Are Kalshi longshots overpriced? 19,838 trades say yes
Kalshi prices favourites honestly and overcharges for longshots by about 1.5 points. Measured across 19,838 trades in one week.
Short answer: yes, by about 1.4 percentage points, consistently, across 6,218 trades in a single week.
A price on a prediction market is a probability claim. Pay 15¢ for a contract and the market is telling you it pays out about 15% of the time. That claim is testable, but not by any individual trader, because you thousands of trades spread across the whole book to get a clean read, and by then you have paid for the education.
This lab makes roughly twenty thousand simulated trades a week across every Kalshi market it can reach, at the prices the order book was actually showing. That is a large enough sample to check the claim directly.
What the prices are worth
Of 21,796 settled trades in the week to 3 August 2026:
| Price paid | Trades | Market implied | Actually paid out | Difference |
|---|---|---|---|---|
| 1–9¢ | 1,601 | 5.5% | 4.1% | -1.4 pts |
| 10–19¢ | 2,477 | 14.3% | 12.6% | -1.7 pts |
| 20–29¢ | 2,140 | 24.7% | 23.7% | -1.0 pts |
| 30–39¢ | 2,586 | 34.1% | 34.0% | -0.1 pts |
| 40–49¢ | 2,084 | 44.5% | 42.2% | -2.3 pts |
| 50–59¢ | 1,935 | 54.3% | 55.6% | +1.3 pts |
| 60–69¢ | 2,157 | 65.1% | 64.0% | -1.1 pts |
| 70–79¢ | 2,355 | 74.2% | 73.9% | -0.3 pts |
| 80–89¢ | 2,504 | 84.6% | 85.5% | +0.9 pts |
| 90–99¢ | 1,957 | 93.2% | 93.3% | +0.0 pts |
Read the last column as: positive means the market underpriced that range and buyers were paid more often than the price implied; negative means you were charged for optimism.
Kalshi is honest at the top and expensive at the bottom
Above 90¢, the exchange is accurate to within half a point. 1,957 trades at an implied 93.2% paid out 93.3% of the time. If you are buying heavy favourites, the price is telling you the truth.
Below 30¢ it is a different market. 6,218 trades at an implied 15.6% paid out 14.2% of the time, a gap of 1.4 points that appears in every cheap bucket, not one outlier dragging an average.
That is the favourite–longshot bias: the oldest documented finding in betting markets, where long odds are systematically overpriced because people like buying lottery tickets. It has been measured at racetracks since the 1940s. It is alive on a CFTC-regulated exchange in 2026, and you can watch it in the table above.
What it means if you trade there
Buying cheap contracts on Kalshi means paying roughly a point and a half over fair value before you have had a single opinion about the underlying event. Then the fee lands on top. A strategy built on longshots has to overcome both before it earns anything, and in this lab's records, most of them never do.
The inverse is not a free lunch either. Favourites are priced correctly, which means there is no edge in buying them; you simply are not being overcharged. One seat here buys every contract above 90¢ and wins 93% of its trades, and still finishes red, because the exchange fee is charged whether you were right or not.
How this was measured, and what it does not prove
Every trade is simulated. Nothing here moved real money. Entries are taken at the ask the exchange was reporting, and each is charged Kalshi's real fee. Crossing the spread like this is the taker rate, ceil(7 × price × (100 − price) / 10000) cents a contract; a resting order pays nothing on all but about 130 series, which the fee study lists in full.
Three limits worth stating plainly. This is one week, and one week of any market can lie; the table is republished every Sunday, and the number to trust is the one that holds up across months. Fills are verified against recorded order-book depth only where the lab records depth, under 10% of trades, so the rest are entered at a reported price that cannot be re-checked afterwards. And a simulated fill never suffers the thing a real one does: being filled precisely because someone better informed wanted the other side.
The full per-seat board is published every week, and the underlying per-trade file is available to members, so none of this has to be taken on trust. If a figure here does not survive your own arithmetic, say so.
The Seats lists every strategy this lab runs, with its real entry thresholds and the Kalshi tickers it trades.