Flyer! EA - New Thread! | Page 12 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

My trade explorer Flyer > 1000% return for May 3 replies

Flyer! EA - New Thread - Part 2 147 replies

Fozzy, Pip Boxer, Flyer 4 replies

Help Test my new EA, FLYER! 243 replies

Flyer! Version 2.00 is released. 0 replies

Attachments: Flyer! EA - New Thread!
Exit Attachments

Flyer! EA - New Thread!

  • #221
  • Nov 15, 2006 5:43am Nov 15, 2006 5:43am
Hi GirlFlyer,

While this thread seems to have fallen out of favour, I thought that you may be interested in updated results from one of the test runs that I'm still runing.

Results look promising
Attached File(s)
File Type: zip 20061115FXDD.zip   10 KB | 336 downloads
Just my 2 pips worth
  • #222
  • Nov 15, 2006 8:20am Nov 15, 2006 8:20am
Quoting saltcafe
Hi GirlFlyer,

While this thread seems to have fallen out of favour, I thought that you may be interested in updated results from one of the test runs that I'm still runing.

Results look promising
From my testing, one needs to exit out of positions manually to avoid big losses. The results are very good if the pairs trend very well, but that had not been the case for some time....
  • #223
  • Nov 16, 2006 2:40am Nov 16, 2006 2:40am
Quoting et_phonehome_2
From my testing, one needs to exit out of positions manually to avoid big losses. The results are very good if the pairs trend very well, but that had not been the case for some time....
The way I look at is a little bit different ... if the pairs have NOT been trending and this IS a trending EA, then the sheer fact that it hasn't lost money is a success in itself. Once the pairs start to trend, which they inevitably will, then this should prove to be a small gold mine.
Just my 2 pips worth
  • #224
  • Nov 17, 2006 10:58am Nov 17, 2006 10:58am
Hey everyone!

I'm back with some limited access to the internet until I get my permanent connection. But, I'm moved into my place in Denmark and all is going well.

Look to see an update to the EA soon!

Good luck all!

GirlFlyer
  • #225
  • Nov 17, 2006 11:32am Nov 17, 2006 11:32am
Quoting GirlFlyer
Hey everyone!

I'm back with some limited access to the internet until I get my permanent connection. But, I'm moved into my place in Denmark and all is going well.

Look to see an update to the EA soon!

Good luck all!

GirlFlyer
Welcome Back GirlFlyer... we are some who still wait for your return ;-)
  • #226
  • Nov 17, 2006 11:34am Nov 17, 2006 11:34am
Girlflyer,

Nice to see you back again, it seems like ages. I hope to see the new EA soon.....
  • #227
  • Nov 17, 2006 12:46pm Nov 17, 2006 12:46pm
Quoting GirlFlyer
Hey everyone!

I'm back with some limited access to the internet until I get my permanent connection. But, I'm moved into my place in Denmark and all is going well.

Look to see an update to the EA soon!

Good luck all!

GirlFlyer
Nice to have you back ... hope the move didn't wreck too much havock with your senses. Having moved countries (continents) numerous times myself, I always hate to have to pack all of the crap up. These days I travel light 2-3 bags and my trusty old laptop ... speaking of which, maybe it's time for me to upgrade?!?!

In regards to the next version, then are you still planning to combine the two versions earlier released (A & B). And if so what is the methodology for it?

If I'm not mistaken then the two version differed by the exit strategy and I thought that you were merging them by opening up two half sized postions and exiting one with the A strategy and the other half with the B strategy? Am I on the right track here? or something different?
Just my 2 pips worth
  • #228
  • Nov 20, 2006 10:40am Nov 20, 2006 10:40am
Quoting saltcafe
Nice to have you back ... hope the move didn't wreck too much havock with your senses. Having moved countries (continents) numerous times myself, I always hate to have to pack all of the crap up. These days I travel light 2-3 bags and my trusty old laptop ... speaking of which, maybe it's time for me to upgrade?!?!

In regards to the next version, then are you still planning to combine the two versions earlier released (A & B). And if so what is the methodology for it?

If I'm not mistaken then the two version differed by the exit strategy and I thought that you were merging them by opening up two half sized postions and exiting one with the A strategy and the other half with the B strategy? Am I on the right track here? or something different?
Hi SaltCafe & Everybody Else,

Thanks for hanging in there....

Regarding the new version, the reason why I'm taking so long to update this one is because it is a total upgrade. I have thrown a lot of new techniques and ideas into this one with some promising results, but I'm still fine tuning it right now before I let it out. It takes the best of both my old Version A & B, combines some improved ideas for entries and exits and so far the back testing is giving me about 75-80% on the success rates. While this is a nice percentage, the profit factor is still a little low per trade which is causing the EA to pretty much break even over time. Until I get that resolved, I really don't want to let it go.

Salt, I'm not using split trades in the program. As much as that sounds like an idea, keeping track of every open order in the computer can be a bit of a programming nightmare and right now, the EA is getting complicated enough. The early versions were pretty simple, but now the code is growing and the possibility of getting errors in to the code is getting higher, also.

To explain the EA now in a post is too lengthy. The principle behind the EA is still the same, but there a lot of other factors now that decide the entries and exits. I will write it all out in detail in a document when I finish.

Hang in there everyone!

GirlFlyer
  • #229
  • Nov 20, 2006 10:48pm Nov 20, 2006 10:48pm
Nice to see you back. I hope you are still planning on including a manual input for selecting trading hours in any new version. I wouldn't mind seeing one in the 105A version. As I have been having excellent success with it the last 3 weeks to enter trades and then monitoring the trades and manually closing the trades. I sometimes think writing EA code to profitably exit trades is probably far more difficult than writing EA code for entering trades.

dbcmkx
  • #230
  • Nov 21, 2006 3:35am Nov 21, 2006 3:35am
Quoting GirlFlyer
Salt, I'm not using split trades in the program. As much as that sounds like an idea, keeping track of every open order in the computer can be a bit of a programming nightmare and right now, the EA is getting complicated enough. The early versions were pretty simple, but now the code is growing and the possibility of getting errors in to the code is getting higher, also.
Hi GirlFlyer,

In relation to the split trades idea then I was thinking of implementing it through either two seperate magic numbers or a comment for a particular order. Then when you have the logic to close trades, you read what the magic number is (or comment) and based on that decide how to close the trade. It sounds relitivelly simple, but I guess the diificulty is always in the detail.

Anyway, let me know if you would like a hand in this? (If you want to pursue this path)

Thanks, Salty.
Just my 2 pips worth
  • #231
  • Nov 21, 2006 8:43am Nov 21, 2006 8:43am
Hi Marci,

Glad to hear you finished the move. I haven't moved countries and that alone was a bear so i can appreciate moving even further and the boxes just sitting waiting to be unpacked.. Gool luck in the new place. Oh where do we send the house gifts to?

Rick /New Jersey USA
  • #232
  • Nov 21, 2006 8:44am Nov 21, 2006 8:44am
Quoting crusinnj
Hi Marci,

Glad to hear you finished the move. I haven't moved countries and that alone was a bear so i can appreciate moving even further and the boxes just sitting waiting to be unpacked.. Gool luck in the new place. Oh where do we send the house gifts to?

Rick /New Jersey USA
Hope that was clear that I moved twice in the past two years but within the same State and as mentioned above it was a bear...

Good Luck again
  • #233
  • Nov 22, 2006 9:44am Nov 22, 2006 9:44am
Quoting dbcmkx
Nice to see you back. I hope you are still planning on including a manual input for selecting trading hours in any new version. I wouldn't mind seeing one in the 105A version. As I have been having excellent success with it the last 3 weeks to enter trades and then monitoring the trades and manually closing the trades. I sometimes think writing EA code to profitably exit trades is probably far more difficult than writing EA code for entering trades.

dbcmkx
It is in my opinion. The right/wrong exit can make all the difference between a profitable and a losing EA. Interesting thing I'm finding out as I go through my tests that a 1 point change in your entry/exit point decisions can make a difference in your profit by sometimes 1000's of dollars.
  • #234
  • Nov 23, 2006 2:58am Nov 23, 2006 2:58am
Quoting GirlFlyer
It is in my opinion. The right/wrong exit can make all the difference between a profitable and a losing EA. Interesting thing I'm finding out as I go through my tests that a 1 point change in your entry/exit point decisions can make a difference in your profit by sometimes 1000's of dollars.
Over a period of time ... those pips here and there definatelly do add up. However if a one pip difference here or one pip difference there is making such a significant difference in the results then it is an area of concern. Without seeing the results it's hard to comment, however I hope that it doesn't mean that problem actually turns a profitable EA into a loosing one. That would mean that the EA would too sensitive to the anticts that brokers use and that wouldn't be nice.

Salty.
Just my 2 pips worth
  • #235
  • Nov 24, 2006 5:51am Nov 24, 2006 5:51am
From my testing of the EAs [both versions], you will see trades which are very profitable only when there is a trend, but if the EA does not exit out, the position will turn out to be a losing one. No matter how good a position is, we want to at least have some type of SL triggered to B/E or trail it to at least keep some of the profits. Of course, all of the settings used during my test was using the default settings as recommended.
  • #236
  • Nov 24, 2006 10:34am Nov 24, 2006 10:34am
Quoting saltcafe
Over a period of time ... those pips here and there definatelly do add up. However if a one pip difference here or one pip difference there is making such a significant difference in the results then it is an area of concern. Without seeing the results it's hard to comment, however I hope that it doesn't mean that problem actually turns a profitable EA into a loosing one. That would mean that the EA would too sensitive to the anticts that brokers use and that wouldn't be nice.

Salty.
I should clarify myself, it is not really a 1pip decision as far as the overall price is concerned. The EA has quite a few different settings involved and the changing of one of those settings by one point can affect the results dramatically because that one point can cause entry points depending on the setting by multiple pips.

Anyway, good news is by Monday I'll have the new version 2.00 ready. I've finally got it where I want it to be and now I'm just cleaning everything up and finding some optimal settings.

Preview of what will be happening!

First let me say that Flyer! has evolved into its own methodology has developed into more than what it started as. I hope that everyone enjoys the strategy. But onwards now.


#1 - Manual setting of trading hours is now possible. Works in GMT time. Set the hours you don't want it to trade, and no trading will happen.

#2 - Program is more adaptive to market now and takes into account trends a little better. Entry points remain generally the same idea but I've got it designed to jump in a little quicker if possible. Also, if a quick move causes a signal, but quickly reverses, the EA won't respond to it if it is counter trending. Additionally, both long and short positions can be possible at the same time.

#3 - Exit strategy is completely updated. Several possibilities are now used for exits. On large moves with little or no reversal, the EA will hang on to the position generally as seen in Version 1.05A. This creates huge profit potentials. However, if the price action begins a reversal, it now will jump out if the profit falls too quickly or too much (usually if a quick reversal occurs, 50-75% of the maximum profit is achieved). Overall, the EA now will always get out at a profit regardless if the price exceeds around 35 pips profit or so from the entry price.

As in previous versions, if the price immediately turns around from the entry and goes bad without ever taking a profit of significance, the S/L will exit the trade.

However, with the way things are now, you will not likely see profitable positions turn into losing positions unless the market reacts extremely quick. I think you will find this the most innovative feature of the EA.

#4 - Settings for the EA for the tested pairs will be already installed in the EA. You won't have to set them. There are many more variables than what I had before and having the user set them would be difficult to explain, so I've just made it so that you don't have to mess with the majority of them. You can still change things such as S/L values, etc. But the more complicated ones are a no worry factor.

Hope this whets your appetite for what is to come.... and if you want to drool a little more, the EA showed on backtests (no data problems this time and with 90% modelling quality) with the usual 8% investment, GBP/USD turned $10000 into $30000 in 9 months.

Look for the release sometime Monday! Good luck all!
  • #237
  • Nov 24, 2006 1:46pm Nov 24, 2006 1:46pm
Hi, I'm new here. I would like to ask a question. let say I got USD 500 only in my account and I am playing mini, How do I set your EA?
  • #238
  • Nov 26, 2006 9:35am Nov 26, 2006 9:35am
Quoting GirlFlyer
I should clarify myself, it is not really a 1pip decision as far as the overall price is concerned. The EA has quite a few different settings involved and the changing of one of those settings by one point can affect the results dramatically because that one point can cause entry points depending on the setting by multiple pips.

Anyway, good news is by Monday I'll have the new version 2.00 ready. I've finally got it where I want it to be and now I'm just cleaning everything up and finding some optimal settings.

Preview of what will be happening!

First let me say that Flyer! has evolved into its own methodology has developed into more than what it started as. I hope that everyone enjoys the strategy. But onwards now.


#1 - Manual setting of trading hours is now possible. Works in GMT time. Set the hours you don't want it to trade, and no trading will happen.

#2 - Program is more adaptive to market now and takes into account trends a little better. Entry points remain generally the same idea but I've got it designed to jump in a little quicker if possible. Also, if a quick move causes a signal, but quickly reverses, the EA won't respond to it if it is counter trending. Additionally, both long and short positions can be possible at the same time.

#3 - Exit strategy is completely updated. Several possibilities are now used for exits. On large moves with little or no reversal, the EA will hang on to the position generally as seen in Version 1.05A. This creates huge profit potentials. However, if the price action begins a reversal, it now will jump out if the profit falls too quickly or too much (usually if a quick reversal occurs, 50-75% of the maximum profit is achieved). Overall, the EA now will always get out at a profit regardless if the price exceeds around 35 pips profit or so from the entry price.

As in previous versions, if the price immediately turns around from the entry and goes bad without ever taking a profit of significance, the S/L will exit the trade.

However, with the way things are now, you will not likely see profitable positions turn into losing positions unless the market reacts extremely quick. I think you will find this the most innovative feature of the EA.

#4 - Settings for the EA for the tested pairs will be already installed in the EA. You won't have to set them. There are many more variables than what I had before and having the user set them would be difficult to explain, so I've just made it so that you don't have to mess with the majority of them. You can still change things such as S/L values, etc. But the more complicated ones are a no worry factor.

Hope this whets your appetite for what is to come.... and if you want to drool a little more, the EA showed on backtests (no data problems this time and with 90% modelling quality) with the usual 8% investment, GBP/USD turned $10000 into $30000 in 9 months.

Look for the release sometime Monday! Good luck all!

let say I got USD 500 only in my account and I am playing mini, How do I set your EA
  • #239
  • Nov 27, 2006 11:30am Nov 27, 2006 11:30am
Hello Everyone:

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o> </o>

Here it is, the long awaited version 2.00 of Flyer!

<o> </o>

This has been a while in coming for several reasons, mainly because of my move from <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:City w:st="on">Prague</st1:City> to <st1:country-region w:st="on"><st1lace w:st="on">Denmark</st1lace></st1:country-region>, but also because of the difficulty in coming up with an EA that can be consistently profitable.

<o> </o>

This new version is much more complicated than any of the Version 1 variations. The principles behind the EA are still the same, but allows for more flexibility on entries and exits although it is biased towards finding better exits as I felt the previous versions lacked in that area the most.

<o> </o>

I have only been able to fully back test the EA on EUR/USD, GBP/USD, and USD/CHF. For those of you wishing to use other pairs, you will need to come up with some settings for yourself that work. Feel free to message or email questions.

<o> </o>

Of the three pairs I have tested, Flyer! performs by far the best with GBP/USD. Strangely this notoriously volatile pair gives the best results. I believe this has something to do with the fact that cable is usually trending somewhere… it doesn’t like to sit and consolidate like many of the other pairs, which will degrade the performance of the EA.

<o> </o>

For a more complete description of the EA, read the attached document, but below is a summary of the EA’s features.

<o> </o>

 

  1. It’s generally a moving average crossover EA with some variations.
  2. Exit strategy changes dynamically according to the market conditions.
  3. You can now limit the trading hours of the EA.
  4. Trailing Stop Feature can be enabled/disabled.
  5. Most data settings are now pre-set in the code to reduce complexity for the user.
  6. As the market changes, you can modify the settings to adjust the entry/exits to match the market behavior.

<o> </o>

Attached you will find the documentation for the EA, some sample back tests, and some charts of selected trades, and the code. Hope everyone enjoys!

<o> </o>

GirlFlyer

<o> </o>

Attached File(s)
File Type: doc Flyer.doc   34 KB | 629 downloads
File Type: mq4 FlyerV200.mq4   32 KB | 618 downloads
File Type: zip FlyerSamples.zip   55 KB | 570 downloads
  • #240
  • Nov 27, 2006 11:40am Nov 27, 2006 11:40am
Quoting diablo80
Hi, I'm new here. I would like to ask a question. let say I got USD 500 only in my account and I am playing mini, How do I set your EA?
First of all, I don't advise anyone to trade real money with any EA until you have fully tested it on a demo account and it works to your satisfaction.

Aside from that, the settings and instructions are in the documentation and/or the code.

Beyond that, if you have any specific questions, I will try to answer them as timely as possible.

GirlFlyer
1 trader viewing now