The Hidden Round-Trip Cost of Crypto-Futures Trading: Fees + Funding, With Real Numbers

Most traders obsess over entries and exits and treat costs as a rounding error. On crypto futures, costs are not a rounding error. Fees and funding are structural, they hit every position, and they compound with how often you trade. A strategy that looks profitable on the chart can be flat or negative once the real round-trip cost is subtracted. This post does the arithmetic honestly so you can see where the money actually goes.

The three costs that hit every futures position

Every perpetual-futures trade carries three deductions, and only one of them is obvious:

• Entry fee — charged when you open. Taker (market) orders cost more than maker (limit) orders.
• Exit fee — charged again when you close. Same taker/maker split.
• Funding — a periodic payment (typically every 8 hours) exchanged between longs and shorts on perpetuals. If you hold across a funding timestamp, you pay it or receive it depending on your side and the sign of the rate.

Spot traders only think about the first two. Perpetual traders live with all three, and the third one scales with holding time, not trade count. Both add up in ways that are easy to underestimate.

Worked example: one clean round trip

Take a $10,000 notional long on BTC perps. Assume a common taker fee of 0.05% per side and you cross the spread both times.

• Entry taker fee: 0.05% of $10,000 = $5.00
• Exit taker fee: 0.05% of $10,000 = $5.00
• Round-trip fee: $10.00, or 0.10% of notional

Now add funding. Suppose the rate is a fairly ordinary 0.01% per 8h and you hold the long for 24 hours, crossing three funding timestamps:

• Funding paid: 0.01% x 3 = 0.03% of $10,000 = $3.00

Total round-trip cost: $13.00 on $10,000, or 0.13% of notional. Your trade has to move 0.13% in your favor just to break even before you have made a single cent. On a $10k position that is a $13 hole you start in.

That still sounds small. The problem is what happens when you repeat it.

Frequency is the multiplier

The 0.13% is per round trip. Your account does not care about one trade, it cares about the sum over a month or a year.

• 10 round trips/month at 0.13% = 1.3%/month in costs = roughly 15.6%/year of notional turned over, before funding variance.
• 1 round trip/day (about 20 trading days/month) = 2.6%/month = ~31%/year.
• 5 round trips/day = 13%/month = a cost load that most edges simply cannot outrun.

This is why high-frequency discretionary trading quietly bleeds accounts even when the win rate looks respectable. The edge per trade is fixed and small, the cost per trade is fixed and small, but frequency multiplies both, and costs are the part you pay with certainty. A strategy with a genuine edge can still lose money if it trades often enough that fees eat the edge. If you want to see whether your edge actually survives costs, feed real, cost-inclusive results into a profit-factor calculator rather than counting gross wins.

Two things that quietly make it worse (or better)

Taker vs maker. Using limit orders that add liquidity can cut or even reverse the fee side. If maker fees are 0.02% and taker fees are 0.05%, switching both legs to maker drops your round-trip fee from 0.10% to 0.04%, a 60% reduction on the fee component. Over a year of frequent trading that difference alone is large.

Funding sign and holding time. Funding is not always a cost. If you are short while funding is positive, you receive it. But most retail traders are long in bull phases exactly when funding is most positive, so they pay. The longer you hold, the more funding timestamps you cross, so a swing trade held for a week can accumulate far more funding drag than the fees themselves. Before you size a multi-day hold, it is worth checking the current funding-rate environment, because an elevated rate can turn an 8-hour scalp thesis into a losing carry.

What this means for how you trade

None of this argues against trading futures. It argues for being clear-eyed about the hurdle:

• Your break-even is not zero. It is round-trip fees plus expected funding for your holding period. Build that into your target, not your hope.
• Trade less to keep more, or trade cheaper. Fewer, higher-conviction trades face the cost load fewer times. If you must trade often, maker orders and shorter funding exposure are the levers.
• Size your stop to your edge, not the other way around. A trade whose realistic target barely clears the cost hurdle is not worth the risk. Running the numbers through a risk/reward calculator before entry keeps you from taking trades where costs eat the entire reward.
• Judge results net, always. A track record that quotes gross moves is telling you a story that costs will not honor. This is why every signal in the public Ezath track record is timestamped and hash-chained, including the losers, so the record you audit is the record you would actually have lived.

The uncomfortable summary: a 0.13% per-trade cost sounds trivial and is trivial once. Repeated a few hundred times a year, it becomes one of the largest, most predictable line items in your P&L. The traders who last are not the ones who ignore it, they are the ones who price it in before they click.