Hey gang, I’m trying to find a indictor (standard or custom built) that will help filter out price action to stop my ea taking trades in rough conditions to maximise gains. Can anyone point me in the right direction? Please only mention indicators that you have had first hand knowledge in and seen work for yourself.
Many thanks
darth
I once did this with envelopes and bollinger bands - if the bands are wider than envelopes, market is too volatile.
Its not perfect and has a little lag but maybe worth trying 
ohh interesting what did you set the period too for each?
darth
I don’t remeber exact values, I tried many… I used the same period for both indicators so the only variable is the deviation on bb.
Just for example: bb and env period 21, bb deviation 5, envelopes deviation 0.3
The same filter could be done with just standard deviation indicator and some fixed value it should exceed, but I like the visual appearance of bb and env - more understandable.
yeah i have heard of standard deviation, could you give me an example of it in use, i tried looking but couldn’t find much on it…
darth
Standard Deviation (Volatility) [ChartSchool]
Heres everything about SD. However I meant that bollinger bands and envelopes could be replaced with SD, because bollinger bands is based on standard deviation: take a 20-period sma and add 20-period SD multiplied by 2 - you get the upper band of 20-period BB with 2-deviation.
ahh cool, i’ll take a look
Gday bro. Best of luck with your search. From my experince there are no decant filters. All any of them do is filter out good trades as well as the bad. It just something one has to accept. With that said there is one that has worked for me and thats time. Whether thats no trading in the last half hour of each hour, no trading between New York close and Asian open or no trading on the 3rd of the month. Time seems to be the only valid filter.
Personally IMO don’t focus to hard on the entry signal. Concentrate more on how you manage the trade once on. Hope that helps a bit bro
Bob
I agree with bobbillbrowne, the most important part of the trade is between opening and closing it. Trying to set fixed goals right after opening a trade won’t work, because market changes constantly - you have to change you goals constantly. I use trailing stop that gets its values from other indicators to stay on “real-time” market.
Filtering is always important though. Personally I filter the peaceful periods OUT and focus on the most volatile changes - thats when the big money moves, so market is harder to manipulate and psychology matters the most. After all, technical analysis always relies on psychology.
good stuff guys food for thought. I’ve seen you’ve had some success with ea’s bob, woild you say its down to complex coding or just good research into the system?
darth
Oh and some epic failures too my friend! Definitely good research into the system. I use to believe that the best use of a bot was to do the historical testing needed to develop a good system. But now I have the understanding that if your system is fully mechanical AND completely defined in your trading plan then there is absolutely no reason it would fail as a bot. My personal belief is PA based bots have greater success than indicator based.
Now I come from the school of self taught when it comes to programming. Used a compiler at first and still do (plug - fxdreema.com). But nothing beats DIY when it comes to forex so have been learning this year. So I can’t comment to much that side but defining a single point in time can be quite a complex art. Or it could just be me. Practices makes perfect I guess
And the single biggest mistake people make is they simply don’t know how to press the autotrade button and turn the bloody thing off.
All the best bro
Bob
i didn’t even know you could get a ea to analysis PA?!?!?!
i’m a pure PA trader and that would be exaclty what i need to make my bot work successfully? How does that even work a PA bot? sounds complex.
darth
Hell yeah bro. There are only two points of reference on a chart. The x-axis (time) and the y-axis (price). Everything is a calculated value against that. You just need to be spot on with your logic. Also think in terms of percentages when defining candles. I can’t comment too much as I’m just learning programming but there’s not much you can’t code if you have your logic defined.
How do you identify whether the big move is due to an increase in market activity or a severe lack of liquidity? It’s hard to determine the actual cause of moves in spot FX since volume is just tick volume.
You might be interested in candlestick clustering if you’re into PA based strategies.
Intelligent Trading: Quantitative Candlestick Pattern Recognition (HMM, Baum Welch, and all that)
Thanks bro, more homework :). Good to know the old school membership base is still out there keeping an eye on us greenhorns and throwing out these little snippets of gold. Most appreciated.
Bob
Been away for the last year or so, but I visit from time to time.