Stock Position Sizing: Constrain Scenario Loss and Total Exposure

Translate a documented stock thesis into shares by applying scenario-loss, notional, total-loss, liquidity, leverage, issuer, theme, derivative, and portfolio stress limits, then reconcile actual fills and changing exposures.

For educational purposes only; not investment advice. Investing may result in loss.

Direct answer

Stock position sizing chooses a quantity that fits a defined portfolio, thesis, instrument, and set of adverse outcomes. A defensible plan calculates candidate shares under scenario-loss, notional, total-loss, liquidity, leverage, issuer-family, correlated-theme, and mandate limits, then uses the smallest applicable result. It also states rounding, order, review, exit, and exception rules.

A basic long-stock scenario calculation is Scenario loss per share = Entry price - Planned execution price + Gap and slippage allowance + Per-share costs and Scenario shares = floor(Scenario risk budget / Scenario loss per share). This is a planning scenario, not a maximum-loss guarantee. A stop price is a trigger, not a guaranteed fill; a stop order becomes a market order when triggered, a stop-limit order may remain unfilled, trading can halt, and a long stock can fall to zero.

Size should be based on the portfolio’s loss capacity and aggregate exposures, not only conviction or forecast volatility. The same issuer can appear through common shares, employer stock, funds, options, convertibles, structured products, and collateral. Different issuers can share one customer, commodity, financing source, country, currency, duration, regulation, or economic factor. Count those dependencies before treating positions as diversified.

How it works

Build and maintain the plan in this order:

Define the capital base and objective. Record owner, goal, account, net liquidation value, gross exposure, cash, liabilities, margin loan, collateral, tax status, horizon, liquidity needs, maximum portfolio drawdown, and whether outside assets or employer income belong in the risk perimeter. Do not use gross assets when the policy is based on net equity.
Write the thesis and invalidation evidence. Separate fundamental invalidation, review price, stop trigger, stop limit, time exit, profit-taking rule, and actual execution price. State events requiring review, such as earnings, financing, litigation, regulation, product failure, covenant breach, management change, or a security becoming ineligible. A lower price alone is not improved evidence.
Define instrument exposure and loss mechanics. Identify shares, multiplier, contract terms, delta, gamma, assignment, exercise, conversion, borrow, dividends, corporate actions, settlement, currency, and entity. Long stock can lose notional; a long option can lose premium; short options and leveraged or structured claims can lose far more than premium received. Current delta is not a downside-loss ceiling.
Calculate a scenario candidate conservatively. Use a documented portfolio risk fraction: Scenario risk budget = Portfolio net equity × Risk-budget percentage. Estimate the planned execution price rather than merely the stop trigger, include gap, spread, impact, fees, and FX, and round shares down: Scenario shares = floor(Scenario risk budget / Scenario loss per share). Recalculate after the actual entry fill because entry slippage changes both notional and loss per share.
Apply independent hard and soft caps. Calculate shares under Notional cap, Zero-value or defined disaster cap, single issuer and issuer-family cap, theme and factor cap, liquidity and exit-horizon cap, margin and buying-power cap, legal or mandate cap, and tax or lot restrictions. Use Final shares = min(All applicable candidate-share limits) before minimum-lot rounding; if the minimum tradable lot exceeds the limit, make no trade.
Stress the whole portfolio. Shock price, gap, volatility, correlations, spread, depth, borrow, rates, FX, earnings, halts, defaults, option exercise and assignment, margin requirements, and common drivers. Correlations can rise in stress. Include nonlinear option payoffs and positions held through funds or other accounts; do not sum premiums or static deltas as if they were final losses.
Execute, reconcile, and monitor. Specify order type, time in force, session, venue, maximum participation, partial-fill handling, cancellation, and no-trade conditions. After fills, reconcile shares, average price, fees, notional, scenario loss, total-loss stress, issuer and theme exposure, liquidity, margin excess, and residual risk budget. Resize after price moves, issuance, buybacks, exercises, changed holdings, thesis evidence, or portfolio withdrawals.

Scaling in can reduce one-time entry exposure but does not lower the final planned loss unless total size or exit assumptions change. Scaling out can restore a cap without expressing a negative company view. A position that grows through appreciation creates the same current concentration as one purchased at that size and must be tested against the current policy.

Example

Use a hypothetical unlevered portfolio with $250,000.0000 net equity and explicit policy assumptions:

Scenario and independent caps: a 0.5000% scenario budget is $250,000.0000 × 0.5000% = $1,250.0000. Entry is $80.0000, planned execution is $72.0000, gap and slippage allowance is $3.7500, and estimated round-trip cost is $0.2500 per share, so scenario loss is $80.0000 - $72.0000 + $3.7500 + $0.2500 = $12.0000 per share and candidate size is floor($1,250.0000 / $12.0000) = 104 shares. A 5.0000% single-name cap permits floor(($250,000.0000 × 5.0000%) / $80.0000) = 156 shares; a 4.0000% zero-value cap permits floor(($250,000.0000 × 4.0000%) / $80.0000) = 125 shares; and an exit limit of 2.0000% of 20,000 shares average daily volume over 3 sessions permits 20,000 × 2.0000% × 3 = 1,200 shares.
Portfolio theme constraint sets final size: the theme cap is 10.0000%, or $250,000.0000 × 10.0000% = $25,000.0000, and existing correlated exposure is $17,000.0000, leaving $25,000.0000 - $17,000.0000 = $8,000.0000, or floor($8,000.0000 / $80.0000) = 100 shares. Final size is min(104, 156, 125, 1,200, 100) = 100 shares, notional is 100 × $80.0000 = $8,000.0000, and planned scenario loss is 100 × $12.0000 = $1,200.0000. The unused $50.0000 risk budget is not permission to violate the theme cap.
Stop, gap, and total-loss stress: if the stop triggers but the actual fill is $70.5000, stock price loss is 100 × ($80.0000 - $70.5000) = $950.0000; adding the estimated 100 × $0.2500 = $25.0000 costs gives $975.0000. If unexpected news gaps directly to $60.0000, price loss is 100 × ($80.0000 - $60.0000) = $2,000.0000 before costs, exceeding the scenario budget. At zero, stock loss is the full $8,000.0000; the $1,250.0000 scenario budget never capped that disaster loss.
Options and optimization require separate limits: selling 2 put contracts with a $75.0000 strike, $2.0000 per-share premium, and 100-share multiplier receives 2 × $2.0000 × 100 = $400.0000, but zero-price assignment loss net of premium is 2 × ($75.0000 - $2.0000) × 100 = $14,600.0000, not $400.0000. For a simplified repeated binary bet with win probability 55.0000%, gain 20.0000%, and loss 10.0000%, odds are b = 20.0000% / 10.0000% = 2.0000, and the Kelly fraction is f* = (b × p - q) / b = (2.0000 × 55.0000% - 45.0000%) / 2.0000 = 32.5000%. Changing estimated win probability to 50.0000% gives (2.0000 × 50.0000% - 50.0000%) / 2.0000 = 25.0000%. This sensitivity, plus nonbinary stock returns, correlation, tail loss, estimation error, and investor constraints, prevents the formula from being a stand-alone sizing rule.

Risks

Define owner, goal, account, net equity, gross exposure, cash, liabilities, collateral, and outside assets.
Separate loss willingness from financial capacity, liquidity needs, liabilities, and maximum drawdown.
Distinguish thesis invalidation, review price, stop trigger, stop limit, and actual execution price.
Stress earnings, financing, litigation, regulation, covenants, fraud, default, delisting, and zero value.
Treat a stop as an order trigger, not a guaranteed execution price or maximum loss.
Check session eligibility, trigger source, corporate-action adjustment, halts, gaps, and stale prices.
Include entry and exit spread, impact, fees, FX, borrow, financing, and tax-lot effects.
Aggregate common stock, employer awards, funds, options, convertibles, notes, and collateral by issuer.
Aggregate customer, supplier, commodity, sector, geography, currency, duration, regulation, and funding themes.
Look through funds and structured products rather than trusting security labels or account separation.
Use option multipliers, assignment, exercise, nonlinear payoff, and stress Greeks, not premium or current delta alone.
Limit size by zero-value or defined disaster loss independently from the planned-exit scenario.
Compare shares with ADV, participation, depth, spread, volatility, sessions to exit, and market capacity.
Include margin loan, interest, house requirements, maintenance excess, forced liquidation, and cross-position effects.
Recognize that a broker can raise house requirements and liquidate positions without allowing security selection.
Round down after all candidate limits and reject a trade whose minimum lot exceeds the limit.
Recalculate after actual fills, price changes, issuance, buybacks, exercise, assignment, and portfolio cash flows.
Separate evidence-based adding from averaging down solely because price declined.
Document no-trade conditions, partial-fill rules, review dates, exceptions, and independent overrides.
Do not treat confidence, historical volatility, Kelly output, or one forecast distribution as loss capacity.

Common misconceptions

“A stop price caps the position’s loss.” It is a trigger; execution can be worse, a stop-limit may not fill, trading can halt, and the security can gap toward zero.
“The scenario risk budget is the maximum possible loss.” It covers one stated path; total-loss, gap, liquidity, leverage, options, and correlated-portfolio stresses require separate caps.
“More names or more accounts automatically reduce concentration.” Overlapping issuers, funds, employer exposure, derivatives, and common economic drivers can create one aggregate bet.
“Option premium or current delta fully measures position risk.” Long and short options, assignment, exercise, gamma, volatility, liquidity, and nonlinear payoff can produce very different losses.
“Kelly or conviction gives one correct stock percentage.” Optimization is highly sensitive to uncertain probabilities, payoffs, correlations, tails, horizon, and constraints; capacity caps remain necessary.

Authoritative sources

FINRA, Concentrate on Concentration Risk — issuer, employer, correlated, sector, region, fund-overlap, structured-product, and illiquidity concentration.
FINRA, Risk — loss capacity, liquidity risk, concentration risk, diversification, leverage, short selling, options, and complex-product context.
FINRA, Brokerage Accounts — margin loans, initial and maintenance requirements, house rules, interest, margin calls, forced liquidation, notice, and account shortfalls.
Investor.gov, Types of Orders — market, limit, and stop orders and the distinction between trigger, execution, and price guarantee.
Investor.gov, Trading Halts and Delays — regulatory and nonregulatory halts, pending news, order imbalances, and listing uncertainty.
Investor.gov, Asset Allocation and Diversification — goal, horizon, risk tolerance, allocation, diversification, overlap, and rebalancing context.
The Options Clearing Corporation, Characteristics and Risks of Standardized Options — current exchange-listed option terms, exercise, assignment, settlement, adjustment, and risk disclosures.
Kelly, A New Interpretation of Information Rate — the original long-run growth optimization result and its mathematical assumptions, not a universal stock-allocation mandate.