Direct answer: what “swing definition” means, with a worked example
A swing definition is a rule set for labeling parts of a price chart as swing highs and swing lows based on how price moves over time. A worked example clarifies the procedure by turning vague ideas (“a move up then down”) into explicit steps (what counts as a new high/low, and when you stop and declare a swing).
Below is one transparent numerical scenario that you can replicate. It is informational only and does not assume live market data.
Mechanism or definition: the input rules you must state
To make any swing definition example verifiable, you must define at least these elements:
-
Data and time steps: You need a sequence of prices in order (for example, the close of each bar). Assume a fixed bar interval.
-
Swing size rule: Many definitions require a minimum move to avoid labeling tiny noise as swings. This can be expressed as a percentage, a pip distance, or an absolute price distance.
-
Lookback/neighbor rule: You need a way to decide whether a point is a local high (higher than nearby candidates) or a local low.
-
Decision timing: You must choose whether you confirm a swing only after you see later bars (common) or whether you try to label immediately (usually less reliable).
In the worked example, I will use a simple, explicit rule set that is easy to check by hand.
Evidence or example: a transparent worked scenario
Assumptions (state everything used)
- You observe a sequence of 10 bar closes in chronological order.
- The values represent a single instrument’s price level.
- Minimum swing magnitude: a swing high or low must differ from the previous opposite swing candidate by at least 2 units.
- Local extremum rule:
- A swing high occurs at bar t if its close is higher than both the close at t-1 and t+1.
- A swing low occurs at bar t if its close is lower than both the close at t-1 and t+1.
- Confirmation rule: you can only confirm a swing at t after you see bar t+1.
Input data (bar closes)
Bar 1 to 10 closes:
- B1=100
- B2=101
- B3=103
- B4=102
- B5=98
- B6=99
- B7=101
- B8=100
- B9=97
- B10=98
Step-by-step labeling
Step A: Identify local highs/lows (ignoring the 2-unit rule for the moment).
- Bar 3: 103 is higher than B2=101 and B4=102 ⇒ local high candidate.
- Bar 5: 98 is lower than B4=102 and B6=99 ⇒ local low candidate.
- Bar 7: 101 is higher than B6=99 and B8=100 ⇒ local high candidate.
- Bar 9: 97 is lower than B8=100 and B10=98 ⇒ local low candidate.
Step B: Apply the 2-unit minimum swing magnitude.
- First, accept the sequence starting point as:
- Swing high at B3=103 (no prior opposite swing exists, so we take it as the first labeled extremum for illustration).
- Next opposite candidate is the swing low at B5=98.
- Difference: 103 − 98 = 5 ≥ 2 ⇒ accept swing low B5.
- Next swing high candidate is B7=101.
- Difference from last labeled low: 101 − 98 = 3 ≥ 2 ⇒ accept swing high B7.
- Next swing low candidate is B9=97.
- Difference from last labeled high: 101 − 97 = 4 ≥ 2 ⇒ accept swing low B9.
Resulting labeled swings (what your definition produced)
- Swing high: B3 (103)
- Swing low: B5 (98)
- Swing high: B7 (101)
- Swing low: B9 (97)
How to independently verify
- Write down the same bar closes.
- Check each internal bar (bars 2–9) against its immediate neighbors to confirm local high/low.
- Enforce the 2-unit minimum relative to the last accepted opposite swing.
- Confirm you didn’t label anything at bar 1 or bar 10 under this neighbor rule because both sides aren’t available.
Limitations and risks: where swing definition can fail
Even a correct calculation can produce misleading labels if your rules don’t match how the data behaves.
- Rule sensitivity (mechanics depend on choices)
- Changing the minimum swing magnitude (2 units in the example) can remove or add swings.
- Using only immediate neighbors is sensitive to single-bar noise; a definition with wider lookback can label fewer, larger swings.