This exists to make a portfolio possible. It needs exactly the condition the trend system fails in, which is what makes the pair worth running together — when one is being chopped up, the other is being paid.
It is also the more dangerous of the two, and the danger is specific: a range that breaks is a trend beginning, and a range trader positioned at the edge is on the wrong side of it from the first candle. Every rule below exists to limit that.
---
The rules
Market and timeframes. EURUSD, EURGBP, USDCHF, AUDNZD — pairs that range more than they trend. 4-hour for the range, 15-minute for entry. Avoid gold; its ranges break violently.
Identifying a valid range — all three required:
At least two touches of each boundary, and neither has been closed beyond
Range height is at least 3× the 4-hour ATR. A narrow range is not a range, it is compression before an expansion — trade that with the triangle rules instead
No sequence of higher highs and higher lows, or the reverse
Daily structure is not trending.
If the daily is trending, you are looking at a pullback, not a range. Take it with the trend system.
Setup — all four required:
Price reaches the outer third of the range
It sweeps the boundary — trades beyond it and closes back inside on the 15-minute
The reclaim candle's body is at least 0.7× ATR(14) on the 15-minute
The opposite boundary is at least 2.5× your risk away
Condition 2 is not optional. Entering at the boundary without a sweep puts your stop exactly where everyone else's is, and range boundaries are the densest stop clusters on any chart.
Entry. Market on the close of the reclaim candle.
Stop. Beyond the sweep extreme, plus 0.5× 15-minute ATR. Wider than feels comfortable — that is the price of not being swept twice.
Target. The midpoint of the range for half the position. The opposite boundary for the remainder, moving the stop to breakeven when the midpoint fills.
Skip the trade if:
The daily is trending
Range height under 3× ATR
Reward to the midpoint is under 1× risk
A tier-1 release falls before your target would plausibly be reached
The range has held for more than about three weeks — old ranges break, and the longer one persists the more of the orders defending it have been consumed
Risk. 0.5% per trade — half the trend system's, because break risk is a real, non-normal tail. 3% per currency, 5% total.
Exits. Close immediately if a 4-hour candle closes beyond either boundary. Do not wait for the stop; the range has ended, and the position's entire premise is gone.
---
Worked example — a win
EURGBP 4-hour. Boundaries 0.8420 and 0.8510. Three touches each over nine days. Height 90 pips; 4H ATR 22. 90 ÷ 22 = 4.1× ✓. Daily shows no directional sequence ✓.
Price falls to the lower third. On the 15-minute it trades to 0.8408 — below the boundary — and closes at 0.8424. Body 16 pips; 15m ATR 7. 16 > 4.9 ✓
At 0.5% of $10,000 = $50: $50 ÷ (20 × $10) = 0.025 → 0.02 lots.
Price reaches the midpoint in eleven hours — half closed, remainder to breakeven. Two days later it reaches 0.8502 and is closed near the boundary. ≈ +2.9R.
Worked example — the loss that matters
Same range, two weeks later. It has now held 24 days.
Price reaches the lower third, sweeps to 0.8411, closes at 0.8427. Every setup condition is met.
But the skip list says: a range older than about three weeks is skipped. Most of the orders that defended those boundaries have been filled across seven touches.
A disciplined trader passes. Suppose they take it anyway.
Entry 0.8427, stop 0.8405, risk 22 pips.
Four hours later a 4-hour candle closes at 0.8391 — below the boundary. The exit rule fires: close immediately, at roughly 0.8393. −1.5R, worse than the planned stop because the exit rule sits beyond it.
Price continues to 0.8290 over the following week. A trend has begun.
This is the range trader's characteristic loss — not a normal stop-out, but being on the wrong side of a new trend from its first candle. It is why the risk is half, why the exit rule exists, and why the age filter exists.
---
Running the two systems together
Trend pullback
Range fade
Needs
A trending daily
A non-trending daily
Risk
1%
0.5%
Typical hold
2–5 days
12–48 hours
Characteristic loss
Structure breaks after entry
The range becomes a trend
The daily structure check decides which one is eligible, and it is the same check for both. They cannot both be active on the same instrument at the same time, which is exactly the property that makes them a portfolio rather than two overlapping bets.
Run both, log them separately, and after 100 trades each, check the correlation between their monthly results. If it is not clearly negative, one of them is not doing what it was chosen for.
