Record what precedes the first visible bar
True Range can depend on the previous close as well as the current high and low. A download beginning with one visible bar may omit that earlier close. Treating its high-minus-low as a complete True Range silently adds a starting convention that another chart may not use.
Preserve the bar interval, price unit, previous-close coverage and any initialization rule. The example below uses USD-labelled teaching prices and a three-period window to keep the arithmetic visible. It does not select an indicator setting or reproduce a live market series.
Compare two possible previous closes for one fictional bar
For a bar with high 112 and low 108, the within-bar range is 4. If the supplied previous close is 110, the other two absolute distances are 2 and 2, so True Range is 4. If the previous close is 130, the distances are 18 and 22, so True Range is 22.
Both possibilities are consistent with that same current high and low. If the prior close is missing, the current bar alone cannot choose between them. Do not describe a default range of 4 as independently verified gap-aware data.
| Teaching prior close | High−low / distance to high / distance to low | True Range |
|---|---|---|
| 110 | 4 / 2 / 2 | 4 |
| 130 | 4 / 18 / 22 | 22 |
| Not supplied | Current bar insufficient | Unresolved |
Carry the initialization difference through the recurrence
Suppose the next two teaching True Ranges are 5 and 6. An arithmetic three-value seed using 4,5,6 is 5. Substituting 22 for the first value produces a seed of 11. For a later True Range of 10, the stated Wilder recurrence (previous ATR×2+new TR)/3 gives about 6.6667 or 10.6667 respectively.
The recurrence did not malfunction: it received different seeded histories. Preserve the exact seed, its input window and the point at which the recurrence begins. A short downloaded sample need not match a chart that carries a much longer initialization history.
Distinguish source precision from an output label
Retain sufficient input precision for the comparison and round the displayed result only under a declared rule. Two outputs shown to one decimal place can hide a small difference, while a different seed can create a large one. Neither issue should be repaired by replacing source values to make two screens agree.
Also keep the bar boundaries and adjustment convention. Different bars or price series can produce different ranges even when the lookback label matches. This record does not establish which vendor feed is correct or authenticate the underlying observations.
Report a calculation boundary, not a market prediction
A useful audit states whether the first previous close is supplied, how the seed was formed and which subsequent recurrence was used. Missing initialization evidence remains a documented limitation. The cited indicator reference supports the calculation components, not the fictional data.
ATR is a calculation on supplied price history, not proof of a future direction or a suitable trade. This companion case complements a calculator by exposing an input-history issue; it does not validate the linked commentary offer or recommend a risk setting.
The optional link below leads to this site's separate crypto signals and Telegram updates offer. The record method is usable without that offer. It does not independently verify the offer, establish country eligibility or turn the teaching example into an actual transaction.
Primary sources and scope
- Fidelity: Average True Range
ATR recurrence and the role of the previous close in True Range. No current prices, indicator recommendations or trading results are used. Checked 2026-10-21.