Direct answer
Adjusting unadjusted forex gain/loss means converting a basic profit-or-loss figure based only on price movement into a more complete figure that also reflects other effects that change realized trade results. In the context of average win loss, you typically want the gain/loss numbers you average to be calculated on the same basis for every trade, so the “win size” and “loss size” represent comparable outcomes.
Because the term “unadjusted” can mean different things in practice, the first step is to define what your “unadjusted” amount includes (usually only the price change). Then you apply a consistent adjustment rule that accounts for items that affect the net result, such as transaction-related costs and execution effects, according to the data you actually have.
Mechanics: define the inputs and the adjustment
-
Define “unadjusted gain/loss” An unadjusted forex gain/loss is a profit/loss amount computed from price change alone. It is often derived from the trade’s entry and exit prices, and (if used) the traded size. It excludes costs unless those costs were already embedded in the price series or in your calculation.
-
Decide what “adjusted” should include Common categories of items that can be excluded from unadjusted numbers include:
- Transaction costs (for example, spreads or commissions, depending on how you model execution)
- Swaps/rollover charges if your dataset spans multiple days
- Any other execution-related differences between the assumed and realized fill
You do not need to know the “best” approach; you need an approach that matches your reporting inputs.
- Apply a consistent adjustment method Use the same adjustment logic for every trade in your sample. A practical way to express this is:
- Adjusted gain/loss = unadjusted gain/loss + (net adjustments)
“Net adjustments” is the total of whatever components were missing from the unadjusted calculation, with signs applied consistently (costs reduce results; charges/credits affect results according to their direction).
- Recompute average win and average loss Once you have adjusted trade outcomes, compute:
- Average win: average of adjusted gains across the trades classified as wins
- Average loss: average of adjusted losses across the trades classified as losses
If your average win loss metric also separates “count” and “size,” keep the win/loss classification consistent with the adjusted numbers (for example, a trade that was slightly positive unadjusted could become negative after adding costs).
Example checks: verify the adjustment independently
You can perform consistency checks without needing predictions:
- Reconciliation check: ensure the sum of adjusted results across trades equals the sum of unadjusted results plus the sum of your net adjustments.
- Sensitivity check: if you can, compare averages computed with and without the adjustment; large differences indicate the unadjusted basis was materially missing components.
- Classification check: verify whether any trades change from win to loss (or vice versa) after adjustment; that can change average win loss even if the total stays similar.
These checks help confirm that the adjustment is applied uniformly and that average win and average loss reflect comparable net outcomes.
Relevant limitations and risks
- Ambiguity risk: “unadjusted” and “adjusted” are not universal accounting terms. Two datasets may use different conventions, leading to different averages even with the same method description. - Data availability: you can only adjust using components you have. If you lack transaction cost or rollover details, any “adjustment” becomes an assumption rather than a verified correction. - Time sensitivity: if costs or execution conditions changed during your sample, one adjustment rule may not reflect reality for all trades. - No guarantee of improvement: adjustment makes figures more consistent, but it does not imply future performance changes or that an average becomes “better.