What is a worked example of Divergence Reversal?

Divergence reversal: definition and the key moving parts

Divergence reversal refers to an observed mismatch between price action and a second measure (often an indicator). In practice, the “reversal” part is not automatic; it is a claim that the mismatch may precede a change in price direction.

A worked example is useful because it forces explicit assumptions: what exactly counts as “divergence,” what indicator values you use, how you decide “reversal,” and what you assume about measurement timing. This article uses a hypothetical scenario (no live prices).

Worked numerical scenario (with every assumption)

Assumptions used for the example

1. Market: a currency price series where “price” is measured once per time step (for example, each candle close).
2. Time window: 10 time steps, labeled T1 to T10.
3. Price values: we use a simplified sequence of closes.
4. Indicator: we use a generic momentum-style indicator value “Ind,” computed externally. For this example, Ind values are given (we do not assume a particular formula).
5. Divergence rule (bearish): at two swing points, price makes a higher high while Ind makes a lower high.
6. Reversal rule (practical): after the divergence is identified, we label a “reversal” only if price later falls by at least 1.0 unit from the divergence swing point, within the next 3 time steps.
7. No transaction costs, slippage, or spreads are included, so the outcome is purely descriptive.

Data for the example

We choose two swing highs:

• Divergence swing point 1: T3
• Divergence swing point 2: T6

Price closes:

• T3 price = 100.0 (Ind at T3 = 60)
• T6 price = 101.0 (Ind at T6 = 55)

By the bearish divergence rule:

• Price made a higher high (101.0 > 100.0)
• Indicator made a lower high (55 < 60) So divergence is present at T6.

Later price movement after T6:

• T7 price = 100.6
• T8 price = 100.2
• T9 price = 99.0

Now apply the reversal rule:

• Starting reference: divergence swing point price at T6 = 101.0
• Required drop: 1.0 unit
• T9 price = 99.0, so the drop is 101.0 − 99.0 = 2.0 Since the 1.0-unit drop happens within the next 3 steps (T7–T9), we classify this as a “reversal” in this example.

What the example demonstrates (and what it doesn’t)

This scenario shows a complete, checkable chain:

• divergence criteria were defined,
• the divergence conditions were met numerically,
• and the later price behavior met the reversal threshold.

It does not imply that divergence always leads to reversal, only that the defined conditions were satisfied in this hypothetical dataset.

Limitations, risks, and failure modes to watch for

Divergence reversal is vulnerable to several issues that a worked example helps you name.

1. Subjective swing selection: deciding where swing highs/lows are formed can change whether divergence exists. If you pick different turning points, the divergence may disappear.
2. Threshold sensitivity: the reversal threshold (for example, “drop by at least 1.0 unit”) is arbitrary in this example. Different thresholds can change whether the outcome is labeled a reversal.
3. Indicator noise and smoothing: many indicators fluctuate. If the indicator is volatile, “lower high” and “higher high” readings can reflect noise rather than a structural shift.
4. Timing mismatch: even if reversal eventually occurs, it may happen outside your verification window (here, within 3 steps). Late reversals may be missed by design.
5. No causality guarantee: divergence is an observation about behavior, not a guarantee that price must reverse.

Outcomes vary with market conditions, execution details, and the specific indicator settings and data frequency you use. Historical relationships do not establish future results.

How to verify the concept yourself (without relying on prediction)

To independently verify divergence reversal in any dataset, keep the process explicit:

• Write your definitions: exact divergence rule, which points count as swing highs/lows, and your reversal threshold and time window.
• Recompute with the same definitions across many historical segments.
• Track what fraction of divergence events meet your reversal rule, and how often reversals occur without divergence.

If your definitions are unclear, you cannot reproduce results reliably. If your dataset changes (time period, volatility regime, or indicator settings), your measured relationships can also change.