EURUSD ADX Trend Strength Filter Strategy

News Catalyst

EURUSD has no single headline driving it today, and that is what makes the setup worth watching. The one scheduled eurozone release is Germany's Balance of Trade in the early European session, with a surplus of €19.0B forecast against €21.3B previously. It is a low-impact print, but a narrower surplus fits the familiar story of a soft German export engine, and a miss of that size tends to push the euro lower through the London morning instead of causing one sharp spike. The larger cross-currents come from energy and geopolitics. The IEA says it will speed up its strategic oil reserve release, with 100 million barrels still to come, which eases the import bill of an energy-dependent eurozone and supports the euro. Reports that Syria is weighing military aid to Saudi Arabia in the Yemen war keep a safe-haven bid under the dollar. When the drivers pull in opposite directions like this, EURUSD usually chops until one side wins and then trends hard. An ADX filter is built for exactly that: it stays flat during the chop and only allows trades once trend strength confirms which force won. One headline needs filtering out: Circle bringing EURC stablecoin payments to SAP customers matters for crypto adoption but has no meaningful effect on the spot exchange rate.

Trade Summary

Most trend-following losses happen when there is no trend to follow. Moving-average and breakout signals fire just as often in a sideways market as in a trending one, and the trader has no objective way to tell the two apart in real time. This strategy fixes that with a single gate: no trade is taken unless the market can prove it is trending. ADX (14) is the primary indicator. A reading above 25 and rising shows that directional movement is strong and getting stronger, and the +DI/−DI pair tells you which side is in control. A 50-period EMA confirms that price structure agrees with that direction. A Pin Bar times the entry on a pullback, and a session filter limits trading to the London and New York hours, when EURUSD trends are actually funded.

The system is directionally neutral. Longs and shorts follow mirror-image rules, and ADX decides whether either side is active at all. It suits developing, catalyst-driven trends like the one that could follow today's energy-versus-haven tug of war. It deliberately sits out the long, low-volatility ranges that EURUSD is known for, which is exactly when ADX stays below 20. ATR (14) sets the stop distance, so position risk adjusts automatically between quiet and active sessions.

The Anatomy of the Trade

The Logic: What Inefficiency Are We Exploiting?

In FX, institutional repositioning after a macro shift takes several sessions to complete because large flows are worked into the market over time. Once a trend is genuinely established, it therefore tends to continue. The difficulty is not the continuation itself but telling a real trend from the random drift inside a range. ADX measures directional movement without regard to direction, so it answers that question directly. The edge being harvested is trend persistence, traded only once its strength is measurable. Requiring ADX to be rising, not just above 25, removes the late phase where trend strength is already fading and new entrants supply the exit liquidity for earlier positions.

Each component filters out a different kind of failure, which is what produces the confluence. ADX removes ranging markets. The DI crossover and the 50 EMA together remove trades taken against the dominant side. The session filter removes the thin Asian hours, when EURUSD drifts on almost no volume and ADX readings are least reliable. The Pin Bar refines timing. Instead of buying the extended candle that pushed ADX over 25, the system waits for a pullback toward the EMA to be rejected by a long wick, which shows that the other side tried to retake control and failed. That costs some entries in the fastest trends, but it puts the stop behind a defined rejection level rather than in the middle of an impulse candle.

Setup Requirements

• Primary indicator: ADX with period 14 and its +DI/−DI components. The trend gate is ADX > 25 and higher than it was 3 bars ago.
• Structure filter: 50-period EMA on the 30-minute close. Take longs only when price is above it and shorts only when price is below it.
• Confirmation: A Bullish Pin Bar for longs or a Bearish Pin Bar for shorts. The wick must be at least 2× the body and must test the EMA or the prior swing level.
• Session filter: Entries only between 07:00 and 16:00 UTC, covering the London session and the London–New York overlap.
• Risk management: ATR with period 14, using 1.5× ATR as the stop distance.
• Primary symbol: EURUSD. It is the most liquid currency pair in the world, with the tightest spreads, and it alternates clearly between multi-week ranges and sustained macro trends. That makes it a good fit for a strategy whose whole job is telling those two regimes apart.
• Timeframe: 30-minute. ADX lags more than most indicators, and on 5m or 15m charts it reacts to intraday noise. The 30-minute chart smooths that out while still giving several qualifying setups in a trending week.
• Adaptability: The same logic works on other liquid majors, on gold and on index CFDs. Raise the ADX threshold to around 30 on instruments that trend persistently, so the filter keeps screening out weak moves rather than passing almost everything.

Entry Rules

All conditions below must be true on the same closed 30-minute candle. If ADX qualifies but there is no Pin Bar, there is no trade.

• Long entry: ADX(14) is above 25 and rising over the last 3 bars and +DI is above −DI and price closes above the 50 EMA and a Bullish Pin Bar rejects a pullback toward the EMA and the candle closes between 07:00 and 16:00 UTC.
• Short entry: ADX(14) is above 25 and rising over the last 3 bars and −DI is above +DI and price closes below the 50 EMA and a Bearish Pin Bar rejects a rally toward the EMA and the candle closes between 07:00 and 16:00 UTC.

Enter at the close of the confirmation candle. A Pin Bar is not a Pin Bar until the candle closes. A long wick halfway through the bar is often filled in by the close, and so is the ADX reading that seemed to qualify.

Exit Rules

• Stop loss: 1.5 × ATR(14) from entry, or just beyond the Pin Bar's wick, whichever is further from entry.
• Take profit: 3.0 × ATR(14), giving at least 2:1 reward-to-risk.
• Signal exit: Close the position if ADX falls below 20 or the DI lines cross against the trade. Either one means the trend condition behind the entry no longer holds.

The stop loss is non-negotiable. It goes in at entry and only ever moves in the trade's favour. Trend-following systems lose on more than half their trades. They work only because those losses are capped while the winners are allowed to run.

Risk Management

• Risk per trade: 1–2% of account equity. Start at 1% until you have at least 100 recorded trades.
• Risk-to-reward ratio: A minimum of 2:1, which breaks even at a win rate of roughly 34% before costs. That margin is needed, because filtered trend systems commonly win only 35–45% of the time.
• Position sizing: Size each position from the ATR stop. Example: a $20,000 account risking 1% has $200 at risk per trade. If ATR(14) on the 30-minute EURUSD chart is 9 pips, the stop is 1.5 × 9 = 13.5 pips. At $10 per pip on a standard lot, one lot risks $135, so the position size is $200 ÷ $135 = 1.48 standard lots (148,000 units).
• Maximum concurrent positions: Two, and never two USD-correlated majors in the same direction. Long EURUSD plus long GBPUSD is effectively one short-dollar position at twice the intended risk.

SYMBOL:      EURUSD
TIMEFRAME:   30m
SESSION:     07:00–16:00 UTC

LONG ENTRY:
  ADX(14) > 25                         // Trend is strong
  AND ADX(14) > ADX(14)[3]              // ...and strengthening
  AND +DI > −DI                       // Buyers in control
  AND Close > EMA(50)                   // Structure agrees
  AND Bullish Pin Bar at EMA / swing    // Pullback rejected

SHORT ENTRY:
  ADX(14) > 25                         // Trend is strong
  AND ADX(14) > ADX(14)[3]              // ...and strengthening
  AND −DI > +DI                       // Sellers in control
  AND Close < EMA(50)                   // Structure agrees
  AND Bearish Pin Bar at EMA / swing    // Rally rejected

STOP LOSS:   1.5 × ATR(14) from entry
              // Or beyond Pin Bar wick, whichever is further

TAKE PROFIT: 3.0 × ATR(14)
              // Minimum 2:1 reward-to-risk

SIGNAL EXIT: ADX(14) < 20 OR DI lines cross against trade

RISK:        1% of equity per trade
MAX OPEN:    2 positions

Copy the block above into the Strategy Note field in the Arconomy Strategy Designer so the written rules are saved with the strategy. See working with rules for how the note stays visible next to the rule blocks while you build.

Common Pitfalls

An ADX filter cannot rescue a system that ignores its readings. These five mistakes cause most of the damage when trading ADX on a 30-minute EURUSD chart.

Trading the Range Anyway

EURUSD often spends weeks inside a band of 100–150 pips with ADX stuck between 12 and 20. During those stretches the system correctly produces almost nothing, and the urge to "just take the obvious bounce" is strongest. A low ADX reading is not a reason to switch strategies mid-session; it is the filter telling you the edge is absent. If ranging conditions frustrate you, build a separate mean-reversion system for them. Do not quietly bend this one to fit.

Entering Into Scheduled Releases

ECB decisions, eurozone CPI, US non-farm payrolls and FOMC days produce candles that push ADX above 25 within one or two bars, often in a direction that reverses within the hour. Even a low-impact print like today's German trade balance can briefly widen spreads in a thin market. Take no new entries from 30 minutes before to 30 minutes after a high-impact EUR or USD release. Open positions keep their existing stops, but assume those stops may fill with slippage.

Accepting a Flat or Falling ADX

The rule is ADX above 25 and rising. Traders often drop the second half because a reading of 32 looks strong enough. But an ADX of 32 that has been falling for six bars describes a trend that is running out, and entering then usually means buying the top of the move. The slope of ADX matters as much as its level, and dropping the slope condition turns a trend filter into a late-entry signal. The same goes for accepting a Doji as a Pin Bar or allowing trades during Asian hours. Each relaxation lets back in the noise the system was designed to remove.

Optimising the ADX Threshold

Periods of 14 and threshold levels of 20 and 25 are the conventional ADX settings, and they are valuable because they were not fitted to your data. Sweeping the threshold from 18 to 35 and the period from 7 to 28 will almost always find a pair of values that looks excellent on the past two years of EURUSD and performs no better than average afterwards. If a parameter change improves the backtest but you cannot explain why it should work, you have found a curve fit rather than an edge. Hold back at least six months of data that the optimiser never sees.

Revenge Trading Through a Drawdown

A trend system with a 40% win rate will regularly produce six to nine losses in a row, usually at the start of a new trend, when ADX rises and then fails more than once before the real move begins. Set a maximum drawdown before going live, and treat reaching it as a reason to review the system, never as a reason to increase size. On this system, 10% of equity is a reasonable line.

Build Strategy using Arconomy

You can build the EURUSD ADX Trend Strength Filter in the Arconomy Strategy Designer from nine rule blocks. ADX acts as the gate, and the EMA, Pin Bar and session rules decide when an open gate becomes a trade. Each row below corresponds to one rule block on the canvas.

Backtest Considerations

Test on at least 24 months of 30-minute EURUSD data, and make sure the sample includes both regimes the pair cycles through: long low-ADX consolidations, where the system should be almost inactive, and macro-driven trends, where it should earn its returns. A window containing one large trend will make the results look far better than they are. The ADX, Pin Bar and session filters together are very selective, so check the trade count. Fewer than 100 trades is too small a sample to trust. If you are short, extend the test window rather than loosening the filters.

Focus on three metrics. Profit factor should be above 1.3 after costs. Maximum drawdown should be judged against per-trade risk: at 1% risk, a 10–14% peak-to-trough drawdown is normal for a trend system with a win rate below 45%. Last, look at how profits are distributed. Trend systems naturally rely on a minority of large winners, but if three trades account for the entire profit, the edge depends on rare events. The Arconomy backtesting engine provides the trade-by-trade breakdown you need to check this.

EURUSD costs are low but not zero, and they change through the day. Spreads are tightest during the London–New York overlap and widen after 16:00 UTC and around ECB and US data releases. The session filter avoids the worst of that, but the backtest should still use a spread that varies by session rather than one fixed value. Include slippage on stop exits, because an ATR stop hit during a fast reversal often fills past its level. Finally, check that the Pin Bar detection in the backtest uses the same candle data and session boundaries as your broker. A daily candle that closes at 17:00 New York time changes the swing levels used to validate a Pin Bar.

Credits

The strategy idea originated from the following YouTube channel. Concepts have been adapted and structured for systematic implementation by Arconomy.

SayySooFlyy's short on the Vela strategy covers simulating account growth and trading to a fixed schedule rather than whenever the chart looks interesting. This post turns that schedule-first discipline into a defined London–New York session window and pairs it with a measurable ADX trend gate, so the rules decide when to trade on the EURUSD 30-minute chart instead of the trader.

► Watch on YouTube