BB Squeeze (by Andy)

9 janv. 2018

This is only version 1.0, so largely the script isn't working, however you'll get a feel for the logic if you play around with it.

Comments and suggestions welcomed.

12 janv. 2018

Notes de version

This script uses Bollinger Band width as a key indicator of price movement.

By working out the width of the bands (squeezing or expanding) we get a good indication of impending price movement, or a chnge in short-term trend.

If we then look at the rate of change in this width, sharper variations (such as a steep rise or fall) generally mean a breakout.

By plotting this rate of change (red line) along with an historical band width value (yellow line), we can see when they crossover, a trend becomes reality.

Although this script should honour a 120 bar moving average (to determine the general trend - ensuring we're not going long when the price is declining) this isn't working in this version. I've plotted that MA on the chart so we can choose not to go long if the trend is going in the wrong direction.

Obviously the next step is to apply the same logic to a declining trend, when the crossover happens in the opposite direction.

Comments and suggestions are welcomed!

12 janv. 2018

Notes de version

Added moving averages to determine long term trend

12 janv. 2018

Notes de version

Improved stoploss / take profit calculations. Still some work to be done here.

14 janv. 2018

Notes de version

Updated the logic comprehensively this time. Previously we were using MA's to define whether we could go long or short, however getting 2 averages to agree (especially when shorting) was far too restrictive, so now we'll only honour 1 average for short or long positions. Still having some trouble with currency conversions. it seems a 'tick' is different from one exchange to another, and one currency to another, so there doesn't appear to be any way to consistently convert a tick into a percentage of equity or position size. Hence, you will need to pay around with the stop-loss, profit and triling loss settings to get this working correctly.

15 janv. 2018

Notes de version

Updated timings and added close position triggers on bollinger maximum width. I've been playing around with the numbers a bit, and it seems we get the best results as follows:

1min chart: Bollinger days set between 10-16
15min chart: Bollinger days set between 5-10
1hr chart: Bollinger days set between 20-30

The reality is we really want our bollinger days to be as high as possible, because it results in fewer trades following larger trends. I'll do some more testing and see why the numbers give the best results and report back with the next update.