Equity curve trading is a meta-strategy approach that uses your own strategy’s performance history to decide when to trade and when to stand aside. Instead of running a strategy at constant exposure regardless of whether it is performing in-line with expectations, equity curve trading treats the strategy’s equity curve as a signal — and dynamically adjusts position size or halts trading based on whether that curve is above or below its own trend. It is one of the most practical tools for managing a live algorithmic strategy.
What Is Equity Curve Trading?
The equity curve is a running plot of your strategy’s cumulative profit and loss over time. In a healthy strategy, the equity curve trends upward with drawdowns that recover to new highs. Equity curve trading adds a filter on top of this: trade normally when the equity curve is above a moving average of itself; reduce position size or pause entirely when it falls below.
The underlying logic is that a strategy’s equity curve falling below its own moving average is an early warning that the strategy has entered a regime it was not designed for — or that market conditions have shifted in a way that reduces its edge. Stepping back during these periods reduces exposure to losses that the raw strategy would have taken without question.
Why Equity Curve Trading Works
Most trading strategies are designed for a specific market regime. A mean-reversion strategy outperforms in ranging markets and underperforms in trending ones. A trend-following strategy does the reverse. No strategy performs equally across all conditions.
The equity curve naturally reflects these regime shifts. When a mean-reversion strategy enters a trending market, the equity curve begins declining — the strategy is generating losses it would not generate in its optimal regime. The moving average filter detects this decline earlier than the drawdown threshold would. Pausing or reducing size during the below-average period preserves capital. Resuming when the curve recovers above the average re-engages the strategy in conditions more likely to be favourable.
This approach reduces the depth and duration of drawdowns without requiring the trader to identify regime changes directly. The equity curve does the detection automatically.
How to Set the Moving Average Filter
The moving average applied to the equity curve uses the same mechanics as a price moving average — but instead of smoothing price, it smooths cumulative P&L. Common choices are a 20-trade or 50-trade simple moving average. Shorter periods respond faster to regime changes but produce more false signals (the strategy pauses during temporary drawdowns that would have recovered). Longer periods produce fewer false signals but respond more slowly to genuine breakdowns.
Calibrate the period using backtest data. Look at the worst drawdown periods in the backtest and ask: would a 20-trade MA have reduced exposure during those periods? Would a 50-trade MA have been too slow to respond? The right period balances responsiveness against whipsaw — the scenario where the filter pauses the strategy just before it would have recovered.
Some systematic traders use two thresholds rather than one: reduce position size to 50% when the equity curve falls below the moving average, and halt entirely if it falls 10% below. This creates a graduated response rather than a binary on/off switch — preserving some activity during moderate drawdowns while halting during severe ones.
Risks and Limitations
Whipsaw around the moving average: If the equity curve oscillates just above and below the moving average, the filter generates frequent switch signals. Each switch involves a period of reduced activity — potentially missing recoveries. Use a buffer zone (e.g. resume trading only when the curve is 2% above the moving average, not just above it) to reduce sensitivity at the boundary.
Optimising the filter on in-sample data: Selecting the equity curve MA period that produces the best backtest result is a form of overfitting. The optimal period on historical data is unlikely to remain optimal in live trading. Use a reasonable default (20–50 trades) and accept some inefficiency rather than hyper-optimising to past data. See our post on parameter optimisation for how to avoid overfitting when calibrating strategy inputs.
Compounding pauses: Every pause period is a period of zero gains. If the strategy’s edge is real but the filter pauses it frequently due to normal volatility, the compounded cost of missed gains can outweigh the drawdown reduction. Track the equity curve filter’s own performance: does enabling it improve risk-adjusted returns over the full backtest period, not just the worst drawdown section?
How to Apply Equity Curve Trading in Arrow Algo
Arrow Algo’s platform lets you track live strategy performance through the scenario dashboard and use the data to implement equity curve controls.
To build the filter, track cumulative P&L using the last_profit block across completed trades. Apply an EMA or SMA block to the cumulative P&L output over your chosen lookback period. Add a condition block comparing current P&L to the moving average. Wire this condition as an AND gate on all entry signals — entries only fire when the equity curve condition is met (current P&L above its moving average).
For the graduated approach, add a second condition at a lower threshold (e.g. 10% below the MA) and connect it to a position size multiplier block set to 0.5. When the first condition fails, position size halves. When the second condition fires, entries stop entirely. The two-threshold structure lets you manage exposure proportionally rather than switching on and off.
Backtest the strategy with and without the equity curve filter across the same historical period. Compare maximum drawdown, recovery time, and Sharpe ratio between the two versions. If the filter materially improves risk-adjusted returns without significantly reducing absolute return, it is earning its place in the strategy.
What Are the Key Takeaways?
- Equity curve trading uses your strategy’s own performance trend as a filter — trading normally when the equity curve is above its moving average and reducing or pausing when it is below
- The filter detects regime changes automatically — when a strategy’s conditions have shifted, the equity curve starts declining before the drawdown becomes severe
- Calibrate the moving average period on backtest data using 20–50 trades as a starting range; avoid hyper-optimising to historical results
- Use a graduated two-threshold approach (reduce at first threshold, halt at second) to avoid binary on/off whipsaw
- Track the filter’s own performance: it must improve risk-adjusted returns across the full backtest period to justify its use
- Arrow Algo’s last_profit, EMA/SMA, and condition blocks let you build equity curve tracking and filtering directly inside your strategy
Disclaimer: This content is for educational purposes only and does not constitute financial advice. Trading involves significant risk and you should only trade with capital you can afford to lose. Past performance is not indicative of future results. Always conduct your own research before making any trading decisions.
Ready to build your own automated trading strategies without writing a single line of code? Start for free at Arrow Algo and join thousands of traders who’ve made the switch to systematic trading.
