indicator for trending and ranging

hello, new to the forum, im making an ea and i have pretty good background in programming, for the last few months i have been programming mql

I have tried many methods but i cant seem to get this down pat

what is the best way "specifically" of confirming a trending or ranging market. I have tried adx but it doesnt seem to work very well.

any help would be greatly appreciated.

im trying to create an ea that has different rules based on trending and ranging so having this at the core is very important.

i am using smaller timeframes m5 and m1 to scalp from.

thanks again in advance

There is NO, yeah get this thing into your mind, NO indicator that's perfect. What you can do though is to use money management in combination with your favorite indicator i.e. cut your loses with your MM, if you can do so, your profits will run...

Hey buddy, I often used to wonder the same thing. Because obviously in trending you can make pips simpler, and in ranging just trade in the range.. simple, but only if you could identify what is happening.

But you can't Nothing can predict the future... at least you have the right mindset though and were thinking about that. I've seen EAs which say they determine if the market is trending or ranging then do whatever, but none of them have been decent...just because market has been ranging for an hour, you think it's gonna range all day?

Then from there I hope you find some clarification, as I did and the market made more sense to me.

Sim

This is the dilema i have came to, hence why indicators don't work, therefore ea's dont work.

So now the limitation is thinking outside the box but the principle still exists, its either trending or ranging. I know there is no perfect way of identifying this, and it wont always work, but i do need some basis to go off of, just wondering if you have found anything that cuts down the possibility of being wrong with the trending or ranging.

again, i know its not going to 100% of the time be accurate, but i would be happy with 80%. of course money management is a big part of this as with manual trading, but we have to overcome some type of identification to move onto something that works.

again i realize many have tried and many have failed in this endevour.

so now i have tried a different approach that maybe this will help.

its basically identifying the distance from the 1ma compared to the 200ma.

if it is so far away in either direction it starts a loop.

1 thing i have noticed is that the market likes to draw towards the longer moving average so im using 200ma as a staple on the m5 chart.

so let me explain this further

int distance1 = (ma1 - ma200)*1000;
int i = 5;
string Signal;
while(distance1 > i){
i++;
if(distance1 < i){
Signal = "SELL";
break;
}
}
i = -5;
while(distance1 < i){
i--;
if(distance1 > i){
Signal = "BUY";
break;
}
}

so if you look at this coding, im trying to keep counting the distance from the 200ma. if it goes up it adds or subtracts to the variable i.

if it starts to go back, then it sends the signal to buy or sell
my problem is its not sending the Signal variable back to run this....
if(Signal == "BUY"){ManageBuy();}
if(Signal == "SELL"){ManageSell();}

so how can i break the while loop to return "Signal" back to the main program?

Hmm, you have similar thoughts to what I went through. can you check your profile visitor messages please?

not sure what that is, just joined the forum today, i dont have any messages

actually it says as a junior member its disabled

I seem to recall somebody with much more experience than I making a statment that markets trend 80% of the time. It wasn't made by somebody on one of these forums. Since hearing that statement, I haven't seen anything that would cause me to refute it (except for Sep 08 to Apr 09).

you cant just program a ea for trend based because then you will get false readings if it starts ranging, also ranging is a possible reversal sign

on a smaller timeframe you would notice that it ranges more then on a longer timeframe.

just my 2 cents....

btw im referencing off m5 timeframe

Simple solution to this problem is bollinger bands. You could code a minimum required distance (for targeting trends, a maximum for targeting ranging PA) between the upper and lower bands for a trade to be considered. To determine the distance, I would look historically at the pair(s) of interest (on your intended timeframe, of course) in the context of your trading strategy...that way you know the statistics. I have had success with this approach, as well as when pairing it with ADX. In my experience ADX by itself does not cut it....using BBs allows you to pull the actual PA into consideration by the EA.

'elo!

I am using sth like
MAPrevious
MACurrent
PipsForRange

if ((MAPrevious + PipsForRange * Point) > MACurrent) SignalSell

and vice versa for buy.

MA price should be Open, otherwise you get whipsaw. So, if MA is near horizontal, do not trade - so you have to figure it out how much PipsForRange should be for a pair you trade. If MA is showing slope, GO, GO, GO!
I also found it is quite usefull to use at least 2 time frames - range is defined on ex. D1 (filter), white trades are taken on ex 4H (another filter). Signal is
if (BarPrevOpen > BarPrevClose) SignalSell.

And I also use PipsFromOpen -if price goes above or below the OpenCurrentDay + PipsFromOpen - SignalTrade, otherwise it is considered as Ranging Market.

Just not shure if Open[0] is for current bar of current chart TF or Open Daily price?

Gotta go,

Have fun,

Simon