Every earlier lesson ended with rules for a single trade. This one is about the account. Premium selling has a lopsided payoff (lesson 1), its losses cluster (lessons 4 and 6), and its worst days arrive without warning. Under those conditions, survival depends less on picking the right strike than on never being too large when the bad day comes. The general principles are in the Learn article on stop-loss and position sizing; here they are adapted to short options.
Risk per trade: what to measure
For a stock trade, risk is usually the distance to the stop times the number of shares. For short options that number is unreliable, because prices can gap through any stop. Measure instead:
- Defined-risk trades (credit spreads, iron condors): the maximum loss, width minus credit, times 100 per contract.
- Undefined-risk trades (cash-secured puts, covered calls, the wheel): a stress loss, the loss if the stock gaps a set amount against you. A 25% gap is a reasonable default for a large company; use more for small or volatile stocks, and never less than the stock's largest past one-day move.
Then cap that loss at a fixed fraction of the account, commonly 1–2% per trade. The fraction is a choice of temperament, but it must be fixed in advance and applied to every trade.
Sizing in practice
Illustrative numbers. Account: $60,000. Risk per trade: 2%, or $1,200.
| Trade | Risk measure per contract | Contracts within $1,200 | Total risk |
|---|---|---|---|
| Iron condor, 5-wide wings, credit 1.15 | Max loss $385 | 3 | $1,155 |
| Bull put spread 190/185, credit 1.35 | Max loss $365 | 3 | $1,095 |
| Cash-secured 48 put on a $50 stock, credit 0.90 | Stress loss at a 25% gap to $37.50: (48 − 37.50 − 0.90) × 100 = $960 | 1 | $960 |
Note what the sizing ignores: the premium. Three condors collect $345 and one put collects $90, but both are sized by what they can lose. Note also the cash-secured put's footprint: one contract commits $4,800, 8% of the account, even though its stress loss fits the 2% budget.
Limits for the whole book
Trades that are each sized correctly can still add up to too much. Four book-level limits catch this:
- Total risk. The sum of maximum and stress losses across open positions, for example no more than 10–15% of the account.
- Net delta. Add the share-equivalent Deltas of every position and multiply by each stock's price to get dollar Delta. A book of short puts and bull put spreads is long the market even if each trade felt neutral. Example: two short 0.30-Delta puts on $50 stocks (+60 shares, $3,000) and three 190/185 bull put spreads on a $200 stock (about +8.4 shares each, about $5,000) make roughly $8,000 of long exposure, so a 5% market drop costs about $400 from Delta alone, before Gamma and a rise in IV add to it.
- Margin and buying power. Keep the amount committed well below the maximum your broker allows, often under half of the account, so that a volatility spike that raises margin requirements does not force you to close at the worst prices.
- Concentration. No more than about a quarter of total risk in one sector, and no single stock large enough to define the year.
Then run one stress test before adding any position: the market falls 10% in a week and implied volatility rises 10 points. If the estimated loss on the whole book would exceed what you have decided you can lose in a month, the new position does not go on.
Rolling rules
A roll closes one position and opens another; it is a new trade and must pass the entry rules. Write the limits down:
- Roll only for a net credit, to the same or a better strike.
- No more than two rolls per position, and never beyond about 60 days.
- Never roll into an expiry that contains earnings or another binary event.
- Before rolling, ask whether you would open the new position today with no history. If not, close.
The journal
A journal turns experience into evidence. Record for every trade: date, stock, strategy, days to expiry, short-strike Deltas, IV, IV relative to realized, IV rank, the expected move, the credit, the maximum or stress loss and its share of the account, the reason for the trade, the exit plan, and afterwards the result, the exit reason and whether every rule was followed. After 30–50 trades, review: which setups paid, what the average loss was relative to the average gain (lesson 1), and which rule breaks cost the most.
When to stop
Decide in advance at which points you will step back, for example:
- Down 5% in a calendar month: no new positions until the next month.
- Down 10% from the account's high: halve all sizes and review the journal before trading again.
- Three maximum losses in a row, or any trade opened outside the plan: pause for a week and find the cause.
- Market stress (an inverted index term structure, negative dealer gamma, IV spikes across the watchlist): trade smaller or not at all until it passes.
Managing an open position
How to apply it: a one-page plan
- Universe. A written list of stocks and ETFs with liquid options that you would be prepared to own.
- Setups. Which structures you trade, with their default days to expiry and Deltas (lessons 2 to 5).
- Entry filter. IV rich relative to realized, IV percentile or rank in the upper half, no binary event before expiry, strikes at or beyond the expected move (lesson 6).
- Sizing. 1–2% of the account per trade by maximum or stress loss; book limits for total risk, net Delta, margin and sector.
- Exits. Target 50% of the maximum profit; stop at about twice the credit or a breached short strike; time exit at about 21 days; stock-level exits for the wheel.
- Rolling, journal and stop-trading rules as above, reviewed every quarter.
In Clear Tape
The strategy builder on the Options page deliberately prices one lot and places no orders, so sizing stays your decision: it shows the maximum loss, capital at risk and the position's Greeks per lot, which you multiply by the number of contracts your plan allows. The watchlist scan reads up to 40 stocks from the same watchlist as the Trading systems page and lists IV, IV relative to realized volatility, skew, dealer gamma regime, the walls, the next earnings date and the strategies that currently fit, which makes the entry filter quick to apply. The free index cards show whether SPY, QQQ and IWM are in a positive or negative gamma regime and whether their term structure is inverted, inputs for the stop-trading rules. Pro members can also ask for an AI read of a stock (five a day): Claude reads the numbers on the page and writes one view with levels, scenarios, earnings and volatility, and risks. It is a cross-check, it can misread the data, and it is not a signal. Data reflects the previous session's close. See pricing.
Common mistakes
- Sizing by premium collected or by margin required instead of by what the trade can lose.
- Checking each trade's risk but never the sum, the net Delta or the sector mix.
- Using most of the available buying power, leaving no room when margin requirements rise.
- Rolling indefinitely to avoid recognizing a loss.
- Increasing size after a losing streak to win it back, or after a winning streak because it feels easy.
Summary
An options-income program succeeds when its worst month is survivable. Measure risk by maximum or stress loss, cap it per trade and for the whole book, and write down how you roll, what you record and when you stop, before the market tests those rules. The course in one line: you are selling insurance, so price it with care (lessons 1 and 6), understand what drives it (lesson 2), choose the right policy (lessons 3 to 5) and never underwrite more than you can pay out (lesson 7).
For educational purposes only. Not investment advice.