What “spread assumptions” means in forex
In forex, the spread is the difference between a quoted buy price and sell price for the same currency pair. Many analyses—especially those used to estimate performance—must assume what spread will be paid at the time of trading. “Spread assumptions” are those chosen inputs: the spread value (or rule) used in calculations, backtests, simulations, or decision thresholds.
Spread assumptions matter because they directly represent a cost of entering and exiting trades. If your spread assumption is too optimistic, your estimated results will typically be overstated; if it is too pessimistic, they may be understated. This matters even when other parts of the analysis are accurate, because spread affects the realized profit or loss by changing the effective entry and exit prices.
How spread assumptions work in calculations
A simplified way to see the mechanism is to translate a spread into an effective price disadvantage. When you buy, you often pay at (or near) the ask; when you sell, you often exit at (or near) the bid. If your model assumes a fixed spread, it effectively hard-codes the average transaction cost. If your model assumes “zero spread” or assumes a spread that is only reachable during special conditions, the model ignores the cost you would actually pay most of the time.
Spread assumptions can be expressed in several ways:
- A fixed spread value used for every trade.
- A time-based rule (for example, using a typical spread for certain hours).
- A volatility- or liquidity-linked rule (spreads widen when conditions deteriorate).
- An execution-based assumption (for example, how close the trade fills to the quoted price).
Any of these choices creates a specific assumption set. For every example or calculation, the key question is: “What spread did the model assume, and is that spread realistic for the intended trading conditions?”
Practical relevance: what decisions they affect
Spread assumptions influence more than final profit estimates. They can change:
- Risk sizing in cost terms: if spreads are higher, the same price move produces a smaller net outcome after costs.
- Trade frequency comparisons: with more trades, cumulative spread costs can dominate results.
- Strategy evaluation validity: two approaches that look similar before costs can diverge after costs if their assumed spreads differ.
- Break-even thresholds: even without predicting outcomes, you can estimate what price movement would be needed to cover the assumed spread and any other friction.
To keep the mechanics honest, always state the assumption explicitly: fixed or variable spread, the assumed magnitude, and whether the assumed spread corresponds to actual tradable fills at the modeled time.
Evidence or example (with explicit assumptions)
Consider two evaluation runs of the same basic idea, both using identical price data and timing rules. The only difference is the assumed spread.
Example setup (assumptions must be stated):
- Assumption A: spread is constant at 1 unit per trade.
- Assumption B: spread is constant at 2 units per trade.
- Each run executes 10 trades, with identical entry/exit price changes before costs.
Under these simplifying assumptions, Assumption B adds roughly an extra 10 units of total spread cost compared with Assumption A (one extra unit per trade, ignoring other frictions). The exact number depends on how your model maps spread to effective entry/exit prices, but the direction is consistent: higher assumed spread reduces net results.
This example illustrates why spread assumptions must match the conditions you claim to evaluate.
Material limitations and failure modes
Spread assumptions fail in several common ways:
- Condition mismatch: spreads often widen during high volatility, thin liquidity, or major news. A “typical” spread can be unreachable during stress periods.
- Execution mismatch: quoted spreads do not guarantee fills at the quote. If your model assumes you get the best available price while real fills are worse, costs will differ.
- Timing mismatch: using spread measured at one timestamp while trades are executed at another can introduce systematic error.
- Non-fixed costs: spreads can be dynamic (variable) and may change throughout the holding period; assuming a constant spread ignores this evolution.
- Historical-to-future gap: even if a relationship held historically, it does not guarantee the same spread behavior in the future.