Most options accounts that blow up do not die from one bad strategy. They die from one oversized position. Position sizing decides how much a losing trade costs you, and with options the sizing question is harder than it looks, because the number that matters is not the notional value of the contracts. It is the most you can lose, and how many of your trades can lose at the same time.
Key Takeaway
- Size options trades by maximum loss (or your planned exit loss), never by notional value or premium collected.
- A common rule of thumb is to risk about 1% to 2% of the account on any single trade. It is a guideline, not a law.
- Positions in the same sector or the same index move together in a sell-off, so five trades can behave like one big trade.
- Kelly math is useful as a ceiling that shows when you have no edge. Fractional Kelly with noisy inputs is not a safe target.
- Smaller size is what keeps a normal losing streak from becoming a drawdown you cannot recover from.
Why Options Sizing Is Different From Stock Sizing
With shares, risk is easy to describe: you own 100 shares, the stock can fall, and your stop is somewhere below. Options split that into several different numbers, and traders routinely size off the wrong one.
- Notional exposure: the value of the shares the contract controls. One short put at a $50 strike controls $5,000 of stock.
- Premium collected or paid: the cash that changes hands today. It is small relative to the risk on a short option, which is why it is a poor sizing input.
- Maximum loss: the worst case if the trade goes completely wrong. For a defined-risk spread this is a fixed number. For a naked or cash-secured short option it is large, and for a short call it is theoretically unlimited.
- Planned loss: the loss at which you intend to exit. This is the number a disciplined seller of undefined-risk positions actually sizes against.
The practical rule: pick the loss number that is real for the structure you are trading, then size so that number is a small slice of the account.
Step 1: Define Risk Per Trade in Dollars
Start with a percentage of account value you are willing to lose on one trade. Many traders use 1% to 2%, and some use less while they are learning. Convert it to dollars first, then work backward to contracts.
Here is a hypothetical example on a $25,000 account with a 2% limit, so $500 of risk per trade.
| Hypothetical trade | Risk number used | Loss per contract | Contracts within $500 |
|---|---|---|---|
| $5-wide iron condor, $1.50 credit | Max loss: ($5.00 – $1.50) x 100 | $350 | 1 |
| $10-wide iron condor, $2.50 credit | Max loss: ($10.00 – $2.50) x 100 | $750 | 0 (over budget) |
| $2-wide put credit spread, $0.60 credit | Max loss: ($2.00 – $0.60) x 100 | $140 | 3 |
| Cash-secured put, $50 strike, $1.00 credit, exit if the put costs $3.00 to close | Planned loss: ($3.00 – $1.00) x 100 | $200 | 2 |
Two things stand out. First, the same account can hold very different contract counts depending on the structure, so there is no universal “trade 1 contract” answer. Second, the $10-wide condor does not fit the budget at all. Rather than stretch the limit, a trader following this rule would narrow the wings or skip the trade.
The last row only works if you actually exit at the planned loss. A cash-secured put at a $50 strike has a worst case near $4,900 if the stock went to zero. A stock that gaps down overnight can blow through a planned exit, so for undefined-risk positions it is common to size smaller than the planned-loss math alone suggests.
Step 2: Account for Correlation
The 2% rule assumes each trade can lose independently. In real markets they often cannot. In a broad sell-off, correlations rise and positions that looked different start losing together.
Consider a hypothetical account holding five short put spreads, each risking 2%, all on large technology stocks. On a normal day that looks like five separate bets. In a sharp tech drawdown, it can behave like a single 10% bet. Diversifying by ticker is not the same as diversifying by risk.
A few ways traders handle this:
- Cluster caps: limit combined risk in one sector, theme, or index to a fixed share of the account, for example 5%.
- Total risk cap: limit the sum of maximum losses across all open positions, for example 10% to 20% of the account, so a bad week cannot take most of the account.
- Beta-weighted delta: convert each position’s delta into SPY-equivalent terms and add them up. If the total is large relative to the account, you are more directionally exposed than the individual trades suggest.
- Mix structures: a short-vega position (such as an iron condor) and a long-vega position respond differently to a volatility spike.
Step 3: Respect the Math of Drawdowns
Position size is a decision about how bad a losing streak is allowed to get. Losses compound against you, and recovering from them takes more than the percentage lost.
| Drawdown | Gain needed to get back to even |
|---|---|
| 10% | 11.1% |
| 20% | 25% |
| 30% | 42.9% |
| 50% | 100% |
Now apply it to size. This table shows what ten consecutive full losses would do to an account, assuming each trade loses its entire risk budget.
| Risk per trade | Account after 10 straight losses | Drawdown |
|---|---|---|
| 1% | 90.4% of start | 9.6% |
| 2% | 81.7% of start | 18.3% |
| 5% | 59.9% of start | 40.1% |
| 10% | 34.9% of start | 65.1% |
Ten straight losses is an extreme case, but shorter streaks are routine. A strategy that wins 70% of the time still loses four in a row about once every couple of hundred trades by simple probability, and clustered losses in a volatility spike make streaks more likely than independent odds suggest. At 2% a streak is uncomfortable. At 10% it ends the account.
Step 4: Use Kelly as a Ceiling, Not a Target
The Kelly criterion is a formula for the bet size that maximizes long-run growth given a win rate and a payoff ratio. A common form is:
Kelly fraction = win rate – (loss rate / payoff ratio), where the payoff ratio is the average win divided by the average loss.
Premium sellers often have high win rates and small average wins relative to losses. That is exactly where Kelly is informative. Two hypothetical examples:
- 60% win rate, average win $100, average loss $200. Payoff ratio is 0.5. Kelly = 0.60 – (0.40 / 0.5) = -0.20. A negative result means the strategy has no edge at those numbers, and the correct size is zero.
- 70% win rate, average win $100, average loss $150. Payoff ratio is 0.67. Kelly = 0.70 – (0.30 / 0.67) = 0.25. Full Kelly would risk 25% of the account per trade, and quarter Kelly would still be about 6%.
The second example shows why Kelly is a ceiling. The inputs are estimates from a limited sample, and small errors in win rate swing the output a lot. The formula also assumes independent trades with stable odds, which options trades in a stressed market are not. Many traders use a small fraction of Kelly, and most would not size above the 1% to 2% guideline regardless. You need a record of at least 50 to 100 trades before the inputs mean much, which is one reason to keep an options trading journal.
The first example is the more valuable lesson: if your average loss is twice your average win, a 60% win rate loses money. Sizing cannot fix a strategy with negative expectancy, but it can limit how fast you find out.
Step 5: Scale Size With the Account, Not With Your Mood
Because the rule is a percentage, size scales automatically. A $25,000 account that grows to $30,000 moves the 2% budget from $500 to $600. A drawdown to $20,000 cuts it to $400. That built-in reduction after losses is the point: it slows you down when your edge may be weakest.
The common failure is the opposite. After a winning streak, traders double up because the last ten trades worked, and the largest position meets the first loser. Keep the percentage fixed and let the dollar size follow the account.
Practical Rules of Thumb
- Risk about 1% to 2% of account value on one trade, measured by max loss or planned exit loss.
- Cap combined risk in one sector or theme at roughly 5%.
- Cap total open risk at a level you can survive, often 10% to 20%.
- For undefined-risk positions, size below the planned-loss math to leave room for gaps.
- Close or reduce positions by a rule set in advance, such as the 21-DTE and 50% profit exit, rather than in the moment.
- If a trade does not fit the budget at the minimum size of one contract, skip it or choose a narrower structure.
These numbers are illustrative starting points, not recommendations. The right budget depends on your account size, experience, and tolerance for drawdowns.
Who This Approach Is Not For
On a very small account, a 1% to 2% budget may be $100 to $200, which cannot fund a one-contract position in many underlyings. Traders in that spot usually move to narrower spreads or lower-priced underlyings, and accept that some strategies are not yet practical. Stretching the risk limit to fit a trade is the usual way small accounts end.
Bottom Line
Size by the loss you can actually take, not by the premium or the notional. Keep each trade near 1% to 2% of the account, count correlated positions as one, and treat Kelly as a ceiling. The goal is to still be trading after the worst stretch of your first few hundred trades.
FAQ
Q: What percentage of my account should I risk on one options trade?
A: Many traders use 1% to 2% of account value as maximum loss per trade, and beginners often start lower. It is a guideline, and the right number depends on your experience and how much drawdown you can tolerate.
Q: How do I size a cash-secured put or covered call?
A: Start with the collateral, because it is the capital tied up, then decide the loss at which you will exit and size so that loss fits your per-trade budget. A put that is fully collateralized still has a worst case near the full strike value minus premium.
Q: Does the 2% rule apply to the premium I collect?
A: No. Premium is the reward. The rule applies to the maximum loss, or your planned exit loss, which for a short option is usually several times the credit.
Q: Is the Kelly criterion safe to use for options?
A: Full Kelly is aggressive and depends on win-rate estimates that are hard to measure from a small sample. Most traders use a fraction of it at most, and many use it only to check whether a strategy has an edge at all.
Q: How many positions should I hold at once?
A: There is no fixed number. What matters is combined risk: add up maximum losses across all open trades, group correlated ones together, and keep the total within a cap you can survive.
Keep Learning
See how defined risk works in practice in our cash-secured put guide and the breakdown of cash-secured versus naked puts and margin, or browse more lessons in the Learn hub.