How do I avoid those rangy thing... | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Avoid those brokers that require W-8 signing 18 replies

For those of you who trade pivot points 30 replies

Start Trading: Throw Those Excuses Out the Window 0 replies

For those holding positions over the weekend, check how your currency is doing! 6 replies

Attachments: How do I avoid those rangy thing...
Exit Attachments

How do I avoid those rangy thing...

  • #1
  • First Post: Aug 16, 2008 10:39pm Aug 16, 2008 10:39pm
Hi,
Have some experience, don't feel like a guru yet, but what I'm sure of is:
Rangy market takes more of my money and the best for my wallet would be to avoid them.
Trying to catch the wave gives me so many loses, that spectacular rides hardly cover them.
Here's what I'll backtest now.
1. Wait for trend.
2. Make sure it's trend.
3. Wait for little retracement.
4. Jump in on breakout.
Before breakout it might happen that retracement becomes another rangy period what I want to avoid.

Minus is not too many trades
Plus not too many loosing trades

Minus it's vewy vewy scurvy on lower TF like 15M [I have some trick for this one too ;-) ]

So let's lern by practise. My idea is simple:
ad1. Waiting
ad2. When MACDcrosses below 0 crosses up for Long and v. versa for SHORT
ad3. secret_eyball.ex4 is perfect for that [can't find it on FF]
ad.4 as said and shown on pictures

Good examples, after MACD cross and retracement I've marked Entry level

http://i33.tinypic.com/fp8pjn.jpg http://i38.tinypic.com/b3my3d.jpg


Bad examples, I've tried to explain why there's no entry [no proper breakout after retracement, or cross on MACD didn't confirmed by price action [PA]

http://i34.tinypic.com/1snkua.jpg http://i35.tinypic.com/2mw8lxs.jpg


Of course there is some time for tears...
Retracement did not occure, so this method didn't gave a signal
http://i38.tinypic.com/id50z5.jpg


For some final comments:
This is not a method as it is. It's more for general idea of what am I search for. Have some more complex system for this in my arsenal, but think of share it later in case some nice discussion appeare and some smart update for what I wrote now.
What I mean is i don't wanna mess with general idea now by show some complex complicated method. For now think it's better to show naked idea with some simple indic for ilustration. If you have some better indics than MACD for what I show, any comments, anything I'll really apricieate [My ego will as well ;-) ]

PS.
#CLU8 - stands for crude oil future
  • #2
  • Aug 16, 2008 11:11pm Aug 16, 2008 11:11pm
whatever floats your boat dude

can you see 10 hours into the future? how about 10 minutes?

there is no macd cross that will overcome the randomness of short term chaos. it's better that you don't set your hopes too high...
  • #3
  • Aug 17, 2008 4:13pm Aug 17, 2008 4:13pm
(Just an opinion of another with L plates)

For starters, explaine to me what that mac'd on the chart is telling you. Personaly, after alot of looking in to it I can say that I certainly havent found any benefit to that particular indicator. If you can tell me exactly how it is usefull then maby ill give it a try again, lol.

Seconds... Nobody can forcast if 'this next move will be the start of a sizeable move'. A good choice is to find a low risk decision-point ( retrace to a strong resistance/support), in the direction of the trend, and enter with a tight stop-loss when the peaks and valleys on a lower chart tell you the retrace is over, and hope for the best.
I think the only way to win at this is to keep the losses small and let the profits run, which Im sure everyone in the factory has herd before.

Good luck, trading is tough.

Leon
Living the adventure in my head.
  • #4
  • Aug 17, 2008 5:43pm Aug 17, 2008 5:43pm
Quoting Kaligula
1. Wait for trend.
2. Make sure it's trend.
Well, You don't know for sure it's a trend until the actual trend is over and you missed the move, unless you can predict the future. IMO there's no point in developing systems intended to work on/detect specific market conditions because You never know what the market conditions will be in the future. My experience so far is that no random combination of indicators will put the odds in your favor, instead your system should be based on a logical and specific reason, something that gives You and edge. Personally, I don't consider MACD or any other indicator as a logical or specific reason why a system should work.
"I say high, You say low, You say why, and I say I don't know..."
  • #5
  • Aug 17, 2008 6:50pm Aug 17, 2008 6:50pm
Quoting tdion
whatever floats your boat dude

can you see 10 hours into the future? how about 10 minutes?

there is no macd cross that will overcome the randomness of short term chaos. it's better that you don't set your hopes too high...
tdion - the god of randomness... I always wonder what do you see on charts, and (how) do you make money on it...

1H randomness simply appears 23 hours faster on your screen than Daily randomness.
In other hand if you try to spot f.e. dragonfly on daily why don't you wait for 24 candles long U-shaped pattern on 1H chart? - you won't miss dragonfly that apeared around midday of your chart and will enter faster than others that will see - something I don't remember the name for it - bearish cloud [?] 12 hours later.

You cant measure temperature from watching one single particle, and you can't say a word form watching one single candle. No matter what TF you use.
  • #6
  • Aug 17, 2008 7:09pm Aug 17, 2008 7:09pm
Quoting Leonlorenzo
For starters, explaine to me what that mac'd on the chart is telling you.
It tells me nothing. It just force price on my chart to go up or down
In fact as I mentioned I don't use MACD, however It's nice indic to say that some moving averages are getting closer to each other.

I decided to use MACD here in my example trades - with no reason. It just make chart looking professional and it's nice when you observe something by your eye - to get confirmation from some lousy indic like MACD.

Please be aware of my goal - to get some comments, some idea how to avoid false breakouts, and how to skip into trend faster. MACD was attached not to discuss it. I hoped to make my idea clearer with it. Personally I suppose most of indics are quite the same. They looks awesome on great moves and then loose it all on rangy markets. So I'm trying to discuss how to avoid that rangy periods and comments from you on this topic are welcome the most

Quoting Leonlorenzo
Personaly, after alot of looking in to it I can say that I certainly havent found any benefit to that particular indicator.
From my long experience i found that only benefits in forex market comes from price moving up and down. It's of course very risky etc etc... lol.

Quoting Leonlorenzo
Seconds... Nobody can forcast if 'this next move will be the start of a sizeable move'. A good choice is to find a low risk decision-point ( retrace to a strong resistance/support), in the direction of the trend, and enter with a tight stop-loss when the peaks and valleys on a lower chart tell you the retrace is over, and hope for the best.
Looks nice - gives propabily less signals then my idea, but worth of discussion. Last point of your strategy I find very usefull in most of my trading methods as well

Quoting Leonlorenzo
I think the only way to win at this is to keep the losses small and let the profits run
I'm not sure can I agree... I'll think about it

P.S.
Thanks for comment
  • #7
  • Aug 17, 2008 7:33pm Aug 17, 2008 7:33pm
What you describe is exactly what I do myself.... I take a gamble on what the daily candle will do before it completes.... based on H1 price action

BUT

I have as many failures as I do successes..... My point was that I have no filter.... which is what this thread is about.

Quoting Kaligula
tdion - the god of randomness... I always wonder what do you see on charts, and (how) do you make money on it...

1H randomness simply appears 23 hours faster on your screen than Daily randomness.
In other hand if you try to spot f.e. dragonfly on daily why don't you wait for 24 candles long U-shaped pattern on 1H chart? - you won't miss dragonfly that apeared around midday of your chart and will enter faster than others that will see - something I don't remember the name for it - bearish cloud [?] 12 hours later.

You cant measure temperature from watching one single particle, and you can't say a word form watching one single candle. No matter what TF you use.
  • #8
  • Aug 17, 2008 7:38pm Aug 17, 2008 7:38pm
Quoting Ricardo.
Well, You don't know for sure it's a trend until the actual trend is over and you missed the move
That's quite frustrating, and even more frustrating is entering on top of trend, and exiting, just days/hours/minutes before it appeare to be just retracement. My idea is more to be able to skip rangy times than enter all trends.

Quoting Ricardo.
IMO there's no point in developing systems intended to work on/detect specific market conditions because You never know what the market conditions will be in the future.
Not sure what you mean. If it's about fundamental condiotion it's most of the time that firstly one country is weaker than another or they are quite equal and secondly one currency is undervaluated against another, or not. For tech is the same - there is range or trend, eventually some currency seems to be more active than another during some given session, what will change during another session.

Quoting Ricardo.
My experience so far is that no random combination of indicators will put the odds in your favor
Yes I agree with you 100%. Have same observation.
For me minning of indics is that the most I'm looking for f.e. trend, the more I see it [start to see it everywhere and both directions ]
So my idea is to describe with my own words what do I consider as a trend, and judge all sitiuations accordingly what I write for myself.
Eventually it may happen that some indic is helpfull in identification of what you described, or you can write such indic for yourself and use it
but not blindly follow it
Finally I'm very sceptic about all indics based on thinking - if price closes above previous close is not good indic for uptrend, maybe Average of last 5 of them closing above average of last 8 of them?
No?
So maybe distance between two different EMAs?
No?
So maybe averag of five last differences between them - and so on.
For me proper indicator is created for spotting what your will is to spot, but you are alredy look for your specific pattern for so long, that your mind see it everywhere. [That's an Idea i'd love to hear your comments about - what's your reason for using indic. Did you look for indic that will show some pattern and found it and now use it, or rather you have attached indics randomly, and one appeared to be working one?]

Thanks for thoughts
Chris
  • #9
  • Aug 17, 2008 7:44pm Aug 17, 2008 7:44pm
Quoting tdion
What you describe is exactly what I do myself.... I take a gamble on what the daily candle will do before it completes.... based on H1 price action

BUT

I have as many failures as I do successes..... My point was that I have no filter.... which is what this thread is about.
Yes - this is about looking for filter, I agree that following MACD might give as nice results as cointoss. But I wonder what is your opinion about breakouts in trend direction. Might it be any better than cointoss?
  • #10
  • Aug 17, 2008 7:47pm Aug 17, 2008 7:47pm
I'm sorry I can't be more specific.... I have given the general idea for my own strategy, which sounds similar to what you are exploring.

I think my one piece of advice is not to try to predict the daily candle every single day, but only on days where particular daily candles are forming and look highly probable.
  • #11
  • Aug 17, 2008 7:51pm Aug 17, 2008 7:51pm
Quoting tdion
I think my one piece of advice is not to try to predict the daily candle every single day, but only on days where particular daily candles are forming and look highly probable.
This is real gold. I wish I said that...
  • #12
  • Aug 17, 2008 9:32pm Aug 17, 2008 9:32pm
Ok, here are some charts to illistrate what I was saying before.

Here is the 1hour chart that I would use to find support and resistance.

And then a zoomed out 1min (equivalent to a normal size 5min probably) that I would use to set my stops and enter.

The point is that you may get afew losses of 15ish pips before you get a win, but the target will be the next support up(or close to it -- depends on the valleys and peaks up to it), so when you do win, youll make back the losses and more.

Now, by no means is this a perfect science, BUT, using the breaks of valleys and peaks when caught under/over the support/resistance for an entry seems a sensible move. Placing the stops on the high/low of the previous peak/valley, you are risking very little, have pretty defined trade criteria, targets, and are able to keep emotion to a minimum.

Id be conservative and say that the hit rate is about 25- 30%, but with next majour support(or close) as a target you should still make some profits.

Ive squared the entry areas on the 1min chart, In the 1st square there are 5 possible entries, one would have been a loser.

Good Trading, Leon.

Ps. Even MA's have no value eccept to offer a sence of faulse security. In my opinion ofcourse.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 1hr.gif Size: 14 KB
Click to Enlarge Name: 1min.gif Size: 24 KB
Living the adventure in my head.
  • #13
  • Aug 18, 2008 3:56am Aug 18, 2008 3:56am
Quoting Kaligula
For me proper indicator is created for spotting what your will is to spot, but you are alredy look for your specific pattern for so long, that your mind see it everywhere. [That's an Idea i'd love to hear your comments about - what's your reason for using indic. Did you look for indic that will show some pattern and found it and now use it, or rather you have attached indics randomly, and one appeared to be working one?]
This post by Merlin will illustrate better my opinion on developing and testing a system:

Quoting merlin

2) How likely is it that a system that backtests well will perform poorly?

that depends on the process by which the backtest was done. if you just banged on your keyboard until you came up with a good backtest results, the system will probably fail in the future. if, instead, you wrote down your idea first, and the idea is rooted in some market function or idiosyncrasy that you have observed, then you went and backtest the idea and it looked good without then it has a MUCH better chance of working in the future. it is very important that you know WHY the system works. there should be a reason, and that reason should be understood BEFORE you go to the backtesting. you can fall into a trap of coming up with reasons why the system works AFTER youve built it, but its likely you will just be making up reasons to justify the system.
"I say high, You say low, You say why, and I say I don't know..."
  • #14
  • Aug 18, 2008 9:10am Aug 18, 2008 9:10am
Quoting Leonlorenzo
Id be conservative and say that the hit rate is about 25- 30%, but with next majour support(or close) as a target you should still make some profits.
I like your idea a lot. I used to test different 1H up systems and see what happen if I'll enter on 15M or smaller, when signal on higher TF occured. That's similar to what you do with exeption that you use plain SR. Like it a lot and will test someday
Here's myy advice and if you won't do it, I will.
Try to spot how 1M behave on S/R levels. What does it do when gonna go back, and what does it do when gona break the level. [It should be quicker when break the trend].
Then when you learn it's action on breaks [you poropabily learnt a bit so far] you can bet only one direction, and not on breakouts of your valleys, but on breaks in oposit direction. I marked on your chart what would be vewy vewy nice place for entry.

http://i38.tinypic.com/be6g3t.jpg
  • #15
  • Aug 18, 2008 9:11am Aug 18, 2008 9:11am
Quoting Ricardo.
This post by Merlin will illustrate better my opinion on developing and testing a system:

Very usefull. Never seen that thread before. Thanks
  • #16
  • Aug 19, 2008 7:17pm Aug 19, 2008 7:17pm
Quoting tdion
but only on days where particular daily candles are forming and look highly probable.
discretion?
  • #17
  • Aug 19, 2008 7:19pm Aug 19, 2008 7:19pm
Quoting Kaligula
1. Wait for trend.
2. Make sure it's trend.
3. Wait for little retracement.
4. Jump in on breakout.
Joe Ross gave the name Ross Hook to it. Very dificult to have a good reward risk with it.
  • #18
  • Aug 19, 2008 7:52pm Aug 19, 2008 7:52pm
no. to me, discretion means that i am making my mind up at the time of the trade....

instead what i have is completely mechanical and based 100% on rules.

Quoting giraia_br
discretion?
  • #19
  • Last Post: Aug 19, 2008 8:14pm Aug 19, 2008 8:14pm
Quoting giraia_br
Joe Ross gave the name Ross Hook to it. Very dificult to have a good reward risk with it.
Now I'm testing.
In my opinion if you agree to wait just one or two candles - you gonna miss couple of trends, but [hopefully] plenty of ranges.
The goal is to enter retracements only if market is still hot, not any retracement after trend.
So for example if I look at 15 min TF during London open I'll wait for one bar against trend and if next goes above it - I'm in. If it doesn't I cancel entry order.
0 traders viewing now