True Strength Index: double-smoothed momentum with a usable scale

What it measures

The TSI takes the bar-to-bar price change, smooths it twice with exponential averages, and divides by the same double smoothing of the absolute change.

The division is what bounds it between plus and minus one hundred and makes readings comparable across instruments.

Double smoothing removes far more noise than a single pass, which is why the TSI produces a genuinely smooth line without the enormous lag a single long average would need.

How it is calculated

These are the steps Basis performs, verified against the published definition.

1. 1

   Compute the change from the previous close.

2. 2

   Smooth that change with a long exponential average, then smooth the result with a short one.

3. 3

   Do exactly the same to the absolute value of the change.

4. 4

   Divide the first by the second and multiply by one hundred.

Settings

Twenty-five for the long smoothing and thirteen for the short is the Blau original. Shortening either makes it noisier without making it meaningfully faster, because the double smoothing dominates.

How to read it

• The zero line separates positive from negative momentum and crossings are the main regime signal.
• A signal-line crossover gives earlier entries at the cost of more of them.
• Readings beyond plus or minus twenty-five are stretched, though there is no universal overbought level.
• Divergence is more trustworthy here than on most oscillators because the double smoothing removes the noise that produces spurious ones.
• Compared against MACD, the TSI is bounded and MACD is not, which means TSI readings are comparable between instruments and across time while MACD readings are only comparable to their own recent history. If you are scanning many symbols for a momentum condition, that difference decides which one to use.
• The bounded scale also makes it usable in a screener condition, where a threshold such as above zero means the same thing on every symbol you scan rather than needing per-instrument calibration.

Where it misleads

Related indicators