The first tool almost everyone learns, and one of the least tested. A trendline claims that an area matters and that it moves at a certain rate — so the honest test is against the same area held still.
A trendline joins two swing points of the same kind and projects them forward, claiming both that the level matters and that it drifts at a fixed rate; a channel is the same line copied onto the opposing swings.
Two swing lows, each higher than the last, joined by a line and projected forward. When price falls back to that line, buyers are supposed to be waiting.
Put like that, the claim has a hidden half. A trendline says two things at once: that this area matters, and that the area moves at a particular rate. The first claim is unremarkable — price reacts near recent lows. The second is the interesting one, and it is the one nobody tests.
So this page tests it directly. Every line is measured twice: once as drawn, sloping up from the two lows, and once as a flat line at the second low's price. Same swings, same forward rule, same instruments. Whatever the slope is worth shows up as the difference.
A friend has been arriving five minutes later each week. You could plan around "they turn up about half past" or around "they turn up half past plus five minutes for every week since March".
The second is a stronger claim. It might be better, and it might be reading a pattern into three data points. The only way to find out is to make both predictions in advance and see which is closer more often — which is exactly what the table below does, 7,108 times.
A line is built from two consecutive swing points, projected sixty hours forward, and the first bar that reaches it counts as the test. Success means price then carried all the way back to the most recent opposing swing. Failure means an hourly close through the line first.
| Instrument | Lines drawn | Sloped line reached | Sloped held | Flat line held | Difference |
|---|---|---|---|---|---|
| XAUUSD | 1,704 | 83.7% | 26.8% | 21.3% | +5.5 pts |
| GBPUSD | 1,600 | 82.7% | 28.4% | 20.4% | +8 pts |
| EURUSD | 1,572 | 81.8% | 24.9% | 21.5% | +3.4 pts |
| US30 | 1,085 | 75.2% | 25.2% | 21.8% | +3.4 pts |
| NAS100 | 1,147 | 78.8% | 25.2% | 22.7% | +2.5 pts |
What was counted: Two consecutive swing points of the same kind, six to eighty hours apart, forming a rising or falling line. The line is projected sixty hours forward and the first bar reaching it is the test. Held means price then carried back to the most recent opposing swing; broke means an hourly close through the line first. The control is a horizontal line at the second swing's own price, projected and tested by the identical rule, so the only difference between the two results is the slope. Both directions included.
The slope is worth something. On gold the sloped line held 26.8% against 21.3% for the flat one — 5.5 points. It came out ahead on 5 of five instruments, between two and a half and eight points.
There is a second difference worth noticing. The sloped line was reached 83.7% of the time within sixty hours; the flat line only 64.8%. A rising line comes to meet price, so it gets tested far more often. If you want the trade to actually happen, that matters as much as the hold rate does.
Roughly a quarter, for something the whole industry draws, deserves an explanation — and it is the definition of success rather than a verdict on trendlines.
Holding here means price carried the whole way back to the most recent opposing swing. That is a full leg, not a bounce. Plenty of touches produced a perfectly tradeable move that fell short of it and were scored as failures.
The test is deliberately demanding, and identical for both lines, which is what makes the comparison sound. Read the gap between the two columns as the finding. Do not read either column as "trendlines work a quarter of the time" — under a gentler target both numbers rise together.
A channel is a trendline with a copy of itself pinned to the opposing swings — support below, resistance above, both moving at the same rate. Everything above applies to each edge; the parallel line adds no new claim, it just draws the same one twice.
The practical value is not prediction but proportion. When price stops reaching the far edge, the move is losing energy, and that is visible earlier on a channel than on a single line. Treat the upper edge as somewhere to take profit rather than somewhere to reverse — a channel edge is a target, not a signal.
Aisha marks two swing lows on gold, 4,008 and 4,018, six hours apart. Her line rises about one dollar and sixty cents an hour. She also drops a flat line at 4,018.
What happens. Fourteen hours later price falls back and touches the sloped line at 4,041, well above the flat line. She has a trade the flat line would not have given her for another two days, if ever.
What she expects. Slightly better than the flat line — a few points — and considerably more opportunities. Both of those come from the slope, and the second is the larger practical effect.
Her stop. Below the line, at a distance chosen from what price actually does rather than from where the line looks tidy. The Stop Loss Reality Checker gives the share of real gold windows that reach any given distance.
What she refuses to do. Redraw the line when price closes through it. The line was a prediction; it was wrong; a new line through the new low is a different prediction and pretending otherwise is how people end up with a chart full of lines that have never once been wrong.
The flat control costs one click and turns every trendline you draw into an experiment. After thirty touches you will know whether the slope is earning its place on your instrument, instead of taking this page's word for it.
Test your lines on real candles →Run the drill in the free backtester → Free, no sign-in to begin.
Do trendlines actually work?
Better than a flat line at the same swing, by a small margin. On gold the sloped line held 26.8% of the time against 21.3% — 5.5 points — and the slope came out ahead on 5 of five instruments.
Why are the hold rates so low?
Because success is defined strictly: price had to carry all the way back to the most recent opposing swing, not merely bounce. The same demanding target is applied to both lines, so the comparison is fair even though both numbers look modest.
Should I draw trendlines from wicks or bodies?
Either, consistently. This measurement used the extremes. What matters far more than the choice is not switching between them, because switching is how a line gets quietly fitted to the outcome you already know.
How long should a trendline stay on the chart?
Not indefinitely. A projected line will eventually intersect any price, so an eternal line has stopped making a prediction. Sixty hours was the cap here; whatever you choose, choose it in advance.
Is a channel better than a single trendline?
It adds no new claim — the parallel line is the same slope drawn twice. Its value is in showing proportion: when price stops reaching the far edge, the move is losing energy. Treat the far edge as a target rather than a reversal signal.
What if price breaks my trendline?
Then that line was wrong, and drawing a new one through the new low is a separate prediction. Redrawing to keep price on the correct side is the most common way traders convince themselves trendlines never fail.
One extra horizontal at the second swing turns every trendline into a controlled experiment. Thirty touches later you will know what the slope is worth on your instrument, which is more than almost anyone drawing these lines can tell you.
Open the free backtester →Beginner exploration
Open each answer for a plain-language way to read Trendlines and channels, test it carefully and decide what to explore next.
This page focuses on “Trendlines and channels”.Every trendline makes two claims: that an area matters, and that it moves. Measured across 7,108 lines on five instruments, the sloped line held 26.8% against 21.3% for a flat line at the same swing — and was tested far more often.For “Trendlines and channels”, a beginner should identify what the learning guide measures, assumes or teaches before acting on its conclusion.Treat this page's account of “Trendlines and channels” as a learning reference rather than a prediction, signal or promise of future performance.
For “Trendlines and channels”, translate the idea into a definition you could apply the same way on two different charts.While exploring “Trendlines and channels”, work through one example slowly and record which inputs or observations determined the result.Keep your “Trendlines and channels” record honest: list the limitation or counterexample before using the concept in a trading plan.Before leaving “Trendlines and channels”, practise the definition on unseen history and review consistency before judging performance.
Turn one idea from “Trendlines and channels” into a rule with explicit inputs, dates, costs and pass-or-fail conditions.Ask AI to expose missing assumptions in that “Trendlines and channels” test, not to guess the next market move.Use the FXAbsolute AI Backtesting Lab to inspect calculations connected to “Trendlines and channels” and the assumptions behind them.Reproduce any important “Trendlines and channels” result and reserve unseen data before deciding that an apparent pattern is useful.
Continue your exploration of Trendlines and channels with the beginner AI prompt guide, or inspect public calculations in the AI Backtesting Lab.