Direct answer: verify Marubozu information with a source hierarchy and reproducible checks
To verify information about Marubozu, you can (1) adopt a stable definition, (2) cross-check that definition across independent references, and (3) reproduce at least one set of labeled candles using the same calculation assumptions (open, close, and wick handling). Then you confirm limitations: small measurement differences, timeframe changes, and how different chart platforms record candle components can all change whether a candle is labeled “Marubozu.”
Because there are no universally enforced standards for “exactly how to measure,” verification relies on consistency of criteria and reproducibility, not on trust in a single explanation.
Mechanism and definition: what you are trying to verify
Marubozu is commonly described as a single candlestick with a body that dominates the candle, with either very small or absent wicks (upper wick for bearish/upper side; lower wick for opposite side, depending on the exact version). The two key parts you should verify in any source are:
- Body dominance rule: how the source decides the body is “large enough.” Some descriptions use “little to no wick,” others imply a specific wick-size tolerance, and some use relative comparison (body length vs. total range).
- Direction rule: whether the candle is treated as bullish (close above open) or bearish (close below open). Even within stable terminology, sources may differ on whether they require strict equality of open/close with extremes.
To separate stable mechanics from variable conditions, keep your terminology fixed: “Marubozu” refers to a candle-shape classification. It does not, by itself, guarantee future movement.
Evidence and example: reproducible verification steps
Use the same step order every time:
- Write down the criterion in plain rules. Example assumption for reproducibility: “A bullish Marubozu has a long body with an upper wick that is at or near zero relative to the full candle range; a bearish Marubozu has the opposite.” If a source defines a tolerance (for instance, “upper wick must be smaller than X”), copy that rule exactly.
- Collect 2–4 independent descriptions. Prefer references that explain methodology (for example, charting documentation, educational materials that include explicit candle-shape rules). Compare whether they use the same body/wick requirements.
- Create a test set from your own data. Screenshot or export candles for a few dates/timeframes where you think a Marubozu exists.
- Measure with fixed assumptions. For each candle, record: open, close, high, low (or the platform’s equivalent). Then compute body size and wick sizes consistently. Example assumption: “wick length is the distance from the body edge to the corresponding extreme.”
- Apply your written criterion. If your computed wick sizes meet the rule, the candle qualifies under that criterion; if not, the label should not be applied.
- Check agreement. If different sources label the same candles differently, the conflict is likely due to different thresholds or wick-handling conventions—an important part of verification.
Limitations and failure modes: what can go wrong
Even when you follow a consistent workflow, there are material limitations:
- Ambiguous thresholds: if sources say “little to no wick” without a measurable tolerance, two people can reach different classifications.
- Timeframe effects: a candle on one timeframe can include wick noise that disappears on another; classification can change when you switch aggregation.
- Platform differences: candle calculations and display formatting can differ across charting tools, especially regarding rounding and how they present tiny wick components.
- Overgeneralization: a candle-shape label is not a standalone predictive signal. Historical or textbook examples do not establish future results.
Outcomes also vary with trading costs, execution quality, and jurisdictional factors, but those are separate from verifying the visual definition. Verification should focus on whether the candle meets the stated classification rules.
Verification or next question: what to check next
After you verify the definition and reproduce at least one set of labeled examples, the next question is usually whether the “Marubozu” label is operationally consistent across timeframes and platforms. To do that, rerun the same measurement steps on:
- at least two timeframes (for example, a slower and faster chart), and
- at least two charting views/tools (if you have access), using the same measurement rule.