Investigating Differences Between Bollinger Band Calculations

Summary

This post raises a troubleshooting question about discrepancies between Bollinger Bands calculated with a platform’s TA.BOLL function and bands displayed on a chart. The example uses 20-period bands with a multiplier of two on one-minute data. Logged upper-band values differ substantially from the chart’s backtest display, and the author says an exchange chart appears closer to the charting platform’s value.

The post includes a short code excerpt showing how records are retrieved and indicator outputs assigned, but it does not resolve the discrepancy. It does not establish whether the source is candle timing, incomplete bars, record ordering, data feeds, or differences in indicator implementation. It is best read as an example of an unresolved indicator-validation problem, rather than as a worked solution or evidence that one calculation is correct.

Key ideas

• The post compares programmatically calculated Bollinger Bands with values displayed on a chart.
• Its example uses one-minute data and a 20-period, two-standard-deviation setting.
• The logged upper band differs substantially from both the backtest chart and an exchange chart.
• The author provides no diagnosis, so the cause of the discrepancy remains uncertain.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.