The Average Directional Index (ADX) was developed by J. Welles Wilder and published in his 1978 book "New Concepts in Technical Trading Systems." ADX is part of the Directional Movement System, which also includes the Positive Directional Indicator (+DI) and Negative Directional Indicator (-DI). While +DI and -DI indicate direction, ADX measures the strength of the trend regardless of direction — an ADX of 35 means the trend is strong whether price is rising or falling. This makes ADX uniquely valuable as a filter: trend-following strategies (EMA crossover, MACD, Ichimoku) perform best during trending conditions and poorly during choppy sideways markets. ADX quantifies exactly when a trend is strong enough to warrant trend-following entries and when the market is ranging (ADX below 20) and trend strategies should stand aside. For automated crypto bots, ADX as a pre-filter dramatically improves trend-strategy performance by eliminating signals during non-trending conditions.
Related guides: EMA trend following, MACD, Ichimoku, ATR (Wilder's second major indicator).
ADX Formula and Calculation
Directional Movement: +DM = High today - High yesterday (if positive and > |-DM|, else 0) -DM = Low yesterday - Low today (if positive and > |+DM|, else 0) True Range (TR) = max(High-Low, |High-PrevClose|, |Low-PrevClose|) Smoothed averages over period N (typically 14): +DI = 100 × Smoothed(+DM) / Smoothed(TR) -DI = 100 × Smoothed(-DM) / Smoothed(TR) DX = 100 × |+DI - -DI| / (+DI + -DI) ADX = Smoothed average of DX over N periods Standard period: 14 Scale: 0 to 100 (higher = stronger trend)
ADX Interpretation Thresholds
| ADX Value | Market Condition | Strategy Implication |
|---|---|---|
| Below 20 | No trend / ranging / sideways | Avoid trend-following entries — use mean reversion or stand aside |
| 20–25 | Trend forming / weakly trending | Trend entries allowed with confirmation — use larger confirmation requirements |
| 25–40 | Moderate trend | Standard trend-following conditions — proceed with EMA, MACD, Ichimoku signals |
| 40–60 | Strong trend | Excellent trend conditions — higher conviction entries, potentially larger position |
| Above 60 | Extremely strong trend | Watch for exhaustion — very strong trends can reverse sharply; tighten trailing stops |
ADX as a Strategy Filter (Most Common Use)
The most effective use of ADX in automated strategies is as a pre-filter — require ADX above 25 before allowing any trend-following signal to trigger. Example: EMA 20/50 crossover AND ADX above 25 → long entry. Without the ADX filter, an EMA crossover strategy generates whipsaw signals during sideways markets. With ADX above 25 filter, only EMA crosses occurring during genuine trending conditions execute. Backtests consistently show this filter improves Profit Factor and reduces trade count (fewer signals, but higher quality). See our EMA guide for the base strategy to filter.
ADX + DI Crossover Signals
A more active ADX signal uses the +DI/-DI crossover as the entry trigger with ADX above 20 as the qualifier:
- Long entry: +DI crosses above -DI AND ADX rises above 20 (confirming the directional move has trend strength)
- Exit: -DI crosses above +DI OR ADX falls below 20 (trend losing strength)
This approach generates more signals than the pure ADX filter model but requires ADX to confirm trend participation in the DI crossover direction. Combine with EMA 50 as a higher-timeframe bias filter. See our timeframe guide.
ADX Mean Reversion Inverse (Ranging Market Signal)
ADX below 20 is a signal of a ranging market — where mean-reversion strategies outperform. Use: when ADX is below 20 AND RSI touches oversold (below 35) → mean reversion long. The ADX confirms the market is ranging, making the RSI mean reversion entry appropriate. This is the inverse of the trend filter — activating mean-reversion entries when trend-following entries are suspended. See our RSI guide.
Configuring ADX in DennTech
- Navigate to Strategy → ADX / Directional Movement
- Period: 14 (standard Wilder period)
- Trend threshold: 25 (minimum ADX to allow trend-following signals)
- Signal mode: Filter (apply to existing strategy) or Standalone DI Crossover
- Enable rising ADX requirement (ADX must be rising, not just above threshold) for stronger filter
- Backtest with and without ADX filter to quantify the signal quality improvement
All 25 strategies at strategies page. Full docs at DennTech docs. Start at pricing page.
Frequently Asked Questions
- Should ADX be calculated on the same timeframe as the strategy it is filtering?
- Yes — ADX should be calculated on the same timeframe as the primary strategy for consistent signal timing. A Daily ADX filter applied to a 4H EMA strategy creates a timing mismatch: Daily ADX may be below 25 (daily ranging) while the 4H trend is clearly directional. For multi-timeframe setups, use ADX on both the signal timeframe (for entry qualification) and the higher timeframe (for broader trend context). The standard approach: 4H ADX above 25 as the primary filter for 4H strategy signals, with Daily ADX above 20 as an additional higher-timeframe context check. See our timeframe guide.
- Does ADX work differently on crypto compared to traditional financial markets?
- ADX was developed for commodity and futures markets with defined trading sessions. Crypto markets trade 24/7 without session resets, which means ADX accumulates directional movement continuously. In practice, crypto ADX works effectively with the same thresholds (below 20 = ranging, above 25 = trending) with one notable difference: crypto trends can sustain extreme ADX readings (60+) for longer periods during bull market phases than traditional markets. This means ADX above 60 should trigger trailing stop tightening (trend exhaustion watch) rather than immediate exit. The core ranging/trending distinction holds well across crypto pairs with sufficient liquidity. See the strategy library at the strategies page. Compare editions at the pricing page.
- Can ADX be used to switch automatically between trend and mean-reversion strategies in DennTech?
- Yes — this is called a regime filter. Configure DennTech to activate the EMA crossover (trend strategy) when ADX is above 25 and rising, and activate RSI mean reversion when ADX is below 20 and falling. This creates an adaptive system that deploys the appropriate strategy type for current market conditions. The regime filter approach requires careful backtesting to ensure the transition thresholds (25 for trend activation, 20 for mean reversion activation) produce consistent results on your specific pairs. The 5-point gap between thresholds (20 and 25) creates a neutral zone that avoids rapid oscillation between strategy modes. See our stress testing guide for validating regime filter implementations. Start at the pricing page and explore the live demo.
Trend tools: ADX (this guide), EMA, MACD, Ichimoku. All strategies at the strategies page.
For a related trend-strength indicator, read the Elder Ray Index strategy guide.