Direct answer: what “divergence” means
In the context of “MT4 Basics,” divergence typically means that the chart’s price action and another plotted series (often an oscillator derived from price) move in different directions over a similar time window. In simple terms: price might make a higher high while the indicator makes a lower high, or price makes a lower low while the indicator makes a higher low.
Mechanics: how divergence is constructed
Divergence is not a single built-in mystery in MT4; it is a visual concept you create by comparing two lines that are both functions of price data.
-
You choose the price reference. This is usually the high/low of candles on a chart (for example, the latest peak or trough).
-
You choose the indicator reference. Commonly this is an oscillator-like line that is calculated from price (for example, a bounded momentum-style series). The indicator produces values each bar based on its formula and its chosen settings (such as lookback length).
-
You identify comparable turning points. To call something divergence, you are implicitly matching peaks with peaks (or troughs with troughs). That requires a rule for “similar time,” which can be subjective.
-
You compare directions. Typical patterns are:
- Bearish divergence: price forms a higher high, while the indicator forms a lower high.
- Bullish divergence: price forms a lower low, while the indicator forms a higher low.
Even in this simple model, divergence depends on what counts as the “next” peak or trough and how many bars you include in the comparison window.
Evidence or example: a simple, checkable scenario
Assume you look at five-candle swing points on a chart and you label two recent peaks. You observe:
- Peak A: price high is 1.2000, indicator value is 60.
- Peak B: price high is 1.2050 (higher high), but the indicator value is 58 (lower high).
Under the divergence definition above, this mismatch is what people call divergence. However, the construction includes assumptions:
- Your peaks are defined by the five-candle rule.
- The indicator is calculated with a fixed setting.
- You are using the same chart timeframe for both comparisons.
If you change any of those assumptions (timeframe, lookback settings, or the swing rule), the visible “divergence” may appear different.
Limitations and risks: confirmation limits and failure modes
1) Confirmation limits (divergence is not self-verifying)
Divergence is an observation about the past relationship between two plotted series. It does not, by itself, guarantee that a new price move must follow.
A common failure mode is treating divergence as a standalone signal rather than a hypothesis. You may see divergence, but price can continue in the same direction, or the indicator can “catch up” later.
2) Hindsight bias
Because divergence is identified after seeing turning points, it is easy to unintentionally select cases where the mismatch looks persuasive. This is hindsight bias: once a move happens, earlier observations can be retrofitted as “divergence.”
A practical way to reduce this risk is to define a rule in advance (how you pick peaks, what timeframe you use, and what qualifies as divergence) and apply it consistently.
3) Data and calculation variability
Different feeds, broker data differences, or platform settings can change candle formation and indicator values. Even without changing the market, changing indicator parameters can alter whether “peaks” align and whether the mismatch is visible.
4) Market regime sensitivity
The relationship between price momentum and the indicator is not constant across all conditions. In some environments, oscillators can remain biased or “stick” near extremes, creating frequent divergences that do not reliably correspond to reversals.
Verification or next question: how to check divergence independently
To verify divergence claims yourself, you can use a time-agnostic checklist:
- Define the comparison rule: how you select peaks/troughs.
- Fix indicator settings and timeframe.
- Check multiple instances, not only the ones that seem to “work.”
- Evaluate what happens after the identified turning points, recognizing that historical mismatches do not ensure future outcomes.
If you want to go one step further, the next question is: “Compared to what?” For example, divergence can be compared against recent trend direction, support/resistance levels, or broader context—without assuming that divergence alone predicts outcomes.