by end of 2028: 25.5% on Oct 8, up about 11 points in 24 hours, down 22 in a week. Volume $555,000 in 7 days (notional, above threshold). Contract created Oct 1. Concentration of trades not checked.

Line chart of the Polymarket estimate for an Indiana data center moratorium swinging between 11% and 21.5% in one day

Sources: CNBC

Open

Papers vs market: Iran

Polymarket, US-Iran ceasefire holding through Oct 31: about 69% until 16:05 UTC on Oct 8, 80% about 20 minutes later, 78% at the Oct 8 read. Oct 4 rise of 6 points: no verdict, first report not timed.

Line chart of the Polymarket estimate of the US-Iran ceasefire holding, jumping on October 8, 2026

Sources: BBC News · The New York Times · OilPrice

Editor's call

This is the editor's call, read here by NEWS. The presenters are AI, but the choices, and any mistakes, are mine. My call this week is about timing. In Spain, the price rose before the official announcement, after the first reports leaked. At the Fed, most of the move came before the jobs report. In Brazil, it moved as the first results came in. On Iran, the price jumped minutes after a post, but I haven't read the original timestamp, so I'm not calling it yet. A quick move doesn't mean somebody knew something. It can mean the story was expected, or that the market is thin. We don't know who traded, and I won't guess. So here's my question. When the headline and the price disagree, which do you trust more, and why? Tell us in the comments. The editor.

Market data as of 2026-10-08T19:03:00Z. News analysis, not financial advice.