Position sizing is the part of a trading method that decides how much to buy or sell on each trade, and Van Tharp's CPR model reduces the calculation to one line: position size equals the cash you are willing to risk divided by the risk per share. Risk Rs 6,000 on a stock where your stop sits Rs 50 below entry, and you buy 120 shares; that is the entire formula. Tharp held that this "how much" decision, not stock selection, is what determines whether you meet your objectives, and that until you know your system deeply you should risk only about 1% of your equity per trade. This is part 7 of Artha's 10-part series on Tharp's Super Trader (McGraw-Hill, 2009); the model is his, the rupee examples are ours, and this is education, not advice. Parts 5 and 6 supply the R and expectancy vocabulary used here.
Chapter 1Why is position sizing more important than picking stocks?
Because even an outstanding system goes bankrupt when sized badly, while a modest system sized well can meet sensible objectives. Tharp's extreme illustration: a system that wins 95% of the time still destroys anyone who risks their whole account on a single trade, since the 5% loser eventually arrives. Less extreme versions of the same failure happen daily: a trader with a genuine edge risks 10% or 20% of capital per position, hits an ordinary losing streak, and never recovers.
The academic world brushes against the same truth under another name. A well-known Financial Analysts Journal study by Brinson and colleagues attributed about 91% of portfolio performance variability to asset allocation, defined as how much sits in stocks, bonds, and cash. Tharp's reading was that "how much" is precisely position sizing, and that most professionals citing the study never notice that the driver is quantity, not selection. His marble game, described in part 6, demonstrates it live: identical trade sequences, wildly different outcomes, purely from sizing choices.
Chapter 2What is the CPR model?
CPR is a memory aid for the three quantities in the sizing equation:
- C is the cash you will risk on this trade, your total risk in rupees. If you risk 1% of a Rs 5,00,000 account, C is Rs 5,000.
- R is the risk per share: entry price minus the initial stop price. Buy at Rs 500 with a stop at Rs 450 and R is Rs 50 per share. This is the same R that defines the trade's 1R in R-multiple terms.
- P is the position size you solve for: P = C / R.
With C = Rs 5,000 and R = Rs 50, P is 100 shares. The position's market value is Rs 50,000, but your risk, if the stop is honoured, is only Rs 5,000, one-tenth of the exposure. That distinction between investment and risk is one of the most clarifying by-products of the model: a large position with a tight stop can carry less risk than a small position with no stop at all.
Chapter 3How does the 1% guideline work?
Tharp's baseline recommendation for anyone who does not yet deeply know their system: risk about 1% of current equity per trade, so a Rs 10,00,000 account risks Rs 10,000 per trade, whatever the instrument. If one stock's stop is Rs 25 away, you buy 400 shares; if another's is Rs 100 away, you buy 100 shares. Every position then carries the same 1R of account risk, which is what makes R-multiples poolable across your journal.
Because the 1% is taken from current equity, size shrinks in drawdowns and grows in advances, a built-in survival mechanism. A worked sequence (illustrative arithmetic): start with Rs 10,00,000 and risk 1%. Three straight -1R losses take equity to roughly Rs 9,70,300, with each successive risk slightly smaller than the last. A -5R disaster then costs about Rs 48,500, dropping equity near Rs 9,21,800. A +10R winner then adds about Rs 92,200. After this rough ride of four losers and one winner, the account stands near Rs 10,14,000, up about 1.4%, still fully in business. The same sequence risked at 10% per trade would have been devastating: the -5R trade alone would have removed about 37% of what remained.
What do the worked examples look like across instruments?
The formula is universal; only the inputs change.
Stock example: Rs 3,00,000 account, willing to risk 2%, so C is Rs 6,000. Stock at Rs 500, stop at Rs 450, R is Rs 50. P = 6,000 / 50 = 120 shares, costing Rs 60,000. Cost is 20% of the account; risk is 2%.
Tight-stop example: day trade at Rs 300 with a Rs 3 stop, risking 0.5% of a Rs 4,00,000 account, so C is Rs 2,000 and R is Rs 3. P is 666 shares, costing very nearly Rs 2,00,000, half the account deployed, yet only Rs 2,000 at risk if the stop holds. Tight stops buy big positions with small risk, and they fail exactly when gaps or slippage blow through them, which is the trade-off part 8 explores.
Lot-size example: derivatives trade where the stop distance works out to Rs 12,000 of risk per lot. With C of Rs 5,000, P = 5,000 / 12,000 = 0.42 lots. You cannot trade 0.42 lots, and the honest conclusion is that this trade is too large for the account at this risk level. Tharp built exactly this trap into his teaching examples: sometimes the correct position size is zero, and the formula is how you find out.
What are the other position sizing models?
Percent risk is Tharp's baseline, but his framework treats sizing as a family of models chosen to fit objectives. Common members include equal-units sizing, where you allocate fixed rupee amounts per position regardless of stops; percent-volatility sizing, where the position is sized so that a normal day's movement, measured by ATR, equals a set fraction of equity, useful when you want portfolio positions to feel equally "loud"; and variations that differ in how they measure the equity base itself when multiple positions are open. Each model produces different behaviour in streaks and different drawdown profiles.
The deeper point is the purpose: position sizing is the instrument through which you pursue the objectives written in your business plan. A trader whose objective caps drawdown at 15% needs smaller risk per trade than one who accepts 40% swings for higher return, even on the identical system. Tharp's approach for serious students was simulation: take your system's R-multiple distribution, run thousands of simulated trade sequences at different risk levels, and observe which sizing gives an acceptable probability of meeting your target without breaching your drawdown limit. The system provides the expectancy; sizing tunes the outcome distribution to your goals.
Chapter 6How does risking too much destroy a positive-expectancy system?
Through losing streaks interacting with the mathematics of recovery. Even the +0.8R marble system from part 6 loses 80% of the time, making eight-loss streaks unremarkable. At 1% risk, such a streak costs under 8% of equity, annoying and survivable. At 10% risk, it removes more than half the account, and a 50% drawdown needs a 100% gain merely to break even, a hole most traders never climb out of, financially or psychologically. Oversizing converts a winning system into a losing account without a single bad signal, which is Tharp's core argument for why "how much" outranks "what and when."
How Nora helps
Nora can run the CPR arithmetic with your numbers, show how different risk percentages change drawdown depth on a simulated R-multiple sequence, and quiz you on sizing decisions across stocks and lots, education in the mechanics, never instructions about actual trades.
App · coming soonWhat this means for you
The CPR model turns the vaguest question in trading, how much, into one division: cash risked over risk per share. Wrapped in the 1% guideline and tied to written objectives, it is the machinery by which Tharp's framework converts a system's expectancy into an account's outcome. What the model cannot supply is the stop-loss that defines R in the first place; that comes from exits, and exits, the most underrated component of any system, are the subject of the next article.
Series credit: this series is based on concepts from Super Trader: Make Consistent Profits in Good and Bad Markets by Van K. Tharp, Ph.D. (McGraw-Hill, 2009). Full credit for the framework belongs to Dr. Tharp and the Van Tharp Institute.