Start with the visible fee gap
Using teaching base rates of 0.02% maker and 0.05% taker, a 10,000 USDT fill costs 2 versus 5 USDT per side. An all-maker round trip is 4 USDT and all-taker is 10 USDT, a visible gap of 6 USDT.
A maker entry with a taker risk exit costs 7 USDT and is often a more conservative budget than assuming a stop can fill passively. Replace the teaching rates with your signed-in tier.
Break-even is not one fill-rate number
If a missed order is chased 0.04% higher, the adverse movement costs 4 USDT on 10,000 USDT. The maker saving is only 3 USDT per side, so one chase already exceeds it.
Track fill share, waiting time, price movement after cancellation and whether the final order became taker. Merely counting how many orders were placed passively is insufficient.
Round-trip cost across four execution combinations
Side by side, the four combinations show more than the 6 USDT spread between the extremes. The interesting rows are the two in the middle, where most real trading lands.
The second row deserves to be the default. Entry can afford patience; a risk stop usually needs certainty of execution. Budgeting from the first row assumes every exit fills passively — and the exit you most need to happen is the one least likely to.
| Entry / exit | Round-trip cost | Share of notional | Saved vs all-taker |
|---|---|---|---|
| Maker / maker | 4.00 USDT | 0.040% | 6.00 USDT |
| Maker / taker | 7.00 USDT | 0.070% | 3.00 USDT |
| Taker / maker | 7.00 USDT | 0.070% | 3.00 USDT |
| Taker / taker | 10.00 USDT | 0.100% | — |
The break-even fill rate: when chasing eats the saving
Let h be the fill rate on a resting order and c the adverse price paid when chasing after it does not fill, as a share of notional. The saving of posting versus crossing is h × (taker rate − maker rate) − (1 − h) × c.
Setting that to zero gives a break-even fill rate of h* = c / [c + (taker rate − maker rate)]. With a 0.03% fee gap and a 0.04% cost to chase, h* = 0.04 / (0.04 + 0.03) = 57.1%. Below a 57.1% fill rate, posting is the more expensive choice.
That threshold is higher than most people assume. It also explains why posting rarely pays in fast markets: those are exactly the conditions where fill rates are lowest and chasing costs most, so both variables move the wrong way at once.
Break-even fill rate h* = chase cost / (chase cost + taker rate − maker rate)| Adverse price when chasing | Break-even fill rate at a 0.03% fee gap |
|---|---|
| 0.01% | 25.0% |
| 0.02% | 40.0% |
| 0.04% | 57.1% |
| 0.08% | 72.7% |
Separate execution scenarios
Slow scaling can allow more time for maker fills; breakout entries and risk exits value certainty. Applying one execution assumption to every order understates tail cost.
Backtest passive, mixed and fully aggressive scenarios. Use actual fills to calculate average fee and slippage for each before comparing net cost.
Pre-execution check
Verify the venue maker/taker definition, account tier, rebates or promotions, and whether an immediately executable limit order becomes taker.
Do not accept price risk materially larger than 3 USDT to save a 3 USDT fee. The fee is visible; failed execution is an opportunity and risk cost.
Official sources and calculation boundary
The maker and taker definitions and the fee structure come from both venues' fee documentation. The break-even formula is derived here, not published by either platform, and both the 0.03% gap and the chase cost should be replaced with your own measurements.
Next checks in this series
Trading cost
The full seven-day cost of a 10,000 USDT perpetual position
Reproducible scenario guide
Liquidation and margin risk
BTC at 5x, 10x, 20x and 50x: how liquidation risk changes
Reproducible scenario guide
Liquidation and margin risk
How adding margin changes a BTC long liquidation estimate
Reproducible scenario guide
Frequently asked questions
Does post-only guarantee a maker fill?
It generally prevents immediate taker execution, but the order may cancel or remain unfilled. Confirm final order status and fees.
Can exchanges be compared using only taker fees?
No. Align account tier, funding, slippage, product availability and withdrawal path as well.
Why is the break-even fill rate so much higher than intuition suggests?
Because the fee gap is small and the chase cost is not. A single side differs by 0.03% between maker and taker, while chasing often costs 0.04% or more. You save a certain small amount and risk an uncertain larger one, so the fill rate has to be high to justify it.
How should I measure my own fill rate?
Record four things for every resting order: whether it filled, what fraction filled, how long it waited, and how far price moved after a cancel. Use the realised fill fraction as h and the post-cancel drift as c. Estimating from memory tends to overstate the fill rate, because unfilled orders are the easier ones to forget.