Log in

Trading routine example with Brain Trend

By JohnLast 3174 days ago Comments (8)
От 21 октомври 2011


Here I add anither shot showing the trading routine example. This shows the combination between an estimation of the market state and the decision to follow or not to follow a signal.

The main idea is as you already know that at we need a fractal dimension structure that would be appropriate for directionnal algorythm. 

The Brain Trend is a directionnal algorythm it will work if the probability of the next bat to be in the same direction is bigger than the probability of reverse movement. 

That is measured by the iVAR (or FGDI) indicator. Look at the shot when the farctal dimension is high the movement is jagged. When we have a break - out we can expect a good directionnal movement. 

And here comes the brain trend. On the shot I use Brain Trend with lenght of 50 on 5 m time frame. The optimizer has told me that for this month the Brain Trend was profitable whatever the fractal conditions can be. And that is a safety measure. And we follow the directionnal indications. No brainer at all.

Of course there will be head fakes, but that is trading that is the way it is. Just do not assume. Follow the directionnal algorythm. You would need to stop using it if the market conditions remain sloppy and the fractal dimension goes higher.

You know that you can use anything directionnal instead of brain trend. It has to work. It is possible to use price action too. But I use an algorythm because it is a personnal choice, I like to have the statistical edge (it is possible to have statistic estimation of the disretionnary price action too but it is a lot of manual work for data gathering). 

It is also possible to let the expert trade when you detect the conditions but supervision is alway necessary.

Here I would show you the results with the Brain Trend expert and of Brain Trend with iVAR filter. It is interesting because the normal Brain Trend without filter makes more profit. The iVAR filter smoothes the equity curve the profit ratio is bigger but the profit is smaller. The reason is that some of the best impulses begin when the fractal dimension is still high. And those impulses make the profit that compensate the losses.

Here you can compare the two charts. The firts chart is the normal brain trend expert. The second chart if the Brain Trend with the iVAR filter. (Brain Trend use only the end of bar information that is why I can use open prices only for testing)


От 21 октомври 2011


От 21 октомври 2011


  • JohnLast 3174 days ago

    You need to know the profitability of Brain Trend on the current market conditions. For example it works this month but it did not work in September. If the equity curve is going downwards DO NOT insist. It is either working on almost all parameters or not working at all. It depends entirely on the market structure the good point is that the market structure remains relativeley stable for a while. 

  • londontrader 3174 days ago

    hey john try this on some boring pairs,exotics

    since this is trend trading

  • JohnLast 3174 days ago

    Every idea is welcome. Thank you for reminding me, I just do not know why I am so much obsessed with the EUR/USD.

  • JohnLast 3160 days ago
    Brain trend trading routine


    This is another example of trading routine of Brain trend.

    We combine here 3 things:

    1. Fractal dimension analysis: as we know there are two zones and every zone has its specific behaviour.

    2. Open orders information that allows us to know where the support and resistance is according to the current information of the market orders in the market according to the Oanda open orders book.

    3. Brain Trend as signal. The system is trying to smooth the noise allowing us to do two things:

    - to stay in the trade as much as possible

    - to gilter out the noisy signals

    When it is combined with the other information the combination becomes a powerfull framework, well at least the best I have. 

    However the back testing of the Brain Trend is giving me some statistical assurance what time frame I would use for basis. Actually the 15 time frame is good but you need contant back testing to be sure.

    I mean that the 15 m time frame with my current smoothing is profitable if I take all the signals in all market conditions (the bad ones with high fractal dimension included).

    What I coud try is to use my discretionary analysis to do better if possible. 

  • JohnLast 3160 days ago

    The settings of the brain trend was not profitable during the range they were profitable before the range. And as we all expected a big volataility it looked logical for me to train for settings for the period just before it as it was including the volatility settings that we were expecting in the future.

    It may sound a little bit strange but this logic is different from the common optimization logic.


  • JohnLast 3160 days ago

    I think to have a look at the optimizator is very important. For example for 5 m time frame the Brain trend was loosing at all possible settings between 20 and 30 october. This thing is either working or not working. It is giving or not giving a solution. And it was profitable on all settings on the 15 m. time period. So for the break - out logically I would use the 15 m time period for confirmation.

    As you can expect during the range it was not working at all.


  • Pv 3079 days ago

    Is it possible to change BraintrendEA...so it will only take trades when brain trend signal appears outside the channel?...sorry don't know how to attach the picture or indicator that I am using...

  • JohnLast 3078 days ago

    You can give a link from google picassa. Yes but the BT EA its idea is to take a signal when there is a break - out from its own channel (you do not see it), so basically you add two things that are the same. 

    In the expert there is a possiblity to take two brain trends and make it multitime frame, for example to take a trade only in the direction of the daily etc. I did not use that functionality yet but its within the code.

    Instead you would test the profitability of your channel alone and you would be perhaps surprised.