Stochastic on M5 only on range days, and how I decide it is a range day

[Fairman]

Stochastic on M5 only on range days, and how I decide it is a range day

the stochastic gets a lot of criticism because it's constantly overbought in a trend. That's true, and it's why I only use it when the market is clearly ranging. The trick is deciding when that is.

My filter has three parts. First, the daily range so far is smaller than usual. Second, price is spending its time between two clear levels, where the highs and lows are touching the same areas. Third, the 20 EMA on M5 is flat. If all three agree, I use the stochastic to time entries near the edges of the range.

At the top of the range, I wait for the stochastic to be in the overbought area and to cross down, together with a rejection candle. At the bottom, the reverse. Stops go just outside the range and the target is the middle or the other side.

When the market breaks the range, I stop using it right away. The signals that were useful a moment ago become dangerous. I treat a close outside the range as the end of the setup for the day.

[Fairman]

For the first part of the filter, I compare the day's range so far with the average daily range from the last couple of weeks. If by mid morning less than about half of the usual range has been used and price is sitting between two clear levels, it is probably a range day. That is a rough check and not perfect, but combined with a flat 20 EMA it filters out most of the trend days where the stochastic would get me into trouble.

[Fairman]

I use the default stochastic settings and the standard overbought and oversold levels. I tried faster settings to get earlier signals, but on M5 that just added noise and more false crosses. The important part for me is the location. A cross down at the top of a range means something. The same cross in the middle of nowhere means very little, so I ignore any signal that does not happen near one of the two edges. Where the signal happens matters far more than the signal itself.