Advertisement IC Markets

ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Optimize MetaTrader 4, MetaTrader 5, cTrader, and TradingView for speed. Discuss Level II Market Depth (DOM), custom hotkeys, and volume indicators.
Post Reply
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by Fairman »

ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

The Average Directional Index, another of Wilder's creations, measures trend strength on a scale from 0 to 100. It says nothing about direction, only how strongly price is trending in whichever direction it is going.

A common reading is that below 20 suggests a weak or absent trend, and above 25 indicates a trend worth trading. Very high readings, above 40 or 50, show a powerful trend, though also one that may be mature.

This is the missing piece for many strategies. Trend-following methods work well when ADX is rising and above 25. Range strategies work better when it is low. Using the wrong approach in the wrong regime is a common cause of losses.

ADX is built from two other lines, +DI and -DI, which compare upward and downward movement. A crossover of the two lines is sometimes used as a directional signal, though it lags.

Like any smoothed indicator, ADX lags. It often confirms a trend after it has been running for a while, and a falling ADX does not mean the trend is reversing, only that it is weakening.

A useful use is as a filter, a switch between modes, not as a signal.

Practical step: mark periods on your chart where ADX was above 25 and below 20, and record how your usual setup performed in each.
It’s Fairman :geek:
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by Fairman »

A practical tip: watch the slope of ADX as much as the level. A reading of 25 that's rising is a strengthening trend. A reading of 40 that's falling is a trend losing steam. The direction of the line tells you more than the number alone. A falling ADX doesn't mean price will reverse. It often means the market is moving into a range or a slower grind. That's useful for choosing between strategies. When ADX is rising above 20 or 25, trend continuation setups make more sense. When it's below 20 and flat, range trading at the edges tends to work better. I'd check it on the timeframe you use for your bias, not your entry chart.
It’s Fairman :geek:
LondonNewsTrader
Posts: 421
Joined: Mon Sep 21, 2026 9:30 am

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by LondonNewsTrader »

ADX has a quirk on intraday charts that matters if you use it as the mode switch: a single release candle can lift it from the teens to above 25 within a few bars, because the directional movement on that one bar is so large. The reading then says "trend" when what really happened was a one-off repricing, and the next hour is often a choppy range while the market digests the number.

If you trade the M5 or M15, check how much of a high reading comes from one bar before trusting it, or simply ignore the filter for the first hour after high-impact data. On H4 and above the effect is much smaller, which fits your point about using it as context rather than a trigger.
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by Fairman »

ADX doesn't tell you direction, so I pair it with something simple like price above or below a moving average. ADX says there's a trend, the average says which way. Neither alone is enough. The DI lines that come with ADX also show direction, with DI plus above DI minus suggesting buyers in control. Another useful detail is that ADX turning down from a high level, above 40 for example, often means the trend is pausing, not reversing. It's a good moment to tighten stops or take partial profits rather than to enter fresh positions.
It’s Fairman :geek:
FXS Prop Desk
Posts: 1689
Joined: Wed Oct 07, 2026 9:52 am

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by FXS Prop Desk »

Because ADX lags, a reading of 40 often appears when much of the trend is already done. Many traders find the change in ADX more useful than the level: a rise from below 20 can mark the start of a trend, while a falling ADX from a high level often signals the trend is tiring.

On lower timeframes, combining ADX with the DI lines gives direction as well as strength, which makes it more usable as a filter.
FXS Official Team | The Idea Lab & Strategy
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by Shadow Trader »

Part of the lag comes from Wilder's smoothing, which is slower than the period suggests. A 14 period Wilder average weights each new bar by 1 in 14, the same as a 27 period exponential average.

ADX then smooths twice, once for the DI lines and again for the ADX line itself. On M15 that means each stage is digesting roughly 27 bars, close to seven hours, which explains why a reading of 40 so often arrives late in the move.

Some intraday traders drop the period to 7 for that reason, which behaves like a 13 period exponential average. The trade off is that the 25 threshold gets crossed far more often, so a single release candle like the one described above can flip it within two or three bars.

One way round it is to keep the standard 14 for the regime question and use only the direction of the line for timing. The turn up from below 20 is the earliest useful thing it says, and the exact level matters less than the turn.

Has anyone tested a 7 period ADX against the standard one on the same pair? The question is whether the earlier signal survives the extra false starts.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: ADX: Measuring How Strong a Trend Is, Not Which Way It Goes

Post by Shadow Trader »

The 25 threshold is borrowed from daily charts, and it does not carry over evenly between pairs and timeframes. Each market has its own normal range of ADX readings, so the same number can mean very different things.

Say ADX on EURUSD M15 sits above 25 for 30 percent of bars over a few months, while GBPJPY on the same timeframe sits above it 45 percent of the time. A fixed 25 then calls nearly half of GBPJPY trending, which is too generous for a filter meant to pick the clear cases.

A simple alternative is a percentile threshold. Take the last few months of readings on the pair and timeframe being traded, and set the trend line at the value that only the top 30 percent of bars exceed.

That keeps the filter equally selective everywhere, and it adapts when a pair goes through a quiet season. The downside is one more number to refresh, maybe once a quarter.

It also gives a cleaner way to judge the turn up mentioned earlier. A turn from the bottom fifth of the pair's own readings is a more consistent signal than a turn from below 20, which may be routine on one pair and rare on another. Has anyone compared percentile settings across pairs like that?
Post Reply