Momentum was the first thing I learnt in technical analysis. But after
practicing the vocation for more than 10 years, I use few momentum
indicators. For me all of them whether it's RSI, MACD or anything else, they
all tell me a similar story. Though I am a believer of confluence, my
approach is not to look at various momentum indicators and look at the
combined weight of evidence, but to harness various degrees of time. I was
always more interested in knowing what is the result of the weekly, daily
and 60 min RSI at the same time. What do they tell me together and not
separately? This is where momentum miserably fails.

There are other reasons. Momentum is noisy. It has so many signals. If
weekly RSI is a sell, daily RSI can be a buy. There can be a conflict.
Third, there is no way of saying prior when an indicator has reached an
extreme reading, how higher or lower can it go? One may say Constance
Brown’s over-reactivity on momentum works. I would say it’s brilliant and I
use it, but over-reactivity also can’t harness multiple degrees of time.
There can be a negative reversal on 60 minutes and positive reversal on
daily.

What’s my process? First I filter stocks through numeric rankings. Second I
look for performance cycles (Jiseki) signals. Third I look at the price
structure through Elliott and only after that (fourth) draw momentum. Just
to make my case I have illustrated TCS here. Daily, 60 min RSI are oversold
and weekly RSI is at 40. A conventional approach would be to stick around
for a price confirmation break of the potential Head and Shoulder neckline.
Now if I look at momentum the signals are unclear, but if I look at Jiseki
performance cycles, the sell signal is clear. Jiseki says TCS is ready to
crack.

Mukul Pal, CMT, Orpheus Capitals, Global Alternative Research