- The fee you pay as a taker is largest near 50¢ and shrinks toward the extremes. On Kalshi, 100 contracts at 50¢ cost $1.75 in fees; at 90¢ they cost $0.63.
- As a share of what you put up, the fee is the reverse: 3.5% at 50¢ but 6.3% at 10¢. Cheap longshots are the most expensive bets to take.
- Resting (maker) orders pay nothing on Polymarket and, by default, nothing on Kalshi, so how you enter matters as much as what you buy.
- Research on prediction-market accounts finds only 3.14% are statistically skilled. Before you think you have an edge, subtract the fee from it.
Prediction markets betting means buying a contract that pays $1 if an event happens and nothing if it does not, at a price between 1¢ and 99¢. The price is the market's implied probability, so a 62¢ contract says roughly 62%. You only profit over time if your own estimate beats the price by more than the fees. This guide uses the fee formulas published by Kalshi and Polymarket, with worked numbers, plus our tracked-wallet data as of the 2026-10-13 snapshot. It is not advice, and the method is on our methodology page.
The fee formulas in plain terms
A taker is a trader who accepts a price already posted in the order book. A maker is one who posts an order and waits for someone to take it. Both venues charge takers with a formula built on p × (1 − p), which peaks at 50¢ and falls toward zero at 1¢ and 99¢.
- Kalshi: fees = round up(M × 0.07 × C × P × (1−P)), where C is contracts and P is the price in dollars. Under the schedule effective 2026-07-07 the default taker multiplier is 1 and the default maker multiplier is 0 (see the fee schedule in the sources). The rate for resting orders is 0.0175 where a maker multiplier applies. Combo markets use a taker multiplier of 2.
- Polymarket: fee = C × rate × p × (1−p) for takers, and makers pay nothing. The rate depends on category: 0.04 for politics, finance, tech and mentions; 0.05 for sports, economics, culture, weather and other; 0.07 for crypto; and 0 for geopolitics (fee documentation, see sources).
What 100 contracts cost in fees at each price
The table works the formulas for 100 contracts bought as a taker. It is our own calculation, not a quote from either platform, and fees are rounded to the cent here.
| Price | Cost of 100 contracts | Kalshi fee | Kalshi fee as % of cost | Polymarket politics (0.04) | Polymarket sports (0.05) |
|---|---|---|---|---|---|
| 10¢ | $10 | $0.63 | 6.3% | $0.36 | $0.45 |
| 20¢ | $20 | $1.12 | 5.6% | $0.64 | $0.80 |
| 50¢ | $50 | $1.75 | 3.5% | $1.00 | $1.25 |
| 80¢ | $80 | $1.12 | 1.4% | $0.64 | $0.80 |
| 90¢ | $90 | $0.63 | 0.7% | $0.36 | $0.45 |
Two things stand out. In dollars, 50¢ is the costliest price. As a share of the money at risk, 10¢ is the costliest. So the percentage fee is highest on low-priced longshots, the bets least likely to pay.
How much edge you need: a worked example
Suppose a contract trades at 50¢ and you believe the true chance is 55%. Your edge is 5 percentage points, or 5¢ per contract. On Kalshi the taker fee at 50¢ is 1.75¢ per contract (the $1.75 above divided by 100), so your edge after fees is 3.25¢. On Polymarket politics the fee is 1¢ per contract, leaving 4¢. The numbers are hypothetical and show the method: subtract the fee, in cents, from your estimated edge in points.
If your edge is smaller than the fee, a taker trade has a negative expected result even when you are right about the direction. Use the fee calculator to run your own price and size.
Why order type matters in prediction markets betting
Because makers pay no fee on Polymarket and the default Kalshi maker multiplier is 0, a resting order can remove most of the cost above. The trade-off is that it may not fill, or may fill only when the price moves against you. A fee saving is not the same as an edge: you still need a view that beats the price.
In our tracked cohort of 200 profitable Polymarket wallets as of 2026-10-13, 11 (5.5%) earn most of their profit from market making and 175 (87.5%) are directional. The cohort is chosen from leaderboards, so this is not proof that either style is easier. More on the split is in who actually makes money on Polymarket.
Do people make money on prediction markets?
Some do, but the gains are concentrated. A CEPR discussion paper covering 588,000,000 trades and $67 billion of volume reports that the top 1% of users with positive profit took 76.5% of profits (CEPR paper). A separate study finds 3.14% of accounts are statistically skilled and that 60% of lucky winners revert (skill-versus-luck paper). On the all-time overall leaderboard we track, the top 10 of 50 listed wallets hold 40.0% of the profit shown; in politics they hold 61.6%.
That does not mean you cannot profit. It means a leaderboard rank is a weak guide, and fees take a fixed bite from every taker trade you make.
Where this can go wrong
- Fee schedules change; both formulas here are as published at the snapshot date, and the venue's current documentation governs.
- Kalshi rounds fees up per order, so small orders pay proportionally more than the formula suggests. A single contract at 50¢ is charged $0.02.
- Your probability estimate is the hard part. If it is no better than the market price, your edge is zero before fees.
- Thin books mean your fill price can differ from the quote, which is a cost on top of the fee.
- Platforms differ in which categories they list and where you can trade; check availability before you plan around a fee.
Frequently asked questions
Can you actually make money on prediction markets?
Some accounts do, but the research we cite finds profit is concentrated: the top 1% of users with positive profit took 76.5% of profits, and only 3.14% of accounts are statistically skilled. Fees reduce every taker trade, so you need an edge larger than the fee.
Do most people lose money in prediction markets?
The studies we cite do not give a share of losing accounts, so we cannot give you that figure. They show that gains are top-heavy and that many leaderboard winners revert.
Which price has the highest fee?
In dollars, 50¢, where the p × (1−p) term peaks. As a percentage of the money you put up, low prices cost more: 6.3% at 10¢ against 3.5% at 50¢ on Kalshi.
Next steps
Browse live prices on the politics markets page and compare the two venues' rules in Kalshi vs Polymarket, or read What are prediction markets? if the vocabulary is new. For a side-by-side of fees across platforms, see our fee comparison and the compare page.
The short version: the price tells you the market's odds, the fee tells you what it costs to disagree with them, and your edge has to cover both.