Position Sizing Methods
Position sizing methods are systematic frameworks for determining how much capital to allocate to each trade — including fixed fraction, the Kelly Criterion, and ATR-based sizing — to manage overall portfolio risk and avoid single-trade outcomes that significantly damage the account.
Position sizing is often described as the most important yet least discussed component of a trading system. Two traders using identical entry and exit rules but different position sizing approaches can produce drastically different long-term results from the same set of trade signals.
Fixed fraction (or fixed percentage) sizing is the simplest method. The trader risks a fixed percentage of their total account value per trade — commonly 1% or 2%. With a Rs 5 lakh account and a 1% risk rule, the maximum loss per trade is Rs 5,000. If the entry price and stop-loss define a per-unit risk of Rs 50 (for example, a stock at Rs 500 with a Rs 450 stop-loss), the position size is Rs 5,000 / Rs 50 = 100 shares. After a drawdown reduces the account to Rs 4 lakh, the same 1% rule means risking Rs 4,000 per trade — automatically reducing size during losing streaks, protecting the account from compounding losses.
The Kelly Criterion, developed by mathematician John Kelly in 1956, theoretically maximises the long-run growth rate of a bankroll given known probabilities of winning and losing and the payoff ratio. The Kelly fraction is: f = (Win Rate x (Reward/Risk) - Loss Rate) / (Reward/Risk). Practitioners typically used half-Kelly or quarter-Kelly in practice because the full Kelly fraction assumed perfectly accurate probability estimates, which were rarely available in live trading. Over-betting the Kelly fraction led to severe drawdowns despite a theoretically positive-expectation system.
ATR-based sizing (Average True Range) was used to normalise position size across instruments with different volatility levels. The ATR measured average daily price range over a chosen lookback period. If a trader used a 2-ATR stop-loss, the stop-loss distance varied by stock — a high-ATR stock would have a wider stop, automatically reducing position size per fixed-risk-per-trade rules. This prevented a high-volatility stock from taking an oversized bite of account capital on a normal daily move.
For F&O traders in India, position sizing was complicated by lot-size constraints. A Nifty futures lot had a fixed notional value, preventing fractional sizing. The practical workaround was to limit the number of lots per trade to a maximum defined by the risk-per-trade rule, even if it meant occasionally taking only one lot when the strict formula suggested a fraction of a lot.