How Accurate Are Prediction Markets on the Jobs Report? Payrolls, Unemployment and Jobless Claims

By Arthur Angelo, PredictionScout Research · Updated Oct 15, 2026, 7:51 AM ET

Key findings

• Payrolls are the hardest call. The jobs number landed a median 66K from the market median (average 64K). Only 68% of 227 threshold contracts went the way the market leaned a day before. Payrolls beat the market median 13 times and fell short 7 times.
• The unemployment rate is far more predictable. Typical miss 0.07 points; 93% of 188 contracts called right.
• Claims kept coming in low. Weekly initial claims came in below the market median in 44 of 59 weeks (above in 15). Typical miss 6.2K; 86% of 561 contracts called right.
• Biggest payrolls miss: Feb 2026. Market median +50K, first BLS print -92K.
• Settlements verified. Matching first-release values in ALFRED: payrolls 19 of 20, unemployment 20 of 21, claims 51 of 59.

Nonfarm payrolls (monthly jobs added)

Across 20 releases from Jan 10, 2025 to Oct 2, 2026, the market’s median a day before was off by 66K in a typical release (median miss) and 64K on average. The number came in above the market median 13 times and below it 7 times. Of 227 individual “above X” contracts, the side priced above 50¢ a day out was right 155 times (68%).

Highlighted rows missed by more than 100K. Official values: change in total nonfarm payrolls as first published by BLS (ALFRED vintage of PAYEMS on release day).

Unemployment rate

Across 21 releases from Jan 10, 2025 to Oct 2, 2026, the market’s median a day before was off by 0.07 pt in a typical release (median miss) and 0.07 pt on average. The number came in above the market median 13 times and below it 8 times. Of 188 individual “above X” contracts, the side priced above 50¢ a day out was right 175 times (93%).

Highlighted rows missed by more than 0.15 pt. Official values: unemployment rate as first published by BLS (ALFRED UNRATE, first release).

Weekly initial jobless claims

Across 59 releases from Jun 12, 2025 to Oct 8, 2026, the market’s median a day before was off by 6.2K in a typical release (median miss) and 7.4K on average. The number came in above the market median 15 times and below it 44 times. Of 561 individual “above X” contracts, the side priced above 50¢ a day out was right 483 times (86%).

Highlighted rows missed by more than 15.0K. Official values: seasonally adjusted initial claims as first published by the Labor Department (ALFRED ICSA, first release, matched by week-ending date).

Download the data

Release-by-release scorecard (CSV)

One row per settled Kalshi event: the market median 24 hours before, Kalshi’s settlement value, the official first-release value and the miss. Per-contract prices are not included. Free to reuse under CC BY 4.0 with a link back to this page.

Methodology

• Markets. Kalshi’s KXPAYROLLS (monthly change in total nonfarm payrolls), KXU3 (unemployment rate) and KXJOBLESSCLAIMS (weekly seasonally adjusted initial claims) series, from Kalshi’s public API. The headline window starts in January 2025; older events are in the CSV.
• Official data. First-published values from ALFRED: the PAYEMS vintage on release day (payroll change = latest month minus prior month in that vintage), UNRATE and ICSA first releases. Sources: BLS and the Labor Department.
• Market median. For each release we take every “above X” contract’s price 24 hours before trading closed (midpoint of bid and ask from hourly candles when the spread is 10¢ or less, otherwise the last trade; looking back up to 48 hours). Prices are forced to fall as the threshold rises, and the median is the point where the implied chance crosses 50%, interpolated between neighboring thresholds. If the 50% point falls outside the listed thresholds the release is shown but not scored.
• Contracts called right. Each threshold contract counts once: right if the side priced above 50¢ a day before won. Contracts priced exactly 50¢ are skipped.
• Actual value. The value Kalshi settled on (its “expiration value”), which follows the first official release. We check it against the first-release value in ALFRED (St. Louis Fed’s archive of data as first published) and mark any difference.
• Limits. One exchange; thin trading in some older and far-out contracts; the market median is a summary of a full distribution, not a single forecast. Nothing here is investment advice. Last run: Oct 15, 2026, 7:51 AM ET.

Data sources: prices and volumes come from the public APIs of Kalshi and Polymarket, with labor data from the Bureau of Labor Statistics and the Department of Labor. All trademarks belong to their owners. PredictionScout is independent and not affiliated with any exchange. Nothing here is investment, betting, or legal advice.

Related: CPI scorecard · GDP scorecard · release-day odds · All research