Symphonie EA

[Evaluator]

Steve,

Thank you very much for the time and effort in providing this EA. I am excited to try it and see how it compares to the Manual Method of trading Symphonie Trader System. Someone posted this on the Symphonie thread and I am just reading through all the posts and I am glad to see so many people with interest.

After downloading the indikators and the EA I did notice something that causes concern. First, when you load the four indikators into the chart one is unable to change the setting for optimization. Is it possible to have all these new coded indikators have the input settings changable? We have found on the Symphonie thread that sometimes changing the input settings to make more or less sensitive to market conditions provides better results. I realize from an EA perspective this is not something that an EA does but when trading manually it is helpful. Also, I would like to ''evaluate'' these new indikators for functionality.

Second, as to the EA itself is it also possible to incorporate these unput settings into the EA so one may change these conditional settings for the individual indikators to help optimize the EA. I like the setting you have made so far and the 3 exit strategies is wonderful. If you read the post#1 on the main Symphonie Thread you should find updated information as I try to keep it current and up to date.

Here are my current indikator input settings that I think are optimal. Again, it would be nice to be able to modify them in the individual indikators and the EA itself.

Extreme indikator
1,3,2, false, 1,12,21

Emotion indikator
7
50.6
3000

Trendline indikator
30, 5

Sentiment indikator
12

I have found that using mutilple timeframes as confirmation is a must when trading bad whipsaw markets and I use a method I call Symphonie 60-5 there I am watching the 1 hour timeframe for overall market direction (market trend) and then use only 5 min Symphonie Signals that agree with the hourly. This signifcantly cuts down on the whipsaws (consolidation periods) and allows one to take the trades where there is a major market move. Again see post#1 for the link to post #8647 which is a detailed explaination with screenshot explaination. (http://www.forexfactory.com/showthread. ... ost5470997)

Since I am just now reading all the posts I would like to respond to several questions and comments made in your thread on my system.

mbkennel wrote:

A real indicator with some modest value written by a mathey guy was intentionally modified by a very clever and unscrupulous commercial scammer at some point (don't think anybody here is involved).

You are probably corerct. However, I do not know the backgrounds of each indikator other than what was posted. I can tell you that it took me 18 months of combining 100s of indikators into 100s of models before I developed one that fit what I was wanting the system to do. Symphonie is not a perfect top or bottom picker. It is designed to be a follower of market trends and will always enter on the back side of a directional change because Symphonie is looking to enter after the market has decided on a price action directional change so one using Symphonie will never be at the absolute top or bottom of price action. (see attached design flowchart)

As to the coding of the indikators....I have no idea. I am a trader and model builder not a programmer so I leave the coding whether it is good or bad up to you guy the coders and programmers. So, none of the original indikators were coded or modified by me. I only used the input settings for optimization.

batemap wrote;

I also notice that because of the repainting of this indi MajorCycleBuyHistory/MajorCycleSellHistory dots are shown, perhaps the EA needs to read these as well? I say this because I saw the EA change from showing a buy (x candles ago ) then after an indi repaint the EA was showing a sell

Repaints of the extreme indikator has been a major issue and I have tried to be upfront and clear that this is a repainting indikator because it is always try to pick the absolute extremes (therefore the name Extreme indikator) This indikator has also been a mess and from what I understand from people who are coders tell me this indikator has been written and then rewritten over and over again by multiple people and there is alot of crap in the coding of this indikator. People have tried to clean it up but it seems the clean up has not done very much and the last person to post a revised Extreme indikator removed some of the algorithms to the indikator and changed the input settings but it still repaint. The best re-write was seashore where it marks the repaint with a dot on the indikator. It does not change the input setting but just marks the repaint. You can find this indikator on post#1 or post #2545 (http://www.forexfactory.com/showthread. ... ost5124304)

garyfritz wrote;

The "future peeking" problem worried me about Symphonie too. The thing that kept tempting me was Evaluator's reported results. Either he was a liar, or he was producing some incredible results with those indicators.

garyfritz, I can assure you that I am not a liar and those are indeed my results. In good trending markets where trials 1 and 2 were very profitable those are the real results. If you look to trial 3 you will see the results have been very moderate because for the past 3 months the markets have been stuck in major consolidations. Also, why would I lie?? I have nothing to hide and it does what it does and I just trade what the Symphonie Signals tell me. I really do not even look at the results until the end of the week when I make the stats and that is just basically having the Empty4 make the report up and then posting the ending statistics. Many people want me to post trade by trade and the whole point of my NOT doing that is when you look back in history it would just confuse people because of repaint or double down moments, or confirmations that a historical look at the charts will not match up. It is just better to print the overall states and that is why I do it the way I do it. Some people have pushed for this myfxbook.com website and I am not familiar with it and do not feel comfortable letting some other program inside my computer. It is a privacy thing and I do not feel comfortable allowing it into my computer. So, if you don't like my opinion then my response is too bad. It is my preference and you can take it or leave it as you wish. But I can assure you I an not lie on my results. (Mediator that one is just for you)

lhDT wrote;

The problem, as almost all trending strategy is ranging market. If you trade using that strat during that time you will be simply short killed.

That is always the problem since post #1 my first concern has been massive whipsaws and to develop a system that will help me and others overcome the killing effects of whipsaw markets. As I stated in my very first sentence when I started the thread.....

I have been testing many indicators and programs for years to determine which combinations will work the best and most importantly which combinations are the most reliable especially in a whipsaw market like we are having today.

Whipsaws are killers and there must be some way to overcome them and make a system and cooresponding EA out of it that will help one to overcome this obstacle. I did make a post on the thread (again you can find the link in post#1) that discusses the topic of ranging verus trending with interesting statistics for easy identification. You may wish to look it over (post #8129) as it may help in defining parts of the EA to improve performance. (http://www.forexfactory.com/showthread. ... ost5456200)

OK. Enough with my ramblings. I hope you will find some points of interest and help in your quest for a winning EA using the Symphonie Trader System. I am always willing to help in any way or answer any question that may assist you in your quest.

Best of Luck!!!

Evaluator.

[garyfritz]

Evaluator, in case I was not clear: I didn't think you were lying, and I'm sorry if I made it sound that way. You have enough people getting good results that it seemed clear there was something worthwhile there. I looked at your indicators long enough to see that there was repainting, but the multiple-confirmation behavior of your indicators seemed to resolve most repainting issues. I.e. if Extreme was still repainting, then almost certainly the other 3 indicators had not lined up in agreement.

Unfortunately while I can find my way around in an indicator, I am not expert enough to fully understand the causes of repainting. I would guess that the problem comes because the Extreme indicator plots its spike as soon as it sees a potential Extreme condition, but then later market action violates the Extreme condition so the indicator "un-paints" it.

Has anyone tried this? Have the Extreme indicator note that a potential Extreme condition has happen, but check the values of the other 3 indicators and don't paint an Extreme spike until the other 3 indicators agree. (Maybe it could plot a dot when it sees the potential condition, the way it does now after the fact, and replace the dot with a spike when the other indicators line up.) That way it wouldn't paint the spike until it was time to make a trade. Hopefully that condition would NOT repaint.

I don't know enough about the indicators' behavior to know if that Extreme spike is the only problem or not, but it's an idea...

[SteveHopwood]

After downloading the indikators and the EA I did notice something that causes concern. First, when you load the four indikators into the chart one is unable to change the setting for optimization. Is it possible to have all these new coded indikators have the input settings changable? We have found on the Symphonie thread that sometimes changing the input settings to make more or less sensitive to market conditions provides better results. I realize from an EA perspective this is not something that an EA does but when trading manually it is helpful. Also, I would like to ''evaluate'' these new indikators for functionality.

Second, as to the EA itself is it also possible to incorporate these unput settings into the EA so one may change these conditional settings for the individual indikators to help optimize the EA. I like the setting you have made so far and the 3 exit strategies is wonderful. If you read the post#1 on the main Symphonie Thread you should find updated information as I try to keep it current and up to date.

Hi Evaluator

Fantastic to see you here. I hope you will continue to contribute as I am trying to code your system; this is difficult to do without you.

Re the above from your post, I quote this from post 1 here:
"You need the indi versions uploaded here to work with the EA (and read post 2 here as I have acted on it). I have not touched the indi calculation code, apart from the Sentiment code as described in my previous sentence. I have removed all the external inputs that usually appear when you drag an indi onto your chart. EA code requires indicators to be called with a full list of external inputs. This can become complicated when there are a lot, as with the Extreme indi, so removing them makes the coding easier."

So, I did nothing to the code in the indis that I attached for use with the EA. All I did was remove the ability for the user to change the inputs in order to make a horribly complicated EA slightly easier to code. And yes, I can indeed return to the original indis and replicate their inputs in the EA, but this causes a lot of extra work and I needed a good reason to do so. I had no reason to suppose from the thread here that there were good reasons to do so, and am not a mind reader.

I will work on adding the indicator inputs to the EA over the next few days. I was simply trying to save myself some work. Nothing in any of the posts that I read suggested to me that people were experimenting with different settings in the indis. There are lots of other people but only one of me, and this is not the only project I am involved with. I rely on others to step in with information from other posts/threads/forums; I am not a mind reader.

So, more updates to come with inputs that people can play with to their hearts content. It would be great to be able to produce an EA that at least comes reasonably close to trading your system as you would want it to trade.

Cheers

[Evaluator]

Hi Evaluator

Fantastic to see you here. I hope you will continue to contribute as I am trying to code your system; this is difficult to do without you.

Re the above from your post, I quote this from post 1 here:
"You need the indi versions uploaded here to work with the EA (and read post 2 here as I have acted on it). I have not touched the indi calculation code, apart from the Sentiment code as described in my previous sentence. I have removed all the external inputs that usually appear when you drag an indi onto your chart. EA code requires indicators to be called with a full list of external inputs. This can become complicated when there are a lot, as with the Extreme indi, so removing them makes the coding easier."

So, I did nothing to the code in the indis that I attached for use with the EA. All I did was remove the ability for the user to change the inputs in order to make a horribly complicated EA slightly easier to code. And yes, I can indeed return to the original indis and replicate their inputs in the EA, but this causes a lot of extra work and I needed a good reason to do so. I had no reason to suppose from the thread here that there were good reasons to do so, and am not a mind reader.

I will work on adding the indicator inputs to the EA over the next few days. I was simply trying to save myself some work. Nothing in any of the posts that I read suggested to me that people were experimenting with different settings in the indis. There are lots of other people but only one of me, and this is not the only project I am involved with. I rely on others to step in with information from other posts/threads/forums; I am not a mind reader.

So, more updates to come with inputs that people can play with to their hearts content. It would be great to be able to produce an EA that at least comes reasonably close to trading your system as you would want it to trade.

Cheers

Steve,

My apologies if this causes you any problems as my comments were not meant to cause you extra work. I do feel it is important to have the ability to change the input settings. On the Symphonie thread the original trendline was 50,5 then we found that 63,18 was very reliable, now we use 20,5 and 30,5. As to the extreme indikator that is the one that is caused the most problem. The original setting was 1,3,4, false, 1,12,21 but that level seems to make system sluggish so we have moved from 1,3,2, false, 1,12,21 to 1,3,3, false, 1,12,21 and have yet to find a setting that feels right in guaging the market.

That is why I made the suggestion as to changing the input settings for the EA because the results will vary when changing these settings to meet market condition.

I hate to make your life more difficult but you may also wish to code a Symphonie Matrix format to your revised indikators as this will provide you with a ease of use if you wish to test the EA against the actual signals. When people come up with changes I find this a valuable tool in testing because you can see the signals in a glance. Also, ldHT made a wonderful experimental signaling indikator that also tracks in multiple timeframes that may assist you in coding an EA or trying the Symphonie methods in multiple timeframe. KUDOS to ldHT!!!!! (p.s. this is only the .ex4 file and you will have to ask ldHT for the .mq4 as he has only posted the .ex4 version)

I can tell you that the more complex these Signalers have become the more an issue of refreshing has caused because the combination of the four indikators reached the capacity limits of the Empty4 program. But I include these files for your reference and possible you may find them of value.

Evaluator.

[Evaluator]

Evaluator, in case I was not clear: I didn't think you were lying, and I'm sorry if I made it sound that way. You have enough people getting good results that it seemed clear there was something worthwhile there. I looked at your indicators long enough to see that there was repainting, but the multiple-confirmation behavior of your indicators seemed to resolve most repainting issues. I.e. if Extreme was still repainting, then almost certainly the other 3 indicators had not lined up in agreement.

Not a problem and ready to move on. But I wanted to make it clear that I have tried to be upfront and open about the results.

You are exactly right. Symphonie is build on 4 indikators and it seems everyone becomes obsessed with the extreme indikator as being the ''holy grail'' when in reality it is just a signal flare warning you of a possible top ''maybe'' (key word there) forming. Once you become totally focused on the extreme indikator is the moment you start to fail in using Symphonie. One must wait on all four indikators or else you are not grounded and your account will face loss trades. But the indikators are only 1/3 of Symphonie. The 2/3 is the Money Management rules and guidelines that keep on focused on account growth and account balance protection. This ties into the last component of Symphonie and that is ''you'' the trader. Symphonie teaches one to be patient, wait on signals, and look at the overall market conditions for patterns and direction to keep you grounded in how the market is functioning. By discipline, patience one develops trust and confidence in making one a more consistant winning trader.

So there you have it....the long version.

Evaluator.

[SteveHopwood]

Steve,

My apologies if this causes you any problems as my comments were not meant to cause you extra work. I do feel it is important to have the ability to change the input settings.

Hehe. As it happens, I could not resist and am making the changes now. I will be able to upload them in the next few minutes.

In the meantime, be aware of how much I loathe custom indi's and coding for them and so am ignoring the matrix stuff. *Fingers in ears* La la la la la la la etc

[slowkey]

I have the EA now up on a clean DEMO running it on All Cowboy IBFX pairs. It immediately opened 4 positions with minimal draw down. Now this is during the Asian session so I do not expect much. I always chuckle over repainting. Like the holy grail to have that on off switch that does it all for you. Anyway I have been using a MTF Stochastic wallaby that Rob Booker developed with some unusual setting that capture both trend and match up with the extreme cycles nicely. Just smooths out on-off nature of the cycle indicator in my mind. But when you do that you clearly see that the cycle indicator has value in this system when you look at it as just one piece of the whole trading system . The extremes almost all the time match up with extreme trend reversals. With all indicators eventually triggering per rules of the system. Anybody wants the indicator can Google it and find it on Ron's Site. I hide the individual time frame stochastic's and just use the Wallaby line. It is the average of 3 time periods. On the 5 minute I look forward to higher time frames to get trend and filter out the noise.

[Evaluator]

Steve,

My apologies if this causes you any problems as my comments were not meant to cause you extra work. I do feel it is important to have the ability to change the input settings.

Hehe. As it happens, I could not resist and am making the changes now. I will be able to upload them in the next few minutes.

In the meantime, be aware of how much I loathe custom indi's and coding for them and so am ignoring the matrix stuff. *Fingers in ears* La la la la la la la etc

Steve,

I understand. But you must admit to go from this... Screenshot 209 (bottom)...... to this Screenshot 210(top) is many improved because it cleans up your chart window dramatically to keep you focused on the big picture price action by making it many easier to read.

Evaluator.

P.S. The indikators are the same but the Matrix just reorganizes them in a easier format for reading the signals.

[SteveHopwood]

Latest update in post 1.

You can now dump the indi's modified for use with S, which now uses the indi's from E's thread. S includes a list of inputs relevant to the indis - the hidden ones are merely true/false alert inputs.

This should be ok, but sing out if bugs surface.

[Evaluator]

Latest update in post 1.

You can now dump the indi's modified for use with S, which now uses the indi's from E's thread. S includes a list of inputs relevant to the indis - the hidden ones are merely true/false alert inputs.

This should be ok, but sing out if bugs surface.

Steve,

I thought you cleaned up the code in the individual indikators for your EA? Did I misread?

Evaluator.