The normalized oscillator is in a range [0,100], with underlying calculations based on the confirmation signal algorithms which allows him to adapt to various market prices characteristics.
The oscillator can detect the current trend in the price, with an uptrend detected when the oscillator value is above 50 and a down-trend when the oscillator value is under 50.
While this is the main way to interpret the confirmation oscillator, it is also possible to use it in order to detect divergences.
One can make use of the overbought and oversold levels, buying when the oscillator crosses over the overbought level, and selling when it crosses under the oversold level.
It is possible to smooth the two previous oscillators using different smoothing methods, these includes:
Exponential smoothing (EMA)
Triangular smoothing (TMA)
Hull smoothing (HMA)
Lux smoothing (Lux)
EMA by default. The degree of smoothness provided by each method can be controlled by the confirmation smoothing setting, with a higher value returning smoother results (14 by default). Each method is further described below:
The exponential smoothing method makes use of an exponential moving average to smooth the oscillator. This method is relatively reactive but might still preserve some noisy variations of the oscillator.
The triangular smoothing method makes use of a triangular moving average to smooth the oscillator. This method is the smoothest one available but is also the less reactive one.
The Hull smoothing method makes use of a Hull moving average to smooth the oscillator. This method is the most reactive one available.
The Lux smoothing method is an adaptive smoothing method exclusively developed for Lux Algo indicators. This method aims to provide a good amount of smoothness without introducing excessive lag.