SAR (Parabolic SAR): calculation and reading
SAR draws a series of dots below or above price that gradually move closer as the trend develops, and switch sides when price crosses them.
Updated:
Loading chart…
Uses the prices of TSMC (2330) over roughly the last 120 trading days only to show how this indicator is drawn; it is not a stock recommendation. Source: Taiwan Stock Exchange.
Calculation
SAR is updated each day with this formula: new SAR = previous SAR + acceleration factor × (extreme point − previous SAR).
- Extreme point: the highest price of the current trend (in an uptrend) or the lowest price (in a downtrend).
- Acceleration factor: it starts at the “step” (default 0.02) and increases by one step each time a new extreme point is made, up to a cap of 0.2; both values can be adjusted on this site.
- When price crosses SAR, the trend is treated as reversed: SAR jumps to the other side of price, its starting point is set to the previous trend’s extreme point, and the acceleration factor returns to its starting value.
How to read it
- SAR below price means the current phase is judged to be an uptrend; SAR above price means a downtrend.
- The closer the dots are to price, the longer the trend has been running and the larger the accumulated acceleration factor.
Limits
SAR is only meaningful when a trend is clear; in a sideways market price crosses it repeatedly and SAR switches sides often. It describes the price path that has already happened and does not predict the future.
Where the idea comes from
Parabolic SAR was introduced by J. Welles Wilder Jr. in his 1978 book New Concepts in Technical Trading Systems. This article explains it in our own words.
This article only explains the calculation and how to read a chart. It is not investment advice.
See it on this site
Sources
- J. Welles Wilder Jr., New Concepts in Technical Trading Systems (1978)
Citing and copyright
This article is original content of “Tenmizu FinTech”. You are welcome to cite it; please credit the source and link to this page. Concepts or data that come from others are credited under “Sources”; their copyright belongs to the original authors.