Cost Updated 2026-09-12 By Ivo Renner

Binance TradFi perp fees: the four costs of a trade

A round trip in SPYUSDT has four separate cost lines. Binance publishes TradFi Futures maker and taker rates by VIP tier, while promotions and contract-specific funding terms can change. This guide shows where to read those inputs and how to combine them without opening a position to test them.

Fee calculation and sources checked 11 September 2026. Recheck the live fee tab, promotion terms and contract funding panel before using the formulas.

Find the rate that applies now

Binance states in its TradFi Perps FAQ that discounted trading fees apply, and its public Fees & Transactions Overview includes a dedicated TradFi Futures tab. Select that tab, then read the maker and taker cells in the row for your VIP tier. That published row is the starting point; any promotion changes it only according to the promotion's own eligibility, products, dates and discount terms.

No one percentage fits every reader because the table is tiered and a qualifying promotion can alter the displayed schedule. Record the page date, your VIP row, and the exact promotion terms you applied. That makes the estimate reproducible when the schedule changes.

Cost one: the trading fee

The trading fee is charged on notional, not on margin. Post 20 USDT of margin at 10× to open a 200 USDT position and the fee is assessed against 200, not 20. Two positions with the same 200 USDT notional incur the same fee rate and fee amount even if their leverage differs. Higher leverage changes the margin committed and liquidation distance; it does not by itself make two positions equivalent in risk.

The fee has two rates, and which you pay is determined by what your order does to the book, not by what you intended.

Maker is the rate used when an order rests in the book and adds liquidity before it fills. Taker is the rate used when an order removes liquidity immediately: a market order, or a limit order priced aggressively enough to execute on arrival. Read both cells from the TradFi table because promotions can treat the two legs differently.

Two practical consequences. A "post only" order type exists precisely so that an order which would have crossed the spread is cancelled rather than executing at the taker rate; if you are not in a hurry, that is the cheapest configuration change available to you. And, more important here: a stop-loss that triggers as a market order is a taker order, and so is a liquidation. Your exit is frequently the expensive leg even when your entry was patient.

To check what you were actually charged, use fills that already exist in your trade history or export. For each fill, divide the fee by fill price × filled quantity, with consistent fee-asset conversion if necessary. Display precision varies, so do not assume a 5 USDT fill must show 0.00; use the most precise downloadable record available and keep its timestamp.

The public TradFi table remains the source for planning a new trade.

Cost two: the spread

The spread is the gap between the best bid and the best ask. It is not normally itemised as a fee: an immediately executable buy starts at the ask and an immediately executable sell starts at the bid. Its importance varies with the contract, time and order size, so compare it with the other cost lines rather than assuming it is always the largest.

Using the contemporaneous mid-price as the benchmark, each market leg costs roughly half a spread when the book and price are unchanged. Entry plus exit therefore costs about one spread in total. Any execution beyond the best displayed quote belongs in the separate slippage line below; counting the fill's full distance from the mid as slippage would count the spread twice.

The size of that handicap varies enormously across the 72 contracts on the roster, and the specification tables give you no clue whatsoever. Every contract shares the same tick size of 0.01 and the same 5 USDT minimum notional. SPYUSDT and FWDIUSDT occupy identical rows in every published field. They are not remotely the same instrument to trade.

On the deep end — SPYUSDT, QQQUSDT, the mega-cap single names — a lot of people want to trade the same thing around the clock and the book reflects it. On the thin end the picture differs in kind, not degree: the microcaps Binance itself held to 10× while listing everything alongside them at 25×, and the narrow thematic ETFs, can show visible top-of-book depth that is small relative to a position you would consider unremarkable. There the spread is not a rounding error against your trading fee. It can be a multiple of it.

Nobody publishes this. The only way to know is to look at the book on your contract, at the hour you intend to trade it, before you size anything. Two questions answer it: how wide is the bid-ask gap as a fraction of price, and how much size sits within a few ticks of the top? Ask them at 03:00 UTC as well as 15:00 UTC — on an instrument that trades 24/7 against an underlying that does not, the book at three in the morning is a different book.

Cost three: funding

Funding is a payment between longs and shorts, not a trading fee taken by the exchange. If the rate is positive, longs pay shorts; if negative, shorts pay longs. It is calculated on position notional when a contract reaches a funding settlement. The interval is contract-specific: some TradFi contracts use eight-hour settlements, while Binance changed several contracts to four-hour settlements in September 2026. Read the next funding time, interval and cap on the contract panel before counting how many settlements a holding period crosses.

Two features of the TradFi specification make funding behave differently here than on a crypto perpetual, and both are in the launch announcements.

The interest rate component is set to 0%. On Binance's crypto perpetuals, funding combines a premium term with a fixed interest term. Zeroing the interest term means funding on these contracts is purely a measure of positioning imbalance. It is not a financing charge being passed through to you; it is the price of being on the crowded side.

The funding cap and floor are also contract-specific and subject to announced changes. For example, Binance's 10 September 2026 notice moved nine named TradFi contracts from eight-hour funding with a ±2.00% cap/floor to four-hour funding with ±1.00%. Other launch notices specify their own initial interval and limits. Do not transfer one contract's values to another; use the current contract panel and the latest notice for that symbol.

The point that matters for cost accounting is that funding is the only cost line with a clock on it. Entry and exit fees occur on fills; funding can recur at every scheduled settlement for as long as the position is open.

A hypothetical eight-hour contract has 21 scheduled settlements in seven days, while a four-hour contract has 42.

Count the actual settlement timestamps your position would cross.

Where the two cross is not a matter of opinion, and it does not need a Binance number. Funding overtakes your round-trip trading fee after k > 2ft / r settlements — a ratio between your taker rate and the funding rate on your contract, both of which you can read. The arithmetic and a table of break-even horizons are below.

Before estimating an overnight hold, record the contract's current rate, your paying or receiving side, its next settlement time, interval and cap/floor. The funding cost calculator turns those assumptions and a holding period into a USDT figure, and how funding actually works on equity perps explains the mechanism.

Cost four: slippage, and the cost of being liquidated

To keep the accounting clean, define slippage as execution beyond the best quote available when the order arrived: for a buy, fill price minus best ask; for a sell, best bid minus fill price. The bid-ask gap is already counted in the spread line. Slippage appears when an order consumes depth beyond the top of book or the quote moves before execution.

On a market order in a thin contract this is not a subtlety. Your order eats through resting size at successively worse prices until it is filled, and your average execution can sit several ticks from the price you saw. The remedy is not clever: use limit orders where you can tolerate not being filled, and size to the book rather than to your conviction. Our position size planner sizes from a stop distance, but does not read order-book depth or guarantee sufficient liquidity for that size.

Liquidation removes your control over the exit and can add liquidation charges. Isolated margin normally separates the collateral assigned to that position; cross margin exposes the eligible shared futures collateral pool. Neither label guarantees a maximum total loss equal to the initial margin: account for funding, trading and liquidation charges, manual or automatic margin additions, and the applicable deficit rules. Compare margin modes and leverage tiers and liquidation.

Treat the probability of liquidation as a cost line, because it is one. A strategy that survives its fee and funding arithmetic but liquidates once a month has not survived anything.

The total cost of a round trip

Here is the whole thing as one calculation. Every row is a formula rather than a figure, because the inputs are yours. Fill in the four values you can look up — your maker and taker rates from the fee schedule, the current spread from the order book, the current funding rate from the contract page — and the arithmetic is straightforward.

Write N for your position's notional in USDT, fm and ft for your maker and taker rates, s for the spread as a fraction of price, r for the funding rate per settlement, and k for the number of settlements you hold through.

Paper-only example: assume two taker fills, N = 1,000 USDT, taker 0.03%, a stable full spread of 0.04%, funding paid at 0.005% per settlement, and k = 3. The estimate is 1,000 × (2×0.0003 + 0.0004 + 3×0.00005) = 1.15 USDT, plus slippage beyond the best quotes. This assumes unchanged notional and spread on both legs; these are hypothetical inputs, not current Binance rates.

The percentage cost relative to notional does not fall when leverage rises. Higher leverage increases N relative to posted margin, so the same percentage costs consume more of that margin. Funding is the recurring term; entry fee, exit fee, spread and any execution slippage occur on the relevant fills.

When funding overtakes the trading fee

The point at which funding paid equals trading fees is a ratio. The table below is a hypothetical both-legs-taker example; substitute the rates and settlement interval shown for your contract.

For a round trip in which both entry and exit remove liquidity, the trading fees are

fees = N × 2ft

and funding held across k settlements is

funding = N × r × k

Set them equal and N cancels out, which is the whole trick. Position size is irrelevant to the crossing point; so is leverage, since leverage only changes N. What is left is

2ft = r · k  →  k = 2ft / r

Funding has overtaken your round-trip trading fee once you have held through more than 2ft / r settlements. You need one quantity to use this: the ratio of your taker rate to the funding rate per settlement on your contract. Your taker rate is on your own fee schedule. The funding rate is on the contract page. Neither of them has to be published by us, and neither of them can be stale here, because the table below is expressed entirely in the ratio.

Use the ratio column to find k, then convert that settlement count with the interval displayed for the contract. At the same rate ratio, a four-hour schedule reaches the same k in half the calendar time of an eight-hour schedule. If the rate changes between settlements, sum the rates actually paid rather than treating the latest print as constant.

Three honest qualifications on the table.

r is not constant. It is re-set every settlement and it changes sign. What the formula really needs is the average rate you actually pay over your holding period, which you only know afterwards. Use the current rate as an estimate, and check the contract's funding history for the range rather than the last print.

The spread is not in it. The crossing point compares funding against trading fees only. Spread is paid on entry and exit, and on a thin contract it can exceed both. It moves the total, not the crossing point.

The sign matters. If you are on the receiving side, funding reduces the net total instead of overtaking fees as a cost. Trading fees, spread and any slippage still remain. The funding cost projector handles the sign, interval and holding period together; how funding works on equity perps explains the mechanism.

Check a discount without placing test trades

The public TradFi Futures table is the primary source for the published maker and taker rates. If Binance advertises a discount, open the linked promotion and record its exact dates, eligible tiers, eligible contracts and whether it changes maker fees, taker fees or both. A banner by itself does not show that every reader qualifies, and an August screenshot does not establish today's terms.

If your account already has completed fills, the downloadable history can confirm the effective rate that applied at that time. It cannot isolate a TradFi-versus-crypto discount unless the records are comparable, and there is no reason to create new exposure for that comparison.

1. Filter existing fills by symbol, date and order role. Compare maker with maker or taker with taker; do not mix them.
2. Reconstruct notional per fill. Use fill price × filled quantity, then sum partial fills belonging to the same execution if useful.
3. Normalize the fee asset. If fees are recorded in different assets, convert them using a documented timestamp and source before comparing ratios.
4. Calculate the effective rate. Divide the total fee by the corresponding total notional. Keep more precision than the rounded interface display provides.
5. Match the result to dated terms. Check the VIP tier and promotion that applied at the fill timestamp. A difference across dates can reflect a schedule change rather than a product discount.

Estimate costs without trading

Small absolute charges may be shown with different precision across interface views and downloadable records. Do not assume a 5 USDT fill must display 0.00, and do not choose a larger position to force a visible decimal. For an existing fill, use the most precise export available. For a future trade, multiplication is enough.

Suppose, only for a paper estimate, that the public row and an applicable promotion leave you with a 0.03% taker rate. A 5 USDT taker fill would have an estimated fee of 5 × 0.0003 = 0.0015 USDT. At 200 USDT notional the same assumed rate gives 0.06 USDT. Leverage does not change either fee when notional stays fixed.

For spread, take a snapshot of the best bid and ask without placing an order. If the bid is 99.98 and the ask is 100.02, the spread is 0.04 / 100.00 = 0.04%. Under the unchanged-book assumption, crossing on entry and exit costs about one full spread, or 0.40 USDT on 1,000 USDT notional.

Depth beyond the top quote determines additional slippage; estimate it from the visible book for the contemplated size.

For funding, use the rate, next settlement time and interval shown on that contract. Multiply notional by each assumed rate for every settlement the proposed holding period crosses. If you receive funding, enter it as a negative cost. Because the rate can reset, a range based on recent funding history is more informative than one fixed forecast.

Cost structure matters most when the expected move is small relative to spread, fees and slippage. Compare the complete round-trip estimate with the trade thesis before deciding whether the contract and order type are suitable; the calculation itself does not require a rehearsal position.