Making Bollinger Band fading profitable

[dom993]

If you have enough historical data (say 10 years), backtest the basic mean reversion system & record for each entry as much contextual information as you can think of.

Then do statistical analysis, and see if you can come-up with well identified contexts where there is a clear edge in either fading the move towards the band, or going with it.

 

[ofthomas]

am lost, first... it was reading the tape, now using bb... how come you dont stick to one thing and follow through with it until you figure it out 100%? curious.. you seem to be jumping around a lot...

 

[1a2b3cppp]

Quote from ofthomas:

am lost, first... it was reading the tape, now using bb... how come you dont stick to one thing and follow through with it until you figure it out 100%? curious.. you seem to be jumping around a lot...

I have a way that I trade and have been posting live trades for years in my journal threads.

I'm looking for other ideas that might help my understanding of the markets.

 

[atticus]

Only take BB signals in a declining ADX. Not a reco, an example.

 

[bone]

I find that technical studies are usually not applied correctly. John Bollinger himself will tell you that it is the price action near the edges of the bands that tells you to buy or sell. Just fading every band edge is naturally going to get painful. And JB never intended the indicator to be used that way either quite frankly. Ask him.

Think about it in a similar fashion to Market Profile. When historical price "edges" are maintained or even fail off - don't fade it, go with it. The market is telling you that a new fair value price range or dominant player control area is being or has been established.

Conversely, if the "price edge" is met ( the band ) and the market fails off away from it quite dramatically, then go ahead and fade it. Give up a little trading range in exchange for better probability and signal confirmation.

My 2 cents.

 

[IronFist]

I am interested in this, too, since I've tried similar things in the past but without hard stops.

One time this happened:

 

[1a2b3cppp]

I am manually backtesting the last week using a 4 tick stop and not averaging down at all.

Target is the closest tick on the opposite side of the MA.

No reentering until price has crossed the MA.

Here are the first two days. This takes a while so I'll put the other ones up later.

April 29, 2013:

Morning:
http://www.realfibonaccitrading.com/wp-content/uploads/2013/05/April-29-2013-BB-fade-one-entry-1.png

Afternoon:
http://www.realfibonaccitrading.com/wp-content/uploads/2013/05/April-29-2013-BB-fade-one-entry-2.png

April 30, 2013:

Morning:
http://www.realfibonaccitrading.com/wp-content/uploads/2013/05/April-30-2013-BB-fade-one-entry-1.png

Afternoon:
http://www.realfibonaccitrading.com/wp-content/uploads/2013/05/April-30-2013-BB-fade-one-entry-2.png

 

[1a2b3cppp]

May 1, 2013:
First half
Second half

 

[Handle123]

Quote from 1a2b3cppp:

May 1, 2013:
First half
Second half

I am guessing you have not discovered the pattern in which BB usually work? More often that not, when the both the upper band is going down & the lower band is go up(tightening) on a bar by bar basis, using the BB for entry will work. But when one band is going the other direction, it will either stop one out for full loss or will allow one to get out at breakeven it you have a time rule in place, but that is tough if you are working with volume bars.

You can also take a larger timeframe, say five minute bars, and if it retraces to 21ema, and one minute timeframe price retraces to bottom of BB, that makes for a low risk trade shooting for 4-8 ticks depending on time of the day.

 

[1a2b3cppp]

Quote from Handle123:

I am guessing you have not discovered the pattern in which BB usually work? More often that not, when the both the upper band is going down & the lower band is go up(tightening) on a bar by bar basis, using the BB for entry will work. But when one band is going the other direction, it will either stop one out for full loss or will allow one to get out at breakeven it you have a time rule in place, but that is tough if you are working with volume bars.

So you are saying not just when the bands are getting tighter, but when the top one is actually going down and the bottom one is actually going up (since technically they can both be going up or down and the distance between them decreasing), right?