Fibonacci extensions project candidate price levels beyond a completed swing. They can make a target rule explicit, but they do not tell a trader that price will reach, reject, or reverse at the line.
That distinction matters. A retracement asks where a move might pull back inside an existing range. An extension starts with a move and its correction, then projects outward. The existing Fibonacci retracement explainer covers the first job; this article covers the separate A-B-C projection method.
What a Fibonacci extension measures
A Fibonacci extension is a charting construction built from three selected prices: the start of a move, its end, and the end of the pullback. Charles Schwab describes extensions as a way to identify potential support, resistance, or target areas after price has moved outside its prior range; it lists 127.2%, 161.8%, 200%, 261.8% and 423.6% among commonly cited extension levels (Schwab: Fibonacci retracement and extension levels).
The ratios are mathematics, not market evidence. StockCharts explains the sequence-based roots of 23.6%, 38.2% and 61.8%, while Schwab explains that 161.8% belongs to the same ratio family (StockCharts ChartSchool: Fibonacci ratios, Schwab: Fibonacci retracement and extension levels). A platform can draw the line exactly. It cannot establish that the line has predictive power.
Upward A-B-C swing:
A = swing low
B = swing high
C = pullback low
extension at ratio r = C + r × (B - A)For a downward swing, the direction reverses. The important part is not the algebra. It is the choice of A, B and C. A different swing produces a different map.
Why extensions are not retracements
Retracements and extensions use related ratios but answer different chart questions. A retracement measures a counter-move inside the prior A-B leg. An extension projects from the later C point beyond B in an up move, or beyond B in a down move.
| Tool | Inputs | What it maps | What it does not establish |
|---|---|---|---|
| Fibonacci retracement | A completed swing | Possible pullback zones within that swing | That a pullback must stop at a ratio |
| Fibonacci extension | A move plus a pullback | Candidate levels beyond the earlier extreme | That price will reach or reverse at a projected level |
This is a material difference, not cosmetic terminology. A trader may use a retracement to describe where a correction became interesting and an extension to define a possible exit reference. If both tools are drawn after the fact, though, they share the same failure mode: the analyst can choose the swing that makes the finished chart look unusually tidy.
Where the discretion hides
Extensions often look objective because every horizontal line has a precise label. The subjectivity arrives earlier.
- Swing selection: Is A the wick low, the closing low, or a larger-timeframe pivot?
- Pullback completion: What confirms C: the first bounce, a close, a new high, or a minimum reversal distance?
- Ratio set: Are all plotted ratios eligible, or was one selected after it looked best?
- Direction and context: Is the same rule used in trends, ranges, gaps and high-volatility sessions?
- Exit treatment: Does reaching a line mean exit, partial exit, a trailing rule, or simply an observation?
Schwab says Fibonacci tools may be combined with other analysis but that no technical tool reliably predicts future price movement. StockCharts likewise describes Fibonacci levels as alert zones rather than hard reversal points (Schwab: Fibonacci retracement and extension levels, StockCharts ChartSchool: Fibonacci retracements as alert zones). That caution applies even more strongly when the tool's three anchors are discretionary.
The popular explanation that many traders watch the same ratios is plausible as a description of attention. It is not proof of a repeatable payoff after spread, slippage and missed fills.
How you'd actually test Fibonacci extensions
Do not test whether a line looks respected. Test one rule that would have been executable at the time.
- Define the market, timeframe, data source and trading session.
- Define A, B and C with a mechanical pivot rule. For example, every pivot may require a specified number of completed bars on each side.
- Freeze the ratio set before examining results. Do not let the best-looking projected line choose itself.
- State the setup independently of the extension. An extension is usually an exit or reference rule; it is not automatically an entry.
- Define the order, fill, stop, exit and expiry rules, including whether a touch, close or gap through a level counts.
- Include the applicable spread, commission, slippage and swap.
- Compare the extension rule with a simple baseline, such as a fixed reward-to-risk exit or the prior swing range.
- Hold out data before selecting among pivot rules and ratio sets.
| Comparison | Question it answers |
|---|---|
| Extension exit versus fixed reward-to-risk exit | Did the projection add value beyond a simple exit rule? |
| Mechanical pivots versus hand-selected swings | Does the result survive removing hindsight from A, B and C? |
| One declared ratio set versus many tested sets | Was the chosen ratio selected by evidence or by luck? |
| In-sample versus out-of-sample | Did the full rule survive data it did not see during design? |
This is not an argument against using a charting tool. It is an argument against letting a labelled line stand in for a tested method. The same discipline applies to out-of-sample testing and to verifying a cTrader backtest: record the rule, model real costs, and keep the deciding data unseen.
Frequently asked
Are Fibonacci extensions price targets?
They are projected reference levels, not price forecasts. Schwab presents them as potential support, resistance or target areas, and also states that no technical tool can reliably predict future price movement (Schwab: Fibonacci retracement and extension levels).
What is the difference between a Fibonacci retracement and extension?
A retracement maps possible pullback zones within an earlier swing. An extension uses a move and its pullback to project candidate levels beyond the prior extreme.
Which Fibonacci extension level is best?
There is no universal best level. The familiar labels are conventions drawn from a ratio family; choosing one after viewing its historical outcome is selection bias. A declared rule must be compared with alternatives on unseen data.
The stubborn takeaway
Fibonacci extensions can turn a vague target into a visible rule. They become useful only when the rule is precise enough to survive a backtest that has not seen the answer.