Prediction Market Favorites Came True 85% of the Time in October 2026
Prediction markets put a price on everything from Fed decisions to award shows, and the price reads as the chance of the outcome. We record how every market we track settles, so each month we can check those chances against what happened. In October 2026, 4,154 outcomes in 1,144 events that traded at least $10,000 settled with a price on record a week before trading closed.
Outcomes priced at 70% or more happened 85% of the time, against an average price of 87%. Long shots priced at 30% or less happened 10% of the time, against 9%.
The price band furthest from its result was 60 to 70%: those 203 outcomes were priced at 64% on average and 58% happened.
How Calibrated Were the Odds
| Priced a week out | Outcomes | Average price | Happened |
|---|---|---|---|
| 0 to 10% | 1,193 | 3% | 3% |
| 10 to 20% | 334 | 14% | 17% |
| 20 to 30% | 369 | 25% | 28% |
| 30 to 40% | 330 | 35% | 32% |
| 40 to 50% | 410 | 46% | 42% |
| 50 to 60% | 616 | 52% | 52% |
| 60 to 70% | 203 | 64% | 58% |
| 70 to 80% | 197 | 74% | 71% |
| 80 to 90% | 169 | 85% | 79% |
| 90 to 100% | 333 | 96% | 95% |
By Category
| Category | Outcomes | Favorites (70%+) happened | Brier score |
|---|---|---|---|
| Sports | 2,085 | 75% vs 82% priced | 0.194 |
| Culture & Media | 499 | 88% vs 91% priced | 0.080 |
| Politics & Elections | 429 | 87% vs 87% priced | 0.101 |
| Crypto | 340 | 90% vs 90% priced | 0.103 |
| Markets & Companies | 264 | 97% vs 90% priced | 0.094 |
| Economy & Fed | 207 | 85% vs 92% priced | 0.103 |
| Tech & AI | 188 | 84% vs 87% priced | 0.122 |
| World | 125 | 83% vs 82% priced | 0.123 |
The Brier score is the average squared gap between price and result: 0 for perfect foresight, 0.25 for always saying 50%.
A Day Before Close
With a day to go, 13,795 outcomes had a price on record. Those priced at 70% or more happened 80% of the time, against 87%. A price a day out can already reflect a result that is partly known, such as a vote count under way, so the week-out figures are the cleaner test.
How We Measured
Every yes-or-no market on Kalshi and Polymarket that we track, settled yes or no, and traded at least $10,000 over its life. For each we take the last daily price at least seven days before trading closed. Fifteen-minute up-or-down contracts, parlays and multi-game combinations are left out. Each outcome in a multi-outcome event counts once.
The running record, all months together, is on our track record page.
Data: Frenzy Capital. This post is for information only and is not investment advice.
Comments (0)
No comments yet. Be the first.