Position sizing determines how much capital to allocate to each individual trade. It is one of the most impactful decisions in trading strategy design, often more important than the entry and exit signals themselves. A good strategy with poor position sizing can lose money, while a mediocre strategy with intelligent position sizing can survive long enough to be profitable.
Why position sizing matters
The core challenge of position sizing is balancing growth with survival. Betting too aggressively on each trade maximizes potential returns but also maximizes the risk of ruin, a drawdown so severe that the strategy can never recover. Betting too conservatively ensures survival but sacrifices returns. Optimal position sizing finds the sweet spot between these extremes.
Consider two traders using identical entry and exit signals. Trader A risks 20% of capital per trade, Trader B risks 2%. After a streak of five losing trades (entirely possible even for a strategy with a 60% win rate), Trader A has lost approximately 67% of capital and needs a 200% gain to recover. Trader B has lost approximately 10% and needs only an 11% gain to recover. Position sizing determined the outcome, not the signal quality.
Common position sizing approaches
Fixed fractional sizing allocates a constant percentage of current capital to each trade. The most common variant is risking 1-2% of capital per trade, measured from entry to stop loss. This approach automatically adjusts position size as capital grows or shrinks, providing natural scaling.
The Kelly criterion calculates the mathematically optimal bet size to maximize long-term geometric growth rate. The formula is: Kelly fraction = (win rate * average win / average loss - loss rate) / (average win / average loss). In practice, most traders use a fraction of Kelly (half-Kelly or quarter-Kelly) because full Kelly sizing produces very aggressive positions and large drawdowns.
Volatility-based sizing adjusts position size inversely to the instrument's recent volatility. When volatility is high, positions are smaller; when volatility is low, positions are larger. This approach, used in many trend-following systems, normalizes the risk contribution of each trade regardless of the instrument's volatility characteristics.
Equal risk contribution allocates positions so that each trade contributes equally to portfolio risk. This requires estimating each position's expected volatility and sizing accordingly. It is particularly useful in multi-asset strategies where instruments have vastly different volatility profiles.
Position sizing in backtesting
Backtesting with proper position sizing reveals how a strategy's returns and drawdowns change under different allocation approaches. A strategy might show excellent results with 5% risk per trade but become unacceptable with 10% risk per trade. Testing multiple position sizing approaches helps traders find the allocation that best matches their risk tolerance and return objectives.
Practical example
A strategy has a 55% win rate with an average win of $200 and average loss of $150. The Kelly criterion suggests: (0.55 * 200/150 - 0.45) / (200/150) = (0.733 - 0.45) / 1.333 = 0.212, or 21.2% of capital per trade. This is extremely aggressive. Using half-Kelly (10.6%) would still produce significant drawdowns. A conservative fixed fractional approach of 2% risk per trade would be more practical for most traders, even though it sacrifices theoretical growth rate.
How Tektii helps
Tektii allows traders to test their strategies with different position sizing models and compare the resulting performance metrics. The platform supports custom sizing logic within strategy containers so traders can implement any approach, from simple fixed fractional to complex volatility-adjusted models. By making it easy to evaluate how position sizing affects returns and drawdowns, Tektii helps traders make informed decisions about capital allocation.