Understand the 2% risk rule with a EUR/USD sizing example, costs and drawdown calculations. See why a planned loss budget is not a guaranteed cap.
The 2% rule is a way to set a planned loss budget per trade, not a promise that a loss cannot exceed 2%, and not a method that makes a strategy profitable. The percentage applies to account equity—not to the deposit margin or the face value of the position. CME's educational explanation treats 2% as an arbitrary threshold, not a universal optimum. CME: The 2% Rule.
With current equity of USD 10,000, a chosen risk fraction of 2% gives a planned budget of USD 200. That is an example, not a recommendation for your account. An account with existing losses, several correlated positions or money needed for living expenses may not support that budget at all.
Keep three figures separate: equity is the account's current net value, margin is collateral required to open a position, and planned risk estimates the loss between the intended entry and exit plus allowed costs. A low margin requirement does not imply a small possible loss.
The same percentage can produce very different position sizes when stop distance, costs or minimum trade size change. These reproducible examples explain the relationship; they are not trading signals.
Example only: USD account, EUR/USD, 100,000 EUR per standard lot, USD 10 per pip per lot, minimum and increment 0.01 lot. The per-lot cost allowance covers commission and execution deviation not already counted in the price distance. Do not count costs twice. It is neither a broker quote nor a slippage limit.
This is a budget calculation under explicit assumptions, not a guaranteed loss cap. Inputs stay in your browser; no order is sent.
| Equity USD | Risk % | Stop pips | Cost USD/lot | Lots | Est. loss USD |
|---|---|---|---|---|---|
| 10,000 | 2 | 30 | 10 | 0.64 | 198.40 |
| 10,000 | 2 | 50 | 10 | 0.39 | 198.90 |
| 10,000 | 2 | 30 | 30 | 0.60 | 198.00 |
| 100 | 2 | 30 | 10 | 0.00No position | 0.00 |
Each row is independent and assumes USD 10 per pip per lot. Row two changes only the stop; row three only costs; row four the account size. The final row cannot cover the minimum lot: zero means no position.
This is a model, not market history or a backtest. Each trade loses exactly the selected fraction of current equity; no additional fees, gaps or cash flows are included. No probability of a loss sequence is implied. Remaining fraction = (1 − r)ⁿ; gain required to recover = 1 / remaining fraction − 1.
| Losses | 0.5% per trade | 1% per trade | 2% per trade | Recovery gain: 2% model |
|---|---|---|---|---|
| 5 | 97.52% | 95.10% | 90.39% | 10.63% |
| 10 | 95.11% | 90.44% | 81.71% | 22.39% |
| 20 | 90.46% | 81.79% | 66.76% | 49.79% |
| 30 | 86.04% | 73.97% | 54.55% | 83.32% |
The first three percentage columns show capital remaining, not returns. Thirty losses of 2% leave 54.55%, rather than a 60% loss. Recovering requires an 83.32% gain, not 45.45%.
With USD 500 equity, a 2% budget, a 30-pip stop and USD 10 cost per lot, 0.01 lot fits the budget. At USD 100 equity, the USD 2 budget cannot even cover the USD 3.10 estimated loss of 0.01 lot. More leverage does not remove this minimum-unit constraint.
The USD 10 pip value comes from 100,000 × 0.0001. For a non-USD account, convert pip value from the quote currency into the account currency. Gold and other products require their own contract specifications; do not copy EUR/USD parameters.
Long EUR/USD and long GBP/USD can both be exposed to a stronger dollar. Two positions risking 2% each are not automatically diversified. Review existing and proposed exposure together; correlations vary, and summed stop budgets are not a worst-case cap.
Let R represent planned price risk. Assume a 40% win rate, an average 2R win and an average 1R loss. Before costs, expectancy is 0.4×2 − 0.6×1 = 0.2R. An average cost of 0.15R leaves 0.05R; a cost of 0.25R changes it to −0.05R. These are hypothetical inputs, not observed strategy statistics.
Risk percentage scales outcomes; it cannot turn negative expectancy positive. Real evaluation also requires adequate samples, out-of-sample testing, actual costs and the loss distribution—not just win rate or one attractive example.
Definitions and order mechanics: CME and Investor.gov links below. All chart values are calculated from the formulas shown here.
The following sections explain assumptions, order limitations and verification sources.
Assume a USD-denominated account, EUR/USD, a contract of 100,000 EUR per lot, a pip of 0.0001 and therefore USD 10 per pip per lot. Assume the broker permits 0.01-lot steps. Check the actual contract specification; these assumptions do not transfer unchanged to gold, indices, JPY pairs or a differently denominated account.
Now add an illustrative USD 10 per lot for round-trip commission and execution allowance, excluding anything already counted in the entry-to-exit distance. The estimate becomes 200 / (300 + 10) = 0.6452 lots; rounding down gives 0.64 lots, with an estimated loss of USD 198.40. This is a hypothetical calculation, not a broker's fee quote or a guarantee about slippage.
If the minimum tradable size would exceed the budget, skip the trade or use a suitable product; do not move the stop closer merely to force the numbers to fit. The position-sizing guide covers the general process. This page focuses on the limits and consequences of choosing a fixed percentage.
In a simplified model, each completed trade loses exactly 2% of the equity remaining immediately before that trade. With no deposits or withdrawals, the balance after 30 such losses is 10,000 × 0.98^30 = USD 5,454.84. The drawdown is 45.45%, and returning to USD 10,000 would require an approximately 83.32% gain from that lower balance.
If instead each trade loses a fixed USD 200 based on the initial balance, 30 losses leave USD 4,000. These are different sizing rules. Neither calculation predicts the probability of a losing streak or establishes that the remaining capital is adequate. Real losses, fees and execution can differ from the model.
A stop order can be executed at a price different from its trigger price. A stop-limit order has a different trade-off: the price restriction may prevent an exit from executing. Investor.gov explains these mechanics for securities; for an FX or CFD account, verify the broker's actual order terms and product rules. Investor.gov: Stop orders.
Before trading, record the planned loss, estimated costs and a worse-execution scenario. Do not label the first figure “maximum possible loss” unless a specific contractual protection actually establishes that limit and you have checked its conditions.
Three open positions, each with a planned risk of USD 200, add up to USD 600 of planned exposure—not USD 200. Trades with a shared currency driver may lose together. This simple sum is not a worst-case guarantee either, because exits can slip. Decide on an account-level limit as well as a per-trade budget, without treating diversification as certainty.
For the same equity and stop assumptions, 2% allocates twice the planned loss budget. There is no evidence here that it is the right choice for you. Choosing a smaller percentage can reduce the size of a modelled loss, but it cannot turn a negative-expectancy strategy into a positive one.
Define the basis explicitly and use it consistently. This example uses current equity, including unrealised gains and losses. If you use balance while ignoring open losses, the budget may overstate the funds available to bear further risk.
No. Position sizing does not establish broker identity, client-money protection or withdrawal rights. Complete the official licence checks separately.
Record equity, the chosen percentage, contract size, pip value, stop distance, size step, costs, open exposure and the final rounded-down size. After closing, compare planned with actual loss. Investigate the difference before assuming the calculation or execution model was correct.