Position sizing is the decision about how much of an instrument or strategy to hold. It converts a trading idea into shares, contracts or portfolio exposure. The same price move can be tolerable or damaging depending on that quantity, so a trading signal alone does not specify a complete trading strategy.
A sizing rule needs an objective and constraints. A rule based on a possible exit loss answers a different question from one targeting portfolio volatility or long-run growth. None establishes that the underlying trade has positive expected returns.
Turn a loss budget into whole shares
Consider a hypothetical long stock trade, quoted in USD per share. The trader assigns USD 1,000 as the trade's loss budget, assumes entry at USD 100, and assumes an exit at USD 95 if the trade goes against them. For now, both fills are stipulated and all costs are zero.
The assumed loss per share is USD 5. With only whole shares allowed, the risk-based quantity is
Here is the loss budget, is the assumed entry price, and is the assumed losing exit price. The floor rounds down so an extra indivisible share does not push the conditional loss over the budget. CME's position-sizing explanation likewise connects quantity with the chosen dollar risk and distance to the stop, including an instrument's tick value when relevant.
The entry notional is USD 20,000, calculated as 200 shares times USD 100. That is a different quantity from the USD 1,000 loss budget. The budget is not necessarily the trader's account capital; neither the portfolio weight nor the leverage ratio can be inferred without a capital denominator.
The assumed exit is doing real work
A stop-loss rule may motivate the USD 95 assumption, but it does not ensure that fill. For US stock stop orders, FINRA explains that the triggered order becomes a market order and can execute far from its stop price. A stop-limit order adds a price condition while introducing the possibility of no execution.
Keep the 200 shares fixed and compare two hypothetical complete fills:
| Cash movement | Exit at USD 95 | Exit at USD 90 |
|---|---|---|
| Purchase cash paid | USD 20,000 | USD 20,000 |
| Sale cash received | USD 19,000 | USD 18,000 |
| Gross trading loss | USD 1,000 | USD 2,000 |
The worse fill produces twice the planning budget in gross loss, before any fees. It could follow a price gap or poor available liquidity; the example does not assign either a probability. Chan's discussion of discontinuous prices makes the same risk distinction: deciding to exit is different from obtaining the earlier price.
Original hypothetical cash-equity scenarios, with a fixed quantity of 200 whole shares and no costs. The line compares possible fills; it is not a price path or forecast.
Check what the quantity can actually support
Explicit transaction costs consume part of the budget. If the same example adds a stipulated USD 0.25 of round-trip charges per share, the conditional loss per share becomes USD 5.25. Rounding down gives 190 shares, with conditional loss USD 997.50. The next share would raise it to USD 1,002.75. This cost allowance still cannot guarantee the exit fill. Do not add slippage twice if it is already represented by the assumed fill prices.
The resulting quantity is also subject to available capital or financing, concentration limits and executable liquidity. A cash purchase must be affordable including entry charges. A tight assumed exit distance can generate a large calculated quantity while making those other constraints more binding. A zero distance makes this formula undefined; it does not justify an unlimited position. Derivatives require their own payoff and contract-unit mapping rather than reusing a stock's dollars-per-share distance.
At portfolio level, correlation matters: several individually sized positions can lose together. Volatility targeting uses a risk forecast rather than a single exit distance. For other allocation rules, Alpha Sizing Heuristics compares expected-return and volatility-based sizing, then scales the allocations to a portfolio target; its preview is public and the full lesson requires membership.
In a backtest, specify the sizing information available at the decision time, the integer or lot rounding, competing capital demands, costs and execution assumptions. A position that looks controlled only because every stop is filled at its trigger is controlled by the simulation assumption.
Sources
- CME Group, Proper Position Size: risk-distance sizing and instrument value per tick.
- FINRA, Stop Orders: Factors to Consider During Volatile Markets, updated 26 March 2025: US stock stop execution and stop-limit non-execution risk.
- Ernest P. Chan, Quantitative Trading: How to Build Your Own Algorithmic Trading Business (Wiley, 2009), Chapter 6, pp. 95–97 and 105–106: allocation objectives, leverage and discontinuous losses. The numerical examples above are original hypothetical ledgers.