All resolved pollsResolved · 17 AUG 2026Market beat the crowd
US jobless claims for the week ending 15 August
Resolved
The verdict

US jobless claims for the week ending 15 August

US jobless claims for the week ending 15 August 631 verified humans answered. Crowd ran well above market on higher claims, a gap the outcome did not vindicate.

68%Crowd on 210,000 or more
47%Market price
+21Points of divergence

How the crowd voted

631 verified humans · one human, one vote
210,000 or moreCrowd’s pick68%427 voters
Fewer than 210,000What happened32%204 voters
Fewer than 210,000 won, so 32% of the crowd called it — a plurality of 68% did not. The market had 210,000 or more at 47%.

Where the crowd split from the market

Market depth: deep
Market
47%
68%
Crowd
+21 ppts

The crowd was more confident in 210,000 or more than the betting price implied. That gap is the interesting part of this poll, and it went against the result.

The market price is a yardstick, not a recommendation. The crowd’s read is sentiment, not a prediction or financial advice.

What the crowd was thinking

Scenario 1Crowd’s pick
68%427 voters
210,000 or more

The crowd's dominant vision: a labour market softening enough to push weekly filings to elevated territory, with clear majority conviction behind this read.

Scenario 2Came true
32%204 voters
Fewer than 210,000

A minority view within the crowd, holding that claims would stay contained below the threshold, closer to the market's own implied lean.

How well calibrated was the crowd?Brier score, conviction and full methodology0.92
Brier score
0.92

0 is a perfect confident call, 0.25 a coin flip. Higher means the crowd leaned away from what happened.

Conviction
1-in-1

The odds the crowd’s own split implied for its top pick. Shorter odds mean more conviction.

Ballots correct
204 of 631

Ballots whose pick matched the result. Computed from the crowd’s own vote split — no betting market involved.

Crowd sentiment is shown only after a poll closes, never mid-vote. “Crowd called it” means the crowd’s top pick matched the result; it is not a claim the crowd is right in general. Aggregate data from World ID voters — a crowd view, not a prediction or advice.