The forex market in 2020 was not the same market as 2016 or 2023. Central bank policy cycles, geopolitical events, global growth regimes, and technology-driven changes in market microstructure mean that the market's character — its volatility, its trending tendency, the persistence of moves — changes over time. A strategy backtested on five years of historical data performs well in the conditions that dominated that five-year period. When the regime shifts — when the market transitions from sustained directional trends to choppy, news-driven volatility — the same strategy, unchanged, begins to underperform. The professional trader recognises this transition, adapts their approach, and preserves capital until conditions align with their edge again. The amateur trader forces their system onto conditions it was not designed for and calls the resulting losses bad luck.
Markets Are Not Static
Every strategy is implicitly calibrated to a specific market regime — even if the trader who designed it is not aware of this. A strategy that relies on sustained directional momentum — trend following at its core — performs well in trending markets and produces many false signals in ranging markets. A strategy that relies on mean reversion — buying support, selling resistance — performs well in ranging markets and gets destroyed by trending moves.
The key insight is that no single regime lasts forever. Trending markets eventually transition to ranges. Ranging markets eventually break out into new trends. High-volatility regimes return to low volatility. The professional's task is not to predict which regime comes next — it is to recognise which regime is currently dominant and to calibrate their approach accordingly.
Identifying Regime Change
Regime change is identified through both technical and performance-based signals.
TECHNICAL SIGNALS: ADX falling from above 25 toward 15-20: Trend strength is weakening. Ranging conditions developing. Bollinger Bands narrowing significantly: Volatility contracting. Squeeze setting up. Moving averages tangling (20, 50, 200 EMA crossing repeatedly without sustained directional separation): No clear trend structure. Ranging or transitioning regime. Sessions showing no directional follow-through: London open begins a move but New York open reverses it repeatedly. Institutional direction unclear. PERFORMANCE SIGNALS: Win rate drops 10%+ below backtested expectation over 30+ trades: Your setups are forming but the expected moves are not developing. Possible regime mismatch. Setup frequency drops 40%+ from average: The market is not presenting the conditions your system requires. Regime has potentially shifted away from your optimal conditions. Multiple winning trades reversed before second target: Market is showing less persistence than the backtest assumed. Trending regime may have ended.
Adapting Without Abandoning
The critical distinction in market adaptation is between adjusting to confirmed regime change and abandoning a valid system during a normal drawdown. These situations can feel identical from the inside — both involve a period of underperformance. The difference is in the evidence.
Normal drawdown: performance is below expectation over a small sample (under 30 trades) or within the range of the maximum historical drawdown identified in backtesting. The response is to continue executing with the same system and reduced position size.
Regime change: performance is below expectation over a larger sample (30+ trades), the technical signals of regime change are present, and setup frequency has changed. The response is to adapt — not abandon.
Adding Market Condition Filters
The most practical adaptation tool is the market condition filter — a rule that restricts trading to specific market environments where your edge is strongest. Rather than trading all market conditions and accepting reduced performance during non-optimal regimes, a condition filter restricts trading to only the conditions your backtesting shows the system performs best in.
Trend filter: "I will only take long entries when the daily chart shows ADX above 20 and price is above the 50 EMA. I will only take short entries when ADX is above 20 and price is below the 50 EMA." Volatility filter: "I will not trade any pair whose daily ATR is below 50 pips — the market is too quiet for my momentum-based strategy to work." Session filter: "I will only trade the London and New York overlap (13:00-17:00 UTC) — the highest volume period that produces the sustained directional moves my system relies on."
The Adaptation Process
When regime change is confirmed, the adaptation process follows a specific sequence.
Step 1: Confirm regime change. Technical signals present. Performance signals present over 30+ trades. Not a 2-week drawdown — a confirmed shift. Step 2: Reduce position size. From 1% to 0.5% per trade during the adaptation period. Preserve capital. Step 3: Identify the specific change. Is it trending → ranging? (Reduce momentum entries, add mean reversion layer.) Is it low → high volatility? (Widen stops proportionally using ATR.) Is it a sentiment shift? (Apply more macro filters before entry.) Step 4: Backtest the adjustment. Apply the adapted criteria to recent historical data — does the adjustment restore positive expectancy in the new conditions? Step 5: Implement on demo for 2 weeks. Confirm the adaptation works in current live conditions before applying to the live account. Step 6: Implement live at reduced size. Return to full size only after 30+ trades confirm the adaptation is working.