Swing highs and swing lows: the core idea and what they do well
Swing highs and swing lows are chart reference points that aim to summarize market structure by marking local peaks (swing highs) and local troughs (swing lows). In practice, they turn noisy price movement into a sequence of turning points that traders can discuss in plain language: “price moved from this turning point toward that one.”
This concept is most useful when you treat those points as descriptive labels of past price action, not as a predictive mechanism. When you keep that boundary clear, you can use swing points to help explain structure, range behavior, or momentum shifts.
How the limitations show up in real use
Swing highs and swing lows are constrained by the fact that the “swing” is not uniquely defined. Different rules can produce different swing points from the same price series. For example, a swing made up of several nearby highs may be marked as one swing high under a lenient rule, but split into multiple swing highs under a stricter rule.
Because of that, at least three common failure modes can appear:
-
Rule sensitivity (different settings, different swings). If you change the lookback logic, the minimum distance between swings, or the way you treat equal highs/lows, the marked swing sequence can change. That makes comparisons across timeframes or across providers less reliable.
-
Uncertainty from forming bars (in-progress candles). Swing points are typically only confirmed after later price movement. A level that looks like a swing high before subsequent bars can later be invalidated when a higher high appears. This creates a “what you saw vs what you later learned” problem, especially in backtesting.
-
Regime change (structure may stop behaving the same way). Markets can shift between trending, ranging, or high-volatility conditions. A swing-based description that was consistent in one regime may become noisy or ambiguous in another, reducing the clarity of the turning-point sequence.
Evidence and examples: where the concept can mislead
Consider a simplified scenario where you label swing highs and swing lows on a price chart. If price later makes new highs above a previously identified swing high, then the earlier swing high label does not reflect the final “local peak” according to your confirmation rule. That means the historical mark you use for analysis was dependent on future information.
Another example is a sideways market with frequent small reversals. In such conditions, there may be many legitimate “local” peaks and troughs, but not all of them are equally meaningful. The limitation is not that swing highs/lows are wrong; it is that the chart can generate too many competing candidates, and a chosen rule may select some turning points while ignoring others.
These examples highlight an important constraint: historical relationships between swing points and subsequent movement do not automatically establish future results. Costs like bid/ask spreads and execution frictions can also change the outcome of any measurement, even when the structural interpretation seems consistent.
Limitations and risks to verify independently
1) No single “correct” swing map
Swing identification depends on assumptions: the timeframe you use, the required separation of highs/lows, and the confirmation logic. To verify this, you can try marking swings with different reasonable rule sets and observe how often the swing sequence changes.
2) Backtest bias from confirmation timing
If your swing labeling requires later confirmation, then any analysis that implicitly assumes you knew the swing points at the time they were forming can overstate certainty. A practical way to verify this limitation is to compare “confirmed” swings (after the full rule is satisfied) versus “would-have-seen” swings (early-looking candidates).
3) Ambiguity in strong trends or fast volatility
In strong trends, swing lows (or swing highs) may update in frequent, uneven steps, producing a structure that is technically consistent but hard to interpret. In fast volatility, small fluctuations can qualify as swings under lenient rules, increasing noise.
4) Market context and costs matter
Even for purely descriptive use, you should account for non-structural factors when translating structure into real-world outcomes. Outcomes vary with market conditions, costs, execution, and jurisdiction. Historical patterns do not ensure forward performance.