Don’t Count the Bears Out

By -

On Tuesday I talked about how you can improve virtually any system by filtering on ATR basis.

This is so obvious in retrospect, I can’t believe I didn’t notice this earlier in my trading career. High volatility and low volatility periods have completely different psychology, the same way trending and rangebound markets are driven by opposite psychology.

Basically, during decreasing volatility trending moves, counter trend setups are a negative edge.

But counter trend setups become viable again once ATR (as a proxy for volatility) reaches a historic low and then rises.

This is because volatility, once it starts falling, has a statistical tendency to continue falling until it reaches an extreme.

In plain English, once markets start melting up in a boring fashion, they tend to stay boring and melting until they get so boring that we stop paying attention to them… then they go crazy when nobody’s looking anymore.

This phenomenon is discussed extensively in Benoit Mandelbrot’s excellent book. I’m digressing a little, but the bottom line is that volatility is FAR more predictable than price and extremes of both high and low volatility are unsustainable. You wait for an extreme to reliably target market regime change.

What’s a historic low? For system building purposes lowest ATR reading in 90 bars is about right.

You can see on a monthly chart of ES back to the 90’s we clearly have regime change in 2018.

We are in a volatility regime which both short and long setups are viable.

Obviously the bulls are still kicking the bear’s asses, but the bears have had their moments.

What’s my point? We are about a week from the close of the monthly candle and we have a great potential setup forming on the monthly chart.

I greatly favour setups which don’t just jump in, but wait for price to confirm a view by breaking the candle low (for a short, vice versa for a long). Don’t worry I’ll explain in a bit.

Note that we need to wait until the monthly candle closes (entering on half formed candles is half-assed) but it’s nice to be prepared.

The rules of the RTV Sell setup

  1. Highest high in 10 bars
  2. One or more lower closes without breaking the high in (1)
  3. One or more higher closes without breaking the high in (1)
  4. Enter short on break of the low of bar 3, stop a tick above the high in (1)

I’ve put together a handy document for ya!

Here are the exact rules for the mean reversion setups I use.

And here are some examples, from the 2 minute scalping system which is my main focus these days.

What’s got me interested here is that the market has been rallying strongly until about a week ago. Any pullback in an uptrend in a bull market SHOULD be bought heavily. But there isn’t much buying. In fact, what we are seeing looks like distribution.

One more thing.

Look at past endings of bull markets. It is typical for the first retest of a long lasting high to be very deep.

And this point in time doesn’t “feel like” a bear slam dunk. (those never work)

It “feels” like an ass puckering setup to take.

Those are the best ones.

If this type of trading is interesting to you I have a mailing list where I talk about system design and optimization.

It’s nice to be back at the Slope. Best wishes to all of ya!