GUIDE · Kalshi Fee Math: What Fees Actually Do to Your Trades
Kalshi Fee Math: What Fees Actually Do to Your Trades
Every affiliate page that explains Kalshi fees states the formula. Almost none state what the formula does to your profit and loss. This page does the arithmetic in public, with worked examples, so the fee is never a surprise at exit.
The fee schedule below was verified against three independent editions of Kalshi's published schedule, including the September 2022 CFTC rule filing (rule091222kexdcm003.pdf), and the coefficients have been stable since 2022.
The Formula, Verbatim
Kalshi's published schedule states, for the taker side:
fees = round up(0.07 x C x P x (1 - P))
Where C is the number of contracts, P is the price per contract in dollars, and "round up" means rounded up to the next cent. The maker side uses:
fees = round up(0.0175 x C x P x (1 - P))
Two details in that wording matter more than the coefficient:
- C sits inside the ceiling. The rounding applies to the whole trade, not per contract. A tiny fee on a tiny trade rounds up to a full cent regardless.
- INX and NASDAQ100 tickers are half rate. All S&P 500 and Nasdaq-100 series use 0.035 for the taker, not 0.07.
There is no flat percentage. Anyone quoting "Kalshi charges X percent" is quoting a number that does not exist. The fee depends on price, size, and side.
Worked Examples, From the Schedule Itself
Kalshi's own published table confirms the formula against two cases:
| Case | Computation | Fee |
|---|---|---|
| P = $0.50, C = 100 | 0.07 x 100 x 0.50 x 0.50 = 1.75 | $1.75 |
| P = $0.20, C = 1 | 0.07 x 1 x 0.20 x 0.80 = 0.0112, rounded up | $0.02 |
The second row is the one to stare at. A single contract at $0.20 costs $0.02 in fees: that is 10% of the $0.20 stake. The rounding rule exists precisely so that small orders on cheap contracts pay a full cent even when the raw fee is a fraction of one.
Break-Even as a Share of Stake
Because C = stake / P, the fee formula simplifies. Substitute and cancel:
fee = 0.07 x (stake / P) x P x (1 - P) = stake x 0.07 x (1 - P)
So the fee, as a fraction of your stake, is:
break-even = 0.07 x (1 - P)
That is a parabola peaking at $0.50. The most expensive place to trade, as a share of stake, is at even money, where the break-even is 3.5% of your stake. Away from $0.50 the fee share falls, until the rounding rule starts eating again at the longshot end.
| Entry price P | Fee as share of stake (before rounding) |
|---|---|
| $0.90 | 0.7% |
| $0.70 | 2.1% |
| $0.50 | 3.5% |
| $0.30 | 4.9% |
| $0.10 | 6.3% |
Why a Flat Edge Threshold Cannot Work
Here is the trap that catches most new traders. Two trades with an identical edge of 1.20%:
- Trade A at $0.8693: break-even is 0.07 x (1 - 0.8693) = 0.91%. The 1.20% edge clears it. Fine.
- Trade B at $0.2965: break-even is 0.07 x (1 - 0.2965) = 4.92% before rounding, and rounding on a small stake pushes it higher. The 1.20% edge does not come close. This trade cannot win.
Same edge percentage, opposite verdicts. A single flat threshold, such as "only trade 2% edges", sends you into trade B over and over because it cannot tell the two apart. The threshold has to be the price-dependent break-even, computed per trade. Our arbitrage calculator does exactly this.
How Polymarket Compares
Polymarket's model is structurally different, and both are implemented in our scanner with every coefficient cited:
- Polymarket taker: fee = C x rate x P x (1 - P), no rounding documented. The rate is per market, published live at the Gamma API feeSchedule.rate, typically 0.04 to 0.07 by category.
- Polymarket maker: zero. Makers pay nothing, which is a verified zero, not a discount.
- Kalshi maker: 0.0175, still nonzero, still quadratic, never a flat percentage.
The practical difference: on Polymarket the maker side is genuinely free, which is why resting orders are the default advice there. On Kalshi the maker fee exists but is a quarter of the taker rate, so limit orders still win on cost.
What This Means for Your Trades
- Compute the fee per trade, not per platform. The formula depends on price and size, so a "Kalshi fee is X" rule is always wrong.
- Avoid small orders on longshots. The rounding rule turns a 0.5% raw fee into 10% of stake at $0.20. If you trade longshots, size up or accept the fee share.
- Price the maker side in. Even at 0.0175, the maker fee is real. A resting order still needs 0.0175 x (1 - P) of edge just to break even.
- Do not trust a flat edge threshold. Use the price-dependent break-even, as shown above, or you will systematically take the trades that cannot win.
For the full scanner implementation, with sources for every coefficient, see our methodology page.
FAQ
Q: What is Kalshi's taker fee formula?
A: Fees = round up(0.07 x C x P x (1 - P)), rounded up to the next cent, with the ceiling applied to the whole trade. INX and NASDAQ100 tickers use half the rate, 0.035.
Q: What is Kalshi's maker fee?
A: Fees = round up(0.0175 x C x P x (1 - P)). Makers pay a quarter of the taker rate, unlike Polymarket where makers pay zero.
Q: Why are longshot contracts the worst place for small orders?
A: Because the ceiling rounds up to a full cent. A single $0.20 contract has a raw fee of about half a cent, which rounds to $0.02: 10% of your stake. The same rounding does not bite at $0.50.
Q: Is the 3.5% break-even at $0.50 correct?
A: Yes, before rounding. At P = $0.50 the fee is stake x 0.07 x 0.50 = 3.5% of stake. That is the peak of the fee parabola.
Q: Does Kalshi charge a flat percentage?
A: No. The fee is quadratic in price, not flat, and the schedule has been verified across three editions including the September 2022 CFTC filing.
Q: What edge do I need to break even?
A: 0.07 x (1 - P) as a fraction of stake, plus the rounding effect on small trades. Compute it per trade; a flat threshold cannot do this.
Last updated: 2026-08-11