Average True Range (ATR) summarizes recent bar ranges; it does not predict direction or set a universally correct stop. A researcher may choose a multiple of ATR for a stop distance, then derive a hypothetical quantity from a separate loss budget. This is a sizing method built around assumptions, not an edge or a promise that realized loss will stay below the planned amount.
Define which ATR you mean
True range for a bar is commonly the greatest of high minus low, absolute high minus previous close, and absolute low minus previous close. ATR averages those ranges; smoothing and warm-up conventions can differ. TradingView documents Wilder-style RMA as its default ATR smoothing, while the chart’s smoothing setting can select other methods. Record the lookback, smoothing, price units, completed-bar rule and interval, and verify initialization and warm-up in the implementation you use. ATR percentage divides ATR by the current close and multiplies by 100; it helps compare price movement across assets or price levels, but it is not a forecast or stop recommendation. A $1,000 ATR on one BTC timeframe is not the same percentage movement as a $1,000 ATR on another price series. TradingView’s Pine Script v6 documentation identifies true-range built-ins, and its reference supports ta.atr().
Work from a risk budget to a quantity
Assume a fictional BTC entry of $60,000, a completed-bar ATR of $1,000, and a researcher-chosen two-ATR stop distance. The planned stop is $58,000, so price-distance risk is $2,000 per BTC. If the hypothetical account is $10,000 and the chosen risk budget is 0.5%, the budget is $50. CME Group’s general futures position-sizing guidance also starts from the dollar amount between entry and stop and the account amount a trader chooses to risk; that is a calculation framework, not an endorsement of this example or a crypto rule.
Ignoring costs, $50 ÷ $2,000 gives 0.025 BTC. Now suppose the study explicitly budgets another $20 per BTC for combined entry/exit costs. The cost-adjusted estimate is $50 ÷ ($2,000 + $20) = 0.02475 BTC. If the venue’s size increment were 0.001 BTC, round down to 0.024 BTC: the modeled stop-distance loss is $48, plus an assumed $0.48 in costs, or $48.48 total. All prices, costs, risk fraction and size step here are fictional inputs. The calculation assumes a linear one-unit contract and a fill at the stop price; a gap, partial fill or larger cost can exceed the budget.
What changes when the stop distance changes?
Hold the same $50 budget, $1,000 ATR, assumed $20 cost per BTC and 0.001 BTC step constant. The raw quantity is $50 divided by (the ATR multiple × $1,000 + $20). Rounding down gives 0.049 BTC at 1× ATR, 0.032 at 1.5×, 0.024 at 2× and 0.016 at 3×. These rounded quantities model $49.98, $48.64, $48.48 and $48.32 respectively, including the assumed cost. The arithmetic shows the inverse sizing relationship; it does not identify a best multiple. Treat the multiple as a research choice, freeze it before evaluation, and compare predeclared alternatives on later data rather than selecting the most attractive historical result.
For a short position the stop is above entry; the quantity formula uses the absolute entry-to-stop distance either way. For a contract with a multiplier, inverse settlement or a different quote currency, calculate dollar loss per contract using its specification rather than treating units as coins. Include commissions, spread, adverse slippage and funding where relevant. Costs consume part of a fixed loss budget before a price stop is reached.
Make the stop rule causal
If ATR uses a completed candle, only update the stop after that candle closes and state when the new level becomes active. A stop updated using a candle’s final range but treated as active earlier inside that same candle uses information from the future. State whether a volatility expansion can widen the stop, whether it can only tighten, and how a gap through the level is modeled. A larger ATR distance with the same dollar risk budget necessarily implies a smaller quantity under this method.
A fixed multiple does not make every trade equally safe. Volatility can rise after entry; a gap can skip the stop; and leveraged exposure can approach liquidation before an order fills. Position size and required margin are separate calculations: reducing the displayed leverage does not change the price-distance loss for a fixed linear quantity. Verify the exact contract, maintenance-margin tier, trigger reference and order type before interpreting either number. A volatility-based stop is a research rule, not a guarantee.
If a rule uses ATR from a completed candle, keep its signal timestamp and next eligible execution separate in a backtest. Do not use the candle’s final range to place a stop earlier within that same candle. Keep the ATR lookback, smoothing, stop multiple, gap behavior, cost model and quantity step in the experiment record; changing several after inspecting results makes the comparison hard to interpret and can overstate the evidence.
SuperQuant includes ATR calculation code for engine use, but ATR-based position stops are not exposed as an arbitrary user-defined strategy builder in the current interface. The bundled histories are synthetic and the simple execution model is not venue execution. Compare position sizing with stop versus liquidation, then try the local position-size calculator using clearly stated assumptions. This guide is educational, not personal advice.