Proven Methodology (Trend following) - let's create an EA | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Let's create a profitable system 36 replies

BELLA SYSTEM. Let's try it, let's improve it! 118 replies

Is Grid Trading (combined trend following and counter trend) Profitable? 3 replies

Trend following or trend fading? 8 replies

Trend entry system/methodology 5 replies

Attachments: Proven Methodology (Trend following) - let's create an EA
Exit Attachments

Proven Methodology (Trend following) - let's create an EA

  • #1
  • First Post: Apr 27, 2009 10:09am Apr 27, 2009 10:09am
Hi All!

I'm a currency trader still in the process of developing a solid "bag" of profitable methodologies.

I've had a very fortunate opportunity come my way, and I want to work with traders in the FF forums to see the opportunity to its full potential!

Basically, I've met a trader who has 10+ years experience and his methodology is very systematic and easy to understand. The problem is he isn't too technically inclined (can't code EAs, isn't sure how to do robust back/forward-testing, etc)...so we're looking to the FF forum for synergy!

His methodology has been profitable...allowing him to rely on trading profits to live comfortably...on a DAILY TF. We're looking to explore the methodology in other TFs and to increase its profitability if possible.

If you're interesting in working with me on this project...join in! I'm looking for a manageable group of members that can actively participate, share suggestions, share in testing the system, and DEFINITELY someone who can CODE our methodologies/ideas as they evolve and improve.

If you're interested....read on!
  • #2
  • Apr 27, 2009 10:15am Apr 27, 2009 10:15am
AGOtrader.

SOunds good I am interested in being part of the group.

Are there any statistics with regards to the winning percentage of this method?


Quoting AGOtrader
Hi All!

I'm a currency trader still in the process of developing a solid "bag" of profitable methodologies.

I've had a very fortunate opportunity come my way, and I want to work with traders in the FF forums to see the opportunity to its full potential!

Basically, I've met a trader who has 10+ years experience and his methodology is very systematic and easy to understand. The problem is he isn't too technically inclined (can't code EAs, isn't sure how to do robust back/forward-testing, etc)...so we're looking to the FF forum for synergy!...
Believe and Achieve!
  • #3
  • Apr 27, 2009 10:17am Apr 27, 2009 10:17am
On board.
Ready for War
  • #4
  • Apr 27, 2009 10:28am Apr 27, 2009 10:28am
A simple system sounds great and yet I have found that simple systems usually require some discretion for entries or exits - hence why traders are traders and not EAer's. I wonder how much of his success is based on instincts developed through years of trading and just having a feel for the market and how much of his trading success is truely systematic that can be programmed into a black box. An example - right now I have an entry signal except price is trading into a major area of support so I am passing on the trade setup. That comes from experience and I am not sure it is programmable into a black box.
I make money anywhere in the world by taking orders
  • #5
  • Apr 27, 2009 10:44am Apr 27, 2009 10:44am
Premise:

- Follow the TREND
- Enter Trends on small pullbacks
- Use entry points that exploit common placement of STOPS (this usually gives me a "bounce" into the trade, allowing me to be quickly positive and locking the trade with a Stop at Breakeven.

The method looks at bars and gives entry "signals" based on the previous bar or few bars.

Basic Terminology:

Trend Bar - a bar that "continues" the current trend. Specifically, in an UPTrend a bar with a Higher High (in DOWNTrend, a Lower Low). We do NOT enter on these bars.

Pullback Bar - a bar that pulls back the current trend. Specifically, in an UPTrend a bar with a Lower High (than the previous bar) (in DOWNTrend, a Higher Low).

Basic Entry Setup/Signal:

(For ease, I'll only describe UPTrends moving forward, inverse applies for DOWNTrends)

- Wait for the FIRST Pullback Bar to CLOSE. On the NEXT bar, Enter the trade IF/WHEN price moves through the the Pullback Bar's High.

- If the bar the follows the FIRST Pullback Bar FAILS to make a new high (on which we would enter long), that bar is then called Pullback Bar #2 (upon its close).

- A Long trade would then be placed on during the next bar when price moves higher than Pullback #2's High.

- We wait for this setup/signal to happen UNTIL we have a 4th Pullback Bar. After 4 Pullback Bars close....we STOP looking for this trade.

AFTER ENTRY:

So let's assume that we Enter Long (in an UPTrend) once price has moved higher than the HIGH of the Pullback Bar.

Stop Loss Placement: Place STOP at the LOW of either:
a) the Current Bar (the one that's still open when you place trade)
b) the Previous Bar (the most recent Pullback Bar)

Choose either Option A or Option B (above) based on whichever one is LOWER (i.e. giving you a larger stop).

TAKING PROFITS:

This system does NOT attempt to pick a profit target. Instead, this is what it attempts to accomplish:

1) Get QUICKLY into positive territory (by exploiting stops...which will be explained later)
2) Take PARTIAL PROFIT very soon after entering the trade
3) Moving STOP to BREAKEVEN (after partial profit taken)
4) Allow the "free trade" (as Stops are now at B/E) to follow the trend until a meaningful retracement occurs (explained later)
  • #6
  • Apr 27, 2009 10:52am Apr 27, 2009 10:52am
Quoting SedonaPipper
A simple system sounds great and yet I have found that simple systems usually require some discretion for entries or exits - hence why traders are traders and not EAer's. I wonder how much of his success is based on instincts developed through years of trading and just having a feel for the market and how much of his trading success is truely systematic that can be programmed into a black box. An example - right now I have an entry signal except price is trading into a major area of support so I am passing on the trade setup. That comes from experience...
SedonaPipper,

I appreciate your discussion of the nuances of systematic and discretionary trading. I struggle with the most profitable answer to this issue!

Here is what I know/think: You HAVE to use some degree of discretion when trading...at least to the extent of being REASONABLE (i.e. not blindly following your system after major events...like Sept. 11th). BUT, systematic trading has been shown to have many many advantages.

The trader that I work with (who has used this method successfully) FOLLOWS A SYSTEMATIC APPROACH....ABSOLUTELY. I used to ask him "What do you think about the BoE rate cut?" His answer: "I don't care." Me: "What affect will XYZ have?" His answer: "I don't care"

So the good news (in a sense) is that it IS a mechanical system than can almost fully be followed without much (or any) discretion.

The other nice feature is that it uses entry setups that rely on what I condsider "bar to bar" relationships. I.e. Do x this the current bar is higher/lower than the previous bar. This makes the system work on MULTIPLE TIMEFRAMES (I think the word for this is "FRACTAL"!??). Although I suspect/experience that some TFs work better than others...this is what I'm seeking to optimize.

Thanks for your interest...happy trading!
  • #7
  • Apr 27, 2009 11:05am Apr 27, 2009 11:05am
Here is how STOPS are moved once the trade is active:

First, I say STOPS (plural) b/c you want to split the trade into 3 parts.

Initially - ALL Stops are at the low (in an UPTrend) of the current or previous bar (whichever gives a BIGGER stop)

1st STOP - used to take PARTIAL PROFIT as soon as possible. How many positive pips do we need to take profit? It depends on the TF, and the # of pips to the Stop; however, we use about 20 pips as a general rule of thumb.

So...trade is placed...it moves 20 pips positive...we do 2 things:
1) Take Profit on 1/3 of our position
2) Move the other 2 Stops to BreakEven (the price where we entered).

Next... we wait for the next bar to close (our 2 stops are at B/E).

When the next bar closes, we move 1 of the stops UP (trailing our unrealized profit) to the LOW of the most recent CLOSED BAR. (NOTE: We ONLY do this IF this moves our stop CLOSER to current price. Meaning, if stop is ALREADY at B/E, we DON'T move it Lower based on this).

So, IF the Low is Higher than the B/E level, our 2nd stop is moved up to this closed bar's Low, and the 3rd Stop STAYS at B/E...for now.

Moving the 2nd Stop: Whenever a bar closes, we move the 2nd Stop up to that bar's LOW. This "corners" the market...and the Stop will get filled (realizing part of our profit) as soon as a bar makes a LOWER LOW.

Moving the 3rd Stop: We leave the 3rd Stop at B/E until the 2nd Stop is filled. When this happens, we wait for a "swing point" (I don't know what else to call it) to form. Here's what I mean by this:

As our trade is working, consecutive bars are going to have Higher LOWs (until our 2nd stop is filled, which is on a Lower LOW)...then, IF the trade continues to work in our direction...it will form more bars with Higher LOWs. We want to place our 3rd Stop at the LOW of this "LOWER LOW Bar". This point is where the market Pulled Back (in an UPTrend), but then the UPTrend continued. The 3rd Stop stays at this point until a new "swing point" is formed.

I'm going to figure out how to edit/upload charts so all this can be seen VISUALLY!! I hope my explanations are making sense despite lack of graphics!
  • #8
  • Apr 27, 2009 11:08am Apr 27, 2009 11:08am
Count Me in...

Looking froward to it ,and helping along the way.
  • #9
  • Apr 27, 2009 11:32am Apr 27, 2009 11:32am
Chart is attached!

Can someone give a quick tip on how to post/copy the chart in the my post's body...rather than as an attachment that must be opened?

I'm sure it's easy...I'm just missing it.

Thanks!
Attached Image (click to enlarge)
Click to Enlarge Name: bp ex1.gif Size: 16 KB
  • #10
  • Apr 27, 2009 11:45am Apr 27, 2009 11:45am
In DOWNTrend...AFTER 1st Partial Profit is taken...2nd Stop is moved Down on every bar that has a LOWER HIGH. It will get filled by a bar (during this bar) that eventually forms a HIGHER HIGH.
Attached Image (click to enlarge)
Click to Enlarge Name: bp ex 2nd stop.gif Size: 17 KB
  • #11
  • Apr 27, 2009 11:54am Apr 27, 2009 11:54am
Recap:

3rd Stop stays at Breakeven UNTIL the 2nd Stop is filled.

Then, 3rd stop is moved to a "swing high" point (in this DOWN Trend Example).

It is moved lower and lower (in DOWN Trend) to each LOWER "swing high"...until a Higher "swing high" forms and fills the stop.

After the 3rd stop is filled...our entire position has been covered (remember, we split the trade into 3 portions)...and hopefully we're happy with plenty of pips!
Attached Image (click to enlarge)
Click to Enlarge Name: bp ex 3rd stop.gif Size: 17 KB
  • #12
  • Apr 27, 2009 12:18pm Apr 27, 2009 12:18pm
As a reminder, I'm HAPPY to share/contribute/drive this thread, BUT am seeking input from all the talented people on this forum!)

Ultimately, I'm anxious to work with a focused group of traders that can help:

1) code an EA
2) backtest and forward test
3) contribute ideas that make the methodology MORE robust and MORE profitable

Happy Trading to All!!
  • #13
  • Apr 27, 2009 12:26pm Apr 27, 2009 12:26pm
I can definitely help with the tech side, however there is one thing that is not clear - how do you define a "trend" mechanically? In other words, what is the exact definition on when you enter on a pullback.

By the way I trade only daily bars myself, just a sidenote, methods that work very well on daily bars don't usually work at all or work much worse on lower timeframes.
  • #14
  • Apr 27, 2009 12:34pm Apr 27, 2009 12:34pm
Let me explain a major premise of this methodology:

First, the "TREND IS YOUR FRIEND" - overused statement, but still very useful. There are plenty of countertrend systems that work, and still MORE trend-following systems that work!!

Second, a key advantage of this methodology is that the time between 1) Entry and 2) being in Positive Territory (i.e. up ~20 pips) is fairly short (not always, of course, as no methodology is 100%).

This quick "bounce" into positive pip territory allows us to:
1) take partial profits
2) "lock" the rest of the trade so we can't lose (stops at breakeven)
3) and then we can ride the trend for as long as it trends

HOW do we get that "bounce"??

The previous charts showed how I profit from a DOWN Trend, so I'll continue that example:

In every Down Trend, there are traders that WANT to go LONG. These "Bulls" like the market is oversold, forming a "bottom," whatever... So in this downtrend their ARE traders that establish LONG positions (everyone agree???).

These Bulls that go Long, have to place stops to limit losses if they are wrong (and the market continues to go down). We can ESTIMATE where (i.e. at WHAT PRICE) a large accumulation of stops (from all the Bulls) will be placed. The stops are often at the LOW of a bar - they expect prices to go up (hence their long trade) and they will admit they are "wrong" when prices go lower than a recent low.

So....these are the prices that we use to ENTER our SHORT trades. The protective stops for the Bulls/Long Trades are SELL STOPS...triggering a high density of SELL STOPS will push prices lower, fairly quickly, and hopefully quickly enough for us to "lock" the trade.

Recap of advantages:

1) Trade WITH the Trend
2) Get into the trend on a pullback
3) Exploit accumulation of Stops to get quick "bounce" into profitable territory
4) Quickly capture partial (1/3) profits, allowing the rest to ride the trend with no risk of a negative trade (after partial profits locked).
  • #15
  • Apr 27, 2009 12:42pm Apr 27, 2009 12:42pm
Quoting mikkom
I can definitely help with the tech side, however there is one thing that is not clear - how do you define a "trend" mechanically? In other words, what is the exact definition on when you enter on a pullback.

By the way I trade only daily bars myself, just a sidenote, methods that work very well on daily bars don't usually work at all or work much worse on lower timeframes.
Mikkom,

Thanks for jumping in!

First, how do we define a trend: This IS difficult mechanically (as in WITHOUT human discretion)...and in fact, one of the challenges that I'm hoping to gain insight from fellow FF members.

My main goal would NOT be to create an EA (at least intially) that can identify trend. I/we may have to settle on an EA that executes once the user has established the trending direction.

However, there are some thoughts we can consider:

We should probably only consider an UP Trend once NEW HIGHS are reached ("new" is of course relative to how far back you look). Then, we could consider "still" in an UP Trend as long as Higher Highs are produced. Remember, we enter long after 1, 2, OR 3 Pullback Bars. AFTER a 4th Pullback bars closes, we no longer look for Long trades. I'm not saying we're NOW in a DOWN trend (just b/c we got 4 Pullback Bars in the UPtrend), but we should look for a developing Down Trend.

Your other mention about Daily methods NOT working well on smaller TFs. I'm NOT surprised you say this...hence the challenges I've been facing and attempting to correct. It may just not work on smaller TFs, but I'm motivated to explore!

Thanks again for your interest and feedback!
  • #16
  • Apr 27, 2009 1:09pm Apr 27, 2009 1:09pm
Quoting AGOtrader
Hi All!

I'm a currency trader still in the process of developing a solid "bag" of profitable methodologies.

...............

If you're interesting in working with me on this project...join in! I'm looking for a manageable group of members that can actively participate, share suggestions, share in testing the system, and DEFINITELY someone who can CODE our methodologies/ideas as they evolve and improve.

If you're interested....read on!
you have caught my attention.

Ron
  • #17
  • Apr 27, 2009 1:15pm Apr 27, 2009 1:15pm
Hi,
I have built numerous EA's and have built something similar to what you have described, although only a cursory glance, dinners ready. I'll read fully and let you know. My system is modular and very easy to incorporate new ideas etc.
Regards
Steve

Quoting AGOtrader
SedonaPipper,

I appreciate your discussion of the nuances of systematic and discretionary trading. I struggle with the most profitable answer to this issue!

Here is what I know/think: You HAVE to use some degree of discretion when trading...at least to the extent of being REASONABLE (i.e. not blindly following your system after major events...like Sept. 11th). BUT, systematic trading has been shown to have many many advantages.

The trader that I work with (who has used this method successfully) FOLLOWS A SYSTEMATIC APPROACH....ABSOLUTELY....
  • #18
  • Apr 27, 2009 1:55pm Apr 27, 2009 1:55pm
Quoting AGOtrader
Hi All!

I'm a currency trader still in the process of developing a solid "bag" of profitable methodologies.

I've had a very fortunate opportunity come my way, and I want to work with traders in the FF forums to see the opportunity to its full potential!

Basically, I've met a trader who has 10+ years experience and his methodology is very systematic and easy to understand....
If it's a coder what you need, here I am ;-)
I'll send a bottle of wine to my first voucherer ;-)
  • #19
  • Apr 27, 2009 4:51pm Apr 27, 2009 4:51pm
Quoting richboy1972
If it's a coder what you need, here I am ;-)
Hey Richboy (hope your name is fitting )

Thanks for joining it!

Did you get enough of the explanation to start coding a rough form of the methodology? Are you familiar with coding similar methodologies - ones that rely on bar to bar comparison (rather than indicators/oscillators/etc)?

Let me know what you think...
  • #20
  • Apr 27, 2009 8:34pm Apr 27, 2009 8:34pm
Quoting mikkom
I can definitely help with the tech side, however there is one thing that is not clear - how do you define a "trend" mechanically? In other words, what is the exact definition on when you enter on a pullback.

By the way I trade only daily bars myself, just a sidenote, methods that work very well on daily bars don't usually work at all or work much worse on lower timeframes.
How about using the attached indicator to define the trend?
When it crosses consider the trend changed then look for the entry signal as described in the earlier posts.

It may even help if the retracement stops around a fib number.
Attached Image (click to enlarge)
Click to Enlarge Name: eurusd.gif Size: 18 KB
Attached File(s)
File Type: mq4 MAMA_v1.mq4   11 KB | 636 downloads
0 traders viewing now