The Problem with Fixed Sizing in Variable Volatility
Many traders use a fixed dollar amount or a fixed percentage of their bankroll for every prediction. While percentage-based sizing is a good start, it ignores a crucial variable: market volatility. A $100 position in a calm market carries fundamentally different risk than a $100 position in a highly volatile market.
When volatility spikes, price swings become larger. If you maintain the same position size, your portfolio experiences much larger swings in equity. You are inadvertently taking on more risk exactly when the market is most dangerous.
To maintain consistent risk regardless of market conditions, your position size needs to inversely correlate with volatility: trade smaller when volatility is high, and trade larger when volatility is low.
Enter Average True Range (ATR)
Average True Range (ATR) is an indicator developed by J. Welles Wilder that measures market volatility by decomposing the entire range of an asset price for that period. It looks at the current high minus the current low, the absolute value of the current high minus the previous close, and the absolute value of the current low minus the previous close.
Unlike directional indicators, ATR doesn't tell you which way the market is going. It only tells you how much it's moving. An increasing ATR means volatility is rising; a decreasing ATR means volatility is falling.
By incorporating the ATR value into your sizing formula, you create a dynamic system that automatically scales your exposure to match the current market environment.
Calculating ATR-Based Sizing
In traditional trading, ATR is used to place stop-losses. You might set your stop at 2x the current 14-period ATR below your entry. The position size is then calculated to ensure that if the stop is hit, you only lose a predetermined percentage of your account (e.g., 1%).
The formula is: Position Size = (Account Risk Amount) / (ATR * Multiplier). If you want to risk $50, the ATR is $2, and you use a 1.5x multiplier, your position size is $50 / ($2 * 1.5) = $50 / $3 = 16.6 units.
This ensures that whether a stock normally moves $1 a day or $10 a day, your account takes the exact same percentage hit if your thesis is wrong.
Adapting ATR for Prediction Markets
In fixed-payout prediction markets like KeyCandle, you don't set stop-losses because risk is strictly limited to your stake. However, you can still use ATR to scale your stake. When the ATR of your target timeframe is significantly higher than its historical average, the market is historically turbulent.
In highly turbulent markets, price action becomes erratic and less predictable. Traditional candlestick patterns yield lower win rates because noise overwhelms signal. Therefore, when ATR spikes, you should manually reduce your standard stack size (e.g., from 2% to 1%).
Conversely, when ATR is historically low, the market is consolidating. Wait for the breakout, and when it happens, you can deploy your standard or slightly elevated size, knowing the move is likely genuine.
Creating an ATR Volatility Filter
Don't overcomplicate it. Calculate the 30-day moving average of the 14-period ATR. This gives you a baseline for "normal" volatility.
Create a rule: "If current ATR > 1.5x the 30-day ATR Average, cut base prediction size by 50%." This single rule protects your bankroll from the erratic behavior that occurs during sudden market shocks or news events.
ATR sizing forces you to respect the market's current state. It is the mathematical embodiment of the trader's maxim: "When the water is choppy, take a smaller boat."