Transaction Cost Modelling Guide

At a glance

Components: Spread, commission, slippage, impact, financing, borrow
Impact model: Scales with the square root of order size relative to volume
Key ratio: Cost per round trip divided by average winning trade
Rule of thumb: Above 15 percent, the strategy is probably not viable

Key takeaways

Costs are paid on every trade, winners and losers alike, so they subtract directly from expectancy rather than from profits.
The effective spread you actually pay is usually wider than the quoted spread, particularly for larger orders and in fast markets.
Market impact grows roughly with the square root of order size relative to average volume, which means costs rise faster than position size.
Financing, borrow fees, and currency conversion matter for multi-day and leveraged positions and are frequently omitted entirely.
The decisive test is cost as a fraction of the average winning trade; above roughly 15 percent, most strategies do not survive.

The components of trading cost

Component	Applies to	Typical magnitude
Half-spread on entry and exit	Every trade	Varies: 0.01% in index futures to 2%+ in small caps
Commission	Every trade	Per share, per contract, or zero with wider effective spreads
Slippage versus signal price	Market orders and stops	Grows sharply in fast markets
Market impact	Orders large relative to volume	Square-root scaling with size
Financing / margin interest	Leveraged positions held overnight	Annualised rate on debit balances
Borrow fee	Short positions	0.3% to 50%+ annualised
Currency conversion	Foreign-denominated trades	0.2% to 1% per conversion
Exchange and regulatory fees	Depends on venue	Small but non-zero, and per-trade

Practical cost models

Per-trade cost (one side):

   cost = (spread / 2)
        + commission_per_unit
        + impact

Square-root impact model:
   impact = c x sigma x sqrt(Q / V)

   c     = a constant, typically around 0.5 to 1.0
   sigma = daily volatility of the instrument
   Q     = your order size
   V     = average daily volume

Example: buying 20,000 USD of a stock with
   average daily volume 8,000,000 USD, daily vol 2%

   sqrt(20,000 / 8,000,000) = 0.05
   impact = 0.75 x 2% x 0.05 = 0.075%

   Plus half-spread of 0.02% -> about 0.095% per side
   Round trip: about 0.19%

Same strategy at 2,000,000 USD order size:
   sqrt(2,000,000 / 8,000,000) = 0.5
   impact = 0.75 x 2% x 0.5 = 0.75% per side
   Round trip: about 1.54% -> eight times the cost

A cost model suitable for most retail backtests.

The square-root scaling is the important insight: costs do not rise linearly with size. A strategy that works at small size may be entirely unviable at ten times the capital, which is why capacity is a real constraint even for individual traders in less liquid instruments.

Realistic parameters by market

Market	Spread cost	Commission	Typical round trip
Micro index futures	1 tick	Around 0.80 USD	0.02 to 0.05%
Large-cap US equities	0.01 to 0.05%	0 to 0.005/share	0.03 to 0.10%
Liquid index ETFs	0.01 to 0.03%	Usually zero	0.02 to 0.06%
Small-cap equities	0.3 to 2%	Varies	0.5 to 3%
Major FX pairs	0.5 to 1.5 pips	Often in the spread	0.01 to 0.05%
Liquid options	1 to 5% of premium	Per contract	2 to 10% of premium
Liquid crypto	0.02 to 0.1%	0.02 to 0.1% per side	0.1 to 0.4%
Illiquid crypto	0.5 to 3%	Same	1 to 6%

Approximate round-trip costs for modest retail-size orders.

The viability test

Cost ratio = round-trip cost / average winning trade

Strategy A: swing trading ETFs
   Round trip cost   0.05%
   Average win       3.2%
   Cost ratio        1.6%    -> comfortable

Strategy B: intraday equities
   Round trip cost   0.12%
   Average win       0.75%
   Cost ratio        16%     -> marginal; needs a strong edge

Strategy C: scalping with retail costs
   Round trip cost   0.08%
   Average win       0.15%
   Cost ratio        53%     -> not viable

Also compute annual cost drag:
   annual drag = round-trip cost x trades per year

   200 trades/year at 0.12% = 24% annual drag
   That figure must be exceeded before you earn anything.

The calculation that decides whether a strategy can work.

The annual drag calculation is the one most traders never perform, and it is decisive. A strategy trading 200 times a year with a 0.12 percent round-trip cost must generate more than 24 percent gross annually simply to break even. Very few strategies clear that bar.

Reducing costs

Trade less. The most effective lever by a wide margin. Longer holding periods amortise fixed costs across larger moves.
Use limit orders where the strategy permits. You collect the spread instead of paying it, at the cost of missed fills.
Trade liquid instruments. The difference between a large-cap and a small-cap round trip can be a factor of twenty.
Avoid the first and last minutes of the session, where spreads are widest, unless the strategy specifically targets them.
Size below the impact threshold. Keep orders under a small fraction of average volume, typically 1 percent or less.
Choose venues and fee tiers deliberately. Maker rebates and volume tiers can change the arithmetic entirely for active strategies.
Avoid unnecessary currency conversions by holding balances in the traded currency where practical.

Frequently asked questions

How much slippage should I assume in a backtest?

At minimum half the typical spread on each side, plus an impact term scaled by your order size relative to average volume. For strategies using market orders or stops in fast conditions, assume considerably more. The most reliable approach is to measure your own realised fills against the midpoint over several dozen trades.

What is market impact and when does it matter?

The adverse price movement caused by your own order consuming liquidity. It scales roughly with the square root of order size divided by average volume, so it is negligible for small retail orders in liquid markets and becomes the dominant cost once your order approaches a meaningful fraction of daily volume.

Do costs matter for long-term strategies?

Far less. A position trading strategy making ten trades a year at 0.1 percent round trip pays a 1 percent annual drag, which is easily absorbed. The same cost applied to 500 trades a year is a 50 percent drag, which no edge overcomes. Cost sensitivity is essentially a function of trading frequency.

How do I model costs for options?

As a percentage of premium rather than of notional, because option spreads are wide relative to their price. A one-cent spread on a 50-cent option is a 2 percent cost per side. Include per-contract commissions and assume worse fills on multi-leg structures, where you cross the spread on each leg.

Should I include financing costs?

Yes, for any leveraged position or any position held for weeks. Margin interest, futures roll effects, borrow fees on shorts, and currency carry all accumulate over time and can exceed the price move on a multi-week position. Omitting them is one of the most common gaps in retail backtests.

Test this idea before you trade it

Describe the rules in plain language and AlgoTrader AI turns them into a structured strategy blueprint with a configurable historical backtest, cost assumptions, and exportable code.

Build a backtest

Keep reading

Referenced by

Educational use only. This guide explains how a strategy works. It is not investment advice, not a recommendation, and no result described here is a forecast. Test any approach on historical and out-of-sample data, size positions conservatively, and never risk money you cannot afford to lose.