ADX measures trend strength, not direction. Here is what it actually measures, how it was designed, and what most traders get wrong about it.
What is ADX? The Average Directional Index is a technical indicator that measures trend strength on a scale of 0 to 100. It does not measure direction — only how strong the directional movement is. Developed by J. Welles Wilder Jr. in 1978, it is one of the most misunderstood indicators in technical analysis, and the misunderstanding almost always comes down to that single distinction between strength and direction.
ADX measures the strength of directional movement — not the direction itself, and not whether to enter a trade. This distinction is the most important thing to understand about the indicator. It is also where most traders who use ADX go wrong.
A rising ADX means directional movement is becoming stronger, regardless of which direction that movement is pointing. ADX at 35 in a strong uptrend looks identical on the ADX line to ADX at 35 in a strong downtrend. The indicator makes no distinction between the two.
ADX operates on a 0 to 100 scale. Values below 20 indicate a market without strong directional movement — what Wilder called a non-trending state. Values above 25 indicate established trend conditions. Values above 40 represent strong trends. Values above 50 are uncommon and indicate unusually powerful directional conviction.
The direction of the trend — whether it is up or down — comes from the +DI and -DI lines of the Directional Movement Index (DMI), which runs alongside ADX. ADX tells you how strong the trend is. DMI tells you which way it is pointing. These two pieces of information together give a complete picture. ADX alone gives only half of it.
Wilder published ADX in "New Concepts in Technical Trading Systems" in 1978, when most technical analysis focused on price patterns and moving averages. These tools worked well in trending markets and produced poor results in ranging markets. The problem was not the signals — they were functioning correctly. The problem was applying trend-following logic to a market that was not trending.
A moving average crossover system generates profitable trades during a strong trend and losses during a sideways market. But there was no systematic way to determine which condition was operating at any given moment. Traders would discover in hindsight that their system had been deployed in the wrong market type.
Wilder wanted to build an indicator that could answer one question before any other: is there a trend worth following right now? ADX was his answer. It was designed as a filter — a check to run before evaluating any directional entry signal. The question ADX answers is not "should I buy or sell?" It is "is the market in a state where trend-following signals are likely to work?"
This design philosophy is why ADX became foundational to systematic trading. It provides a market classification before signal evaluation, not a signal itself. Wilder published it alongside ATR, DMI, and Parabolic SAR as a complete system — each component addressing a different aspect of the same trading problem.
ADX does not measure price directly. It is derived from two other indicators Wilder published in the same book: the Positive Directional Indicator (+DI) and the Negative Directional Indicator (-DI), which together form the Directional Movement Index (DMI).
+DI measures upward price pressure — the magnitude of upward movement relative to the asset's True Range. -DI measures downward pressure by the same method. When +DI is above -DI, upward movement is dominant. When -DI is above +DI, downward movement is dominant.
The Directional Index (DX) is calculated from both: DX = 100 x |+DI minus -DI| / (+DI plus -DI). It measures the separation between the two directional indicators. When +DI and -DI are equal, DX is zero — no directional conviction in either direction. When one far exceeds the other, DX is high — strong directional conviction.
ADX is the smoothed average of DX over the lookback period, default 14. It measures the average degree of directional conviction over recent bars. This is why ADX rises in both strong uptrends and strong downtrends — it measures the separation between +DI and -DI, not which one is higher.
For the full DMI breakdown, see The DMI Indicator.
Reading ADX requires interpreting both the level and the slope. Level tells you the current strength. Slope tells you whether strength is building or fading.
Below 20: Weak or absent trend. The market is ranging or directionless. Trend-following signals are unreliable in this zone. Mean-reversion approaches are more appropriate.
20 to 25: Transitional zone. A trend may be emerging or a previous trend may be fading. Signals in this range carry more uncertainty than those clearly above or below the threshold.
25 to 40: Established trend present. Trend-following signals have a structural basis for edge. Slope still matters — a rising ADX at 26 is different from a falling ADX at 38.
40 to 50: Strong trend. High directional conviction. Also a zone to watch for exhaustion — ADX cannot rise indefinitely, and strong trends are statistically more likely to be in their mature phase than their early phase at these levels.
Above 50: Exceptional trend strength. Uncommon in most markets. Can precede sharp reversals when the extreme move exhausts. These readings are most common during major macro events or cascading liquidation scenarios in crypto.
The slope rule applies at all levels: a rising ADX at 22 is often more actionable for trend-following than a falling ADX at 35. The direction of change tells you where strength is going, not just where it has been.
ADX has four primary applications in a systematic trading framework.
Regime classification. ADX determines whether a market is in a trending or ranging state before any signal is evaluated. This is Wilder's original design intent. ADX below 20 = ranging. ADX above 25 and rising = trending. Everything follows from this classification.
Signal filtering. Trend-following signals are evaluated only when ADX confirms trending conditions. Mean-reversion signals are evaluated only when ADX confirms ranging conditions. This prevents applying the wrong strategy type to the wrong market state — the primary source of systematic losses in indicator-based approaches.
Confidence weighting. Within a trending regime, higher ADX indicates stronger trend conditions. This can inform position sizing or signal confidence scores. Not all trending conditions are equal — ADX at 40 warrants more confidence than ADX at 26.
Exit timing context. ADX peaking and beginning to decline signals that a trend is losing strength. This is not a precise exit trigger, but it is context for tightening trailing stops or reducing position size in advance of the formal exit signal.
For practical application of ADX in each of these roles, see How to Use the ADX Indicator and How to Read the ADX Indicator.
"ADX above 25 means buy." ADX above 25 means a trend is present. It says nothing about direction. A trader who sees ADX at 30 and immediately enters long is ignoring the directional information entirely. Direction requires the +DI/-DI lines from DMI — not ADX alone.
"ADX below 20 means bearish." ADX below 20 means the market is ranging — oscillating without persistent direction. It is not bearish. It is equally directionless in both directions. Mean-reversion approaches are more appropriate in this state, not trend-following in either direction.
"High ADX is always good for trend entry." Very high ADX (above 45) can indicate a trend approaching exhaustion rather than a prime entry condition. Entering when ADX is already at extreme levels often means entering near the end of the move rather than the beginning. ADX level without context of slope and trend duration is incomplete.
"ADX tells you when to buy." ADX tells you that a trend exists. The entry timing comes from the +DI/-DI crossover (DMI), the Parabolic SAR flip, or price structure. ADX provides the regime context for those signals. ADX alone does not generate entries. It was designed to qualify them.
Lag. ADX is a smoothed average of a derived calculation. By the time ADX confirms a strong trend at 25 or above, the trend has been developing for several bars. The confirmation arrives after the fact. This is the structural tradeoff of all smoothed indicators — lower noise at the cost of delayed signal.
Quiet, slow trends. A gradual uptrend on low volatility produces low ADX readings because True Range is small relative to the directional movement. ADX is calibrated to strong, volatile directional moves. Quiet persistent trends — common in some crypto assets during accumulation phases — can go undetected by ADX while remaining valid trending conditions.
Macro event spikes. A sudden news-driven move can temporarily push ADX above 25 without establishing a genuine trend. The calculation records the directional movement without knowing whether it will sustain for enough bars to constitute a tradeable trend. Combining ADX with price structure confirmation reduces these false regime readings.
The 25 threshold is not universal. Wilder observed the 25 level on daily commodity charts. It has generalized reasonably well across markets, but it is not mathematically guaranteed. On some assets and timeframes, the natural breakpoint between trending and ranging ADX readings may differ. Validating the threshold against historical performance on the specific asset being traded is more reliable than assuming 25 applies universally.
ADX is a good indicator for its intended purpose: classifying whether a market is trending or ranging before evaluating directional signals. It is not a standalone entry signal and should not be used as one. ADX is most valuable as a regime filter — confirming that trend-following conditions exist before applying trend-following strategies. Its primary limitation is lag, since it is a smoothed average of a derived calculation. Used correctly as a regime classification tool rather than a directional signal, it has proven robust across multiple markets and timeframes since 1978.
An ADX reading above 25 means an established trend is present — directional movement is strong enough to support trend-following strategies. It does not indicate direction; ADX at 28 in a strong uptrend looks identical to ADX at 28 in a strong downtrend. To determine direction, the +DI and -DI lines from the Directional Movement Index (DMI) must be checked alongside ADX. ADX above 25 is the regime confirmation gate, not the entry signal itself.
No. ADX does not indicate market direction. This is the most common misconception about the indicator. ADX measures only the strength of directional movement — how strong the trend is, not which way it points. A rising ADX at 35 looks identical whether the market is in a strong uptrend or a strong downtrend. Direction comes from the +DI and -DI lines of the Directional Movement Index. +DI above -DI indicates upward movement dominates. -DI above +DI indicates downward movement dominates. ADX and DMI together provide a complete picture; ADX alone provides only half of it.
ADX was invented by J. Welles Wilder Jr. and published in his 1978 book "New Concepts in Technical Trading Systems." Wilder developed ADX alongside three other indicators in the same publication: the Average True Range (ATR), the Directional Movement Index (DMI), and the Parabolic SAR. ADX was designed specifically to answer the question of whether a trend worth following was present before applying any directional entry signal.
ADX and DMI are related but measure different things. DMI (Directional Movement Index) consists of two lines — +DI and -DI — that measure upward and downward price pressure respectively. When +DI is above -DI, upward movement dominates. ADX is derived from DMI: it measures the separation between +DI and -DI, smoothed over the lookback period. ADX tells you how strong the trend is. DMI tells you which direction it is pointing. Both are needed for a complete picture — ADX without DMI gives trend strength without direction.
ADX in trading is the Average Directional Index, a technical indicator that measures trend strength on a scale of 0 to 100. Developed by J. Welles Wilder Jr. in 1978, it tells you how strong directional movement is — not which direction it is moving. ADX is used primarily as a regime filter: values above 25 indicate trending conditions where trend-following strategies have structural edge, while values below 20 indicate ranging conditions where mean-reversion approaches are more appropriate.
A good ADX reading depends on what you are looking for. For trend-following, ADX above 25 and rising indicates established trending conditions. ADX between 25 and 40 represents a confirmed trend. ADX above 40 represents a strong trend with high directional conviction, though values above 45 can indicate a trend approaching exhaustion rather than an ideal entry point. For regime identification, ADX below 20 signals ranging conditions suitable for mean-reversion approaches. The slope matters as much as the level: a rising ADX at 22 is often more actionable than a falling ADX at 35.