Supertrend Calculation and Why Implementations May Differ
Summary
The document presents a proposed Supertrend formula using the midpoint of the high and low prices, the Average True Range, and a multiplier. It selects the upper or lower band according to whether the close is above or below the midpoint. Two charting-software examples are provided, along with the author’s calculations, which do not match the displayed indicator values.
This is a troubleshooting question rather than a verified implementation guide: it gives no accepted answer or evidence identifying the cause of the discrepancy. The examples show that the stated one-period formula alone may not reproduce a charting platform’s output. Supertrend implementations can depend on details such as how ATR is smoothed, how bands carry forward between bars, and how trend direction switches; these details are not specified in the document. The supplied values and formula are useful for framing a check, but they do not establish which convention the charting software uses.
Key ideas
- The proposed Supertrend bands use the high-low midpoint adjusted by a multiple of ATR.
- The displayed examples do not match the author’s direct calculations from that formula.
- ATR smoothing and rules for carrying bands across bars can affect indicator values.
- The document does not establish the platform’s precise calculation convention.
Tags
Full text
# Supertrend Indicator # Supertrend Indicator I'm trying to implement the Supertrend indicator. Unfortunately, I can't calculate the values that my chart software generates. At the moment, I don't fully understand my problem. You can add the Average True Range and Supertrend indicators here and look at the stock prices, but the values don't add up (Chart ). The formula to calculate the Supertrend should be: if Close > (H + L) / 2: (H + L) / 2 + M * ATR else: (H + L) / 2 - M * ATR For example values to calculate the supertrend provided by the charting software would be: | Date | Open | High | Low | Close | Supertrend | ATR | | 01.05.22 | 13996.82 | 14589.45 | 13380.67 | 14388.35 | 13834.08 | 1226.26 | | 01.06.22 | 14478.37 | 14709.38 | 12618.68 | 12783.77 | 16180.90 | 1312.71 | With the formula I would calculate 16437.58 (instead of 13834.08) and 11038.61 (instead of 16180.90). Can anyone help?
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.