Forex Trend Following: A Rule-Based Time-Series Momentum Guide
Educational content only. This guide is not investment advice and does not promise profits. Forex and CFD trading can lead to losses beyond the amount risked on an individual position unless negative-balance protection applies. Test every rule with realistic spreads, commission, slippage and financing costs before considering live trading.
Trend following asks a deliberately simple question: has a currency pair been moving up or down over a defined lookback period? It then holds a position in the same direction until the signal changes. The idea is not that trends always continue; most individual trades may be small losses. The strategy depends on occasional sustained moves being large enough to offset those losses.
What the evidence supports
Moskowitz, Ooi and Pedersen documented time-series momentum across 58 liquid equity-index, currency, commodity and bond futures. Their results showed persistence over roughly one to twelve months and partial reversal at longer horizons. This is broad cross-asset evidence, not a guarantee that a particular retail FX implementation will remain profitable. See Time Series Momentum, Journal of Financial Economics (2012).
Long samples also show that technical-rule profitability changes over time. Hsu, Taylor and Wang evaluated more than 21,000 rules across 30 currencies with data-snooping controls and out-of-sample tests. Coakley, Marzano and Nankervis similarly found that the number of apparently profitable FX rules falls sharply after robust multiple-testing controls. See Hsu et al. (2016) and Coakley et al. (2016).
A reproducible monthly rule
- Universe: begin with liquid major pairs such as EUR/USD, GBP/USD, USD/JPY and AUD/USD. Do not add pairs after seeing which ones backtested best.
- Data: use end-of-day mid prices from one consistent source. Keep bid/ask data separately so costs can be modelled.
- Signal: on the last trading day of each month, calculate the percentage change from the close 12 months earlier to the current close.
- Direction: hold long when the 12-month return is positive and short when it is negative. A zero reading means no position.
- Rebalance: recalculate monthly. Do not change the rule between signals because of news or a recent loss.
- Exit: close or reverse only at the next monthly signal. A separate emergency stop may be used, but it must be included in every test.
Volatility-scaled position sizing
Calculate the 20-day Average True Range (ATR). Define one unit of risk as two ATRs from entry. A conservative educational example risks 0.25% of account equity per pair: position size = (equity × 0.0025) ÷ stop distance. Cap total open risk across correlated USD positions at 1%. This is an example framework, not a recommendation.
ATR sizing does not prevent gaps, slippage or correlated losses. EUR/USD, GBP/USD and AUD/USD may all express nearly the same USD view. Treat correlated trades as a portfolio rather than independent bets.
How to test it honestly
- Reserve the most recent 20–30% of observations as untouched out-of-sample data.
- Use rolling or expanding walk-forward tests instead of tuning on the complete history.
- Deduct spread, commission, slippage and overnight financing on every trade.
- Report maximum drawdown, turnover, losing streaks and exposure—not only annual return or win rate.
- Stress the lookback at 9, 12 and 15 months. A result that works only at one exact setting is fragile.
- Compare with a no-skill benchmark and document every change before rerunning the test.
When the strategy can fail
Trend following is vulnerable to whipsaws in range-bound markets, sudden policy reversals, crowded positioning and cost drag. Leverage can turn an ordinary signal error into a large account loss. Past momentum is an observed historical tendency, not a law of markets.
Implementation checklist
- Use fixed data and execution timestamps.
- Record the signal before the next bar.
- Keep risk per pair and portfolio risk capped.
- Review performance on a predefined schedule rather than after every loss.
- Stop live trading if execution costs or drawdown exceed tested assumptions.
Test the process before risking money
Affiliate disclosure: The following Tickmill link is an affiliate link. Trade-Forex.info may receive compensation if you register through it, at no additional charge to you. This does not affect the methodology or the risks described above.
Explore Tickmill demo and account options (affiliate link)
Risk warning: CFDs are complex instruments and carry a high risk of losing money rapidly due to leverage. 73% of retail investor accounts lose money when trading CFDs with Tickmill Europe Ltd. Consider whether you understand how CFDs work and whether you can afford the high risk of losing your money.