Direct answer
Spread assumptions in forex describe how you estimate the transaction cost caused by the bid/ask difference (the spread) inside your calculations. In practical terms, a spread assumption is an input used to convert a “mid-price”-style view of the market into entry and exit prices, so cost effects are included in an analysis.
The key idea is separation: the mechanism you use to apply the spread is stable, but the spread itself is variable. If your assumed spread does not match the conditions you are trying to represent (actual market spread, execution quality, and additional costs), your results become unreliable.
Mechanics: definition, inputs, and sequence
What “spread” means
Forex prices are typically quoted as two numbers: a bid price and an ask price. The spread is the distance between them. Trading logic uses different sides of the quote depending on direction: buying uses the ask price; selling uses the bid price.
Because many charts and research views show a single “mid” price concept (roughly centered between bid and ask), spread assumptions define how to translate that into the bid and ask prices you would use for calculations.
What “spread assumptions” do
A spread assumption is a chosen rule for modeling spread effects. Commonly, an analysis will:
- Start with a reference price series (often mid, or a price series treated as mid).
- Apply an assumed spread (fixed, variable by time, or based on a simplified schedule).
- Derive entry and exit execution prices for each trade direction.
- Compute P/L using those derived execution prices.
Even if you do not label it as such, this is the same mechanism whenever you adjust theoretical prices to account for the difference between bid and ask.
Typical inputs
To apply spread assumptions in a transparent way, you need at least these inputs:
- Reference price: what your analysis treats as mid (or as another baseline).
- Assumed spread size: the amount you add/subtract to approximate ask vs bid.
- When the spread is applied: at entry only, at exit only, or at both.
- Units and conversion: how the spread is represented (for example as an absolute price distance or as pips), and how it is converted into the profit/loss calculation.
- Direction mapping: buying uses ask; selling uses bid.
If any of these are left implicit, different people can recreate different models and still believe they used the “same” spread assumption.
Example calculation (with explicit assumptions)
Assume a reference price at a certain timestamp is 100.000 (treated as mid for the purpose of the example). Assume a spread of 0.002 (so half-spread is 0.001).
- If you buy, the modeled entry execution price is mid + half-spread = 100.000 + 0.001 = 100.001.
- If you sell, the modeled entry execution price is mid − half-spread = 100.000 − 0.001 = 99.999.
For exit, you repeat the same mapping using the exit reference price at that time. Costs from spread are therefore embedded in both entry and exit when the model adjusts both sides.
Important: this example does not claim real market behavior. It only shows the stable arithmetic sequence that spread assumptions implement.
Evidence or example scenario: why assumptions can break
Spread assumptions are often used in research contexts where you simulate results from historical data. A failure mode is mismatch between the model and real execution conditions. Common sources of mismatch include:
- Spread variation over time: real spreads can widen unpredictably, while an analysis might use a constant value.
- Execution differences: actual filled prices can differ from the reference series due to order-book dynamics.
- Additional costs beyond spread: fees, commissions, or financing effects can exist and are not captured if you only model bid/ask spread.
- Timing issues: if your model applies the spread at the wrong timestamps (for example, using a spread assumption that does not correspond to the moment execution would occur), the cost embedding becomes inconsistent.
A material limitation
A material limitation of simplified spread assumptions is that they treat spread as an input rather than as a stochastic variable that depends on liquidity and conditions. When the market enters periods of higher volatility or lower liquidity, the spread may behave differently than your assumption, and the modeled cost impact no longer represents what actually happens.
Another failure mode: double counting or omission
If your reference price series already reflects bid/ask (for example, if it is not mid), then applying a spread adjustment again can double count costs. Conversely, if your reference is truly mid but your model forgets to apply the spread on entry or exit, costs may be understated.
Verification and next questions
How to independently verify the facts in your own model
You can verify spread assumptions by reproducing the logic with explicit, checkable steps:
- Write down your reference price definition (mid or not).
- State your assumed spread rule and units.
- Derive the execution prices for both buy and sell directions.
- Confirm whether spread adjustments are applied at entry, exit, or both.
- Recalculate profit/loss using those derived prices.
If you cannot clearly specify items (1)–(4), the spread assumption is not fully defined, and comparisons across analyses become unreliable.
Questions to ask before trusting results
- What does the price series represent (mid-like or side-specific)?
- Does the model include only spread, or also other execution costs?
- Is the spread assumption fixed, time-varying, or schedule-based?
- Are the timestamps consistent with when entries/exits would occur?
These questions help you test whether your spread assumptions represent the mechanism accurately, even though outcomes will still vary with real market conditions and execution quality.