Ask professional traders what separates the survivors from the blown-up accounts and almost none of them talk about entries. They talk about position sizing — how much you risk when you are wrong. Jack Schwager interviewed the best traders alive for Market Wizards and reported the same refrain from nearly all of them: protect the downside first. Van Tharp built a whole framework on it. This guide is that framework, in plain numbers.
The 1% rule
The classic rule from the trading literature: risk no more than 1–2% of your account on any single trade. Not 1% of the account invested — 1% lost if your stop is hit. On a $50,000 account at 1%, a losing trade costs $500, whatever the position size happens to be.
The reason is losing-streak arithmetic. Even a good system loses five times in a row routinely. At 1% risk, ten straight losses draw the account down about 10% — annoying, recoverable. At 10% risk, the same streak destroys nearly two-thirds of the account. And drawdown math is cruelly asymmetric: lose 50% and you need +100% just to get back to even. Sizing small is not timidity; it is what keeps the compounding machine running.
From risk to share count
Position size falls out of three numbers — account, risk percent, and the distance from entry to stop:
Shares = (Account × Risk%) ÷ (Entry − Stop)
Example: $50,000 account, 1% risk ($500), buying at $40 with a stop at $38. The stop distance is $2, so $500 ÷ $2 = 250 shares — a $10,000 position. Notice what the formula does: a tighter stop buys more shares for the same dollar risk, a wider stop buys fewer. The risk stays constant; the size flexes. The position size calculator does this arithmetic, including for shorts.
Think in R, not dollars
Tharp’s greatest export is the R multiple: measure every trade in units of its initial risk. If you risked $500, then a $1,500 win is +3R and a stopped-out loss is −1R. Suddenly your trading history becomes comparable — a +2R trade on a small account and a +2R trade on a big one are the same quality of decision.
R leads directly to expectancy — the average R you make per trade:
Expectancy = Win% × Average win (R) − Loss% × 1R
A 40% win rate sounds bad — until you notice that with 3R average winners it earns 0.4 × 3 − 0.6 = +0.6R per trade. Meanwhile a 90% win rate with tiny wins and −5R disasters loses money. This is the deepest lesson in the trading classics: profitability lives in the sizing and the ratio, not the win rate. Run your own numbers in the risk/reward calculator.
Where the stop goes (so sizing means something)
The formula assumes the stop is real — placed where the trade idea is objectively wrong, not where the loss feels acceptable. The two defensible approaches from the literature: structural (below the support level, swing low, or breakout base that justified the entry) and volatility-based(a multiple of the stock’s average daily range, so normal noise doesn’t tag you out). Setting the stop first and deriving the size from it is exactly backwards from how beginners trade — and it is the correct order.
Portfolio heat: the rule most people miss
Five open positions at 1% each is not 1% risk — it is up to 5% if they all hit stops together, and correlated positions (five semiconductor longs) tend to do exactly that. The pros cap total open risk — “portfolio heat” — at around 5–6%, and count correlated trades as one bet. When you are wrong about a theme, you will be wrong across every expression of it at once.
The practical checklist
- Decide account risk per trade: 1% (2% max, and only with an edge you have measured).
- Find the invalidation point first — the price where the idea is wrong.
- Size with the formula; round down to whole shares.
- Demand a reward at least 2× the risk before entering; know your breakeven win rate.
- Cap total open risk near 5%, counting correlated positions together.
- Log every trade in R. Fifty logged trades tell you your real expectancy.
None of this makes a bad strategy good. What it does is guarantee that no single trade — and no realistic losing streak — can take you out of the game while you find out whether your strategy is good. That is the whole job of risk management.