Anchoring the leg, drawing the levels, and requiring a turn before acting
Five steps, of which the first is subjective and determines everything else, and the fourth is the one most people skip.
Key takeaway
- The anchor choice is the whole method — different swing selections produce entirely different levels
- A level is a place to look, not a signal; the entry requires a confirmed turn
- 78.6% is the natural invalidation, because a retracement that deep suggests the impulse was not an impulse
The anchor decides everything
You draw the retracement from the start of a move to its end. Which start and which end is a judgement, and two traders looking at the same chart routinely pick different swings and arrive at levels several percent apart.
- Use the most recent clean impulse. A move that is visibly one directional push, not a series of overlapping swings.
- Use extremes, not closes. The high and low of the move, consistently — mixing wick and body anchors between charts is how the levels stop being comparable.
- Decide the anchor before you look at where price currently is. Choosing a swing that puts a level conveniently under the current price is the most common way this method is abused.
- If the impulse is not obvious, there is no setup. Ambiguity at this step means every level downstream is arbitrary.
The three levels and what each tends to mean
| Level | Typically read as |
|---|---|
| 38.2% | A shallow pullback; associated with strong trends |
| 50% | The most-watched level, and not a Fibonacci ratio |
| 61.8% | A deep pullback; the last level before the setup weakens |
| 78.6% | Not an entry — the invalidation point |
A strong trend tends to retrace shallowly and a weak one deeply, which means the level price reaches is itself information — a pullback that runs to 61.8% is telling you something about the trend's strength, whether or not it turns there.
The confirmation step, which is not optional
Price reaching a level is not a signal. It is a place to start paying attention. The entry requires evidence that the pullback has actually ended.
- A reversal candle at the level — a hammer, an engulfing bar, a strong close off the low.
- A structure break on a shorter timeframe — the pullback's own sequence of lower highs breaking upward.
- A volume shift — selling drying up into the level, buying appearing on the turn.
- Any of these, applied consistently. Which one matters far less than deciding in advance and not adding a fourth when the first three say no.
The 78.6% stop and the measured target
The stop goes beyond the 78.6% retracement. This is the method's most defensible component and it does not depend on the ratio being special: a move that retraces more than three quarters of the previous leg was probably not an impulse in a trend, and the reason for the trade has gone.
The first target is the prior swing extreme. An extension of the impulse — projecting the original leg's length from the turn — is the conventional second target, and it carries all the same evidential caveats as the retracement levels themselves.
Common questions
- Wicks or closes for the anchor?
- Pick one and use it everywhere. Extremes including wicks is the more common convention and captures the full range of the move. What produces unusable results is switching between them, which is what people do when the levels do not fall where they want.
- What if price stops between two levels?
- It frequently does, which is itself informative about how precise these levels are. Treating each level as a zone rather than a line — a band of perhaps half a percent — matches the actual behaviour better and reduces the temptation to redraw.
- Should I use 23.6% and 78.6% as entries too?
- Adding levels means the pullback is almost always near one of them, which makes the framework unfalsifiable. Three entry levels plus one invalidation is already generous coverage of a pullback; more is not more information.
These are documented methods described for study. Nothing here is investment advice, a recommendation, or a claim about future returns — every system on this page has losing periods, and the pages say where.
Reading about a system is not having one.
Plutux is where you write your own rules down, test them against real data, and keep the record your memory would otherwise rewrite. Join the waitlist for early access.