Kairon AITwo traders buy the same stock at the same price and sell at the same price. One barely notices the loss, the other loses a fifth of their account. The only difference is position size. It is the least exciting and most important decision in any trade.
The formula
Risk-based position sizing answers one question: how many shares can I buy so that, if my stop is hit, I lose exactly the amount I decided in advance?
Shares = (account size × risk %) ÷ (entry price − stop price)
Three inputs, one output:
- Account size: the capital you trade with.
- Risk %: the share of the account you accept losing on this trade. Many traders use 0.5 % to 2 %.
- Distance to the stop: how far the price can move against you before you exit.
Worked example
Your account is $20,000 and you risk 1 % per trade, so $200. You want to buy a stock at $50 with a stop at $46.
- Risk per share: $50 − $46 = $4
- Shares: $200 ÷ $4 = 50 shares
- Position value: 50 × $50 = $2,500, or 12.5 % of the account
If the stop is hit, you lose $200 plus fees and any gap. If you had placed the stop at $48 instead, the same 1 % risk would allow 100 shares and a $5,000 position. The stop decides the size, not your conviction.
Why a fixed percentage works
Risking a fixed share of the account keeps losing streaks survivable. Every strategy has them. With 1 % risk per trade, ten losses in a row cost you roughly 10 % of the account. With 10 % risk per trade, the same streak costs about 65 %, and you need almost a 190 % gain just to get back to where you were.
| Risk per trade | Account after 10 straight losses | Gain needed to recover |
|---|---|---|
| 1 % | ~90 % | ~11 % |
| 2 % | ~82 % | ~22 % |
| 5 % | ~60 % | ~67 % |
| 10 % | ~35 % | ~187 % |
The numbers compound because each loss is taken from a smaller account.
Common mistakes
Sizing by feel. "I'm really sure about this one" is how most oversized positions start. Confidence is not information about risk.
Moving the stop to fit the size. If you want 200 shares and the math says 50, tightening the stop until 200 fits just means you will be stopped out by normal noise. Choose the stop from the chart and the stock's volatility, then let it determine the size.
Ignoring gaps. A stop is not a guarantee. Earnings or news can open the stock far below your stop. Around known events, reduce size or accept that the real risk is larger than the formula says.
Forgetting correlation. Five positions in five semiconductor stocks are closer to one large position than to five independent ones. Add up the risk of positions that tend to move together.
Letting the position grow unnoticed. A winner that doubles also doubles its weight in your account. Decide in advance at what weight you trim.
What about long-term investors?
If you do not use stops, the same logic still applies through the maximum loss you would tolerate before you would re-examine the thesis, and through a maximum weight per position (for example 5 to 10 %). The point is to decide the downside in advance, while you are calm.
Where an AI analysis fits in
An analysis can help you place the stop more sensibly: key support levels, the average daily range of the stock, upcoming events that make a gap more likely. It cannot tell you how much of your money to put at risk. That part is yours, and it is the part that decides whether you are still trading a year from now.
Get a second opinion on your next stock
Five AI agents look at technicals, fundamentals, news and sentiment, then a bull and a bear argue it out. A free account includes one compact AI analysis every month.
This guide is educational and not investment advice.
Get a second opinion on your next stock
Five AI agents look at technicals, fundamentals, news and sentiment, then a bull and a bear argue it out. A free account includes one compact AI analysis every month.