Share your Worse Performing (EA's) & Performance Test Results

Hello All,

The solutions to the most challenging of situations faced by humanity have repeatedly emerged from the least likely of sources.

I have been looking at the possibility of getting EA's (Expert Advisors for Metatrader 4.0 and later) to perform naturally rather than trying get them to perform as I wish or forcing them against their nature.

To this end, I wish to share the results of how I have been able to have an EA do as bad as it can possibly do, then reverse the actions of said trades. I hope that it will be of interest to as many people as possible and in so doing we may just be able to find the solution in a most unconventionalway.

I am not a programmer. I just use EA's and trade manually.
If I may, recall that the best F1 drivers (for example) don't know much if anything about the design of the vehicle in it's entirety, BUT they do know how to get the best out of it on the track (or to do what they will with said machines). This is my approach, as I view programming and EA's (the machines) as a means to an end, little else.

Here goes... lets see who has an EA that can perform worse than I have been able to get from this :

Optimization Results/Settings (Alpari –UK); optimization parameter - Balance:
Back tested Result & Parameters:
PASS-> 128; PROFIT-> 300.98; TOTAL TRADES->25; PROFIT FACTOR-> 1.06; EXPECTED PAYOFF->12.04; DrawDown $-> 4247.25; DrawDown % -> 83.06%; tp=490; sl=500; MaximumRisk=0.57; x1=32; x2=187; x3=53; x4=299; mn=878
Corresponding Forward Test Results:


1. Bars in test 8124
2. Ticks modelled 15243
3. Modelling quality n/a
4. Mismatched charts errors 0
5. Initial deposit 3000.00
6. Total net profit -2528.09
7. Profit Factor : 0.47
Win/Loss = 0.5714
Expectancy: - 0.47
Expected Loss: - 2,217
Performance : - 84.27 %



Gross profit 2229.25
8. Gross loss -4757.34
9. Profit factor 0.47
10. Expected payoff -114.91
11. Absolute drawdown 2718.61
12. Maximal drawdown 4238.78 (93.77%)
13. Relative drawdown 93.77% (4238.78)
14. Total trades 22
15. Short positions (won %) 9 (44.44%)
16. Long positions (won %) 13 (30.77%)
17. Profit trades (% of total) 8 (36.36%)
18. Loss trades (% of total) 14 (63.64%)
a. Largest
19. profit trade 845.14
20. loss trade -1255.88
a. Average
21. profit trade 278.66
22. loss trade -339.81
a. Maximum
23. consecutive wins (profit in money) 2 (375.69)
24. consecutive losses (loss in money) 2 (-2099.97)
a. Maximal
25. consecutive profit (count of wins) 845.14 (1)
26. consecutive loss (count of losses) -2099.97 (2)
a. Average
27. consecutive wins 1
28. consecutive losses 2


Remember! Forum Rule #3: if there is anything commercial that you wish to share, kindly do so via PM or e-mail, it cannot be explicitly shared here, just the rules...

I have attached the grphical representation of this forward test as well as the expert advisor used seperately. If anyone knows how to display the MT4 test graph in a discussion please post the steps. I don't know how to so I have added as PDF.

Looking forward to a spirited and respectful series of interchange and sharing.

Bless

Fear Not! for those that are for us are more than those against. PEACE!

Reverse EA's dont work. They fail for the same reason that profitable ones are so hard to find. Predicting the market is impossible but if you really want a EA that will toast your account just pick a GBP or JPY pair and make the StopLoss too tight. It'll get hit to death!!

A trade with a larger stoploss wont work as well because basically any trade has a 50/50 shot. It can go either way but if you're playing with tighter stop the market noise+spread will make sure you lose more than you win. Spreads actually shift the advantage to the house because a 2 pip spread means your stop gets hit 2 pips early and your takeprofit gets hit 2 pips later plus a reversal screws the numbers up. A trade that closes in reverse actually has a 4 pip disadvantage. -2 pips on the open (reversing this WONT put YOU +2 pips) -2 pips on the (price+spread) stoploss hit. (Its Vegas Baby! - The House Wins Again!!)

Good Luck finding a truly unprofitable EA that can make money in Reverse.

hello Fulltime247 http://cdn.forexfactory.com/images/s...ser_online.gif

Thanks for the input.

Your points are quite logical and much appreciated as part of the process.

Not withstsanding, if you have EA's (based off a solid trading system/strategy of course ) that have stopped being profitable, please share same. The worse the better. Consider it a favour.

I will do the leg work and see if any of your donations can be made to do worse than my posted example.

Secondly, can you share how to post the thumbnail view of the Metatrader perfomance graph? I know how to copy it from metatrader, but I am yet to figure out how to make it part of a forum post as here.

That would help in displaying what we find.

Bless.

I dont think you can in this forum (I know, I know...) - I've been wanting to add my MT4 Stats graph but sigs don't allow html or BB code. If I'm wrong PLEASE someone, anyone correct me.

Hello Fulltime247 :

if you look at this link below you will see exaples of what I am trying to display.


http://www.forexfactory.com/showthread.php?t=102175

Thanks for the sharing.

Bless

Oh, thats just an attachment. If you attach a graphic file it WILL be displayed in the post but you cant attach a html link to a graphic. (Sorry I jumped into paranoid mode a tad fast)

You are brave to be holding a lion that close tou your face, helmet or none

Well I tried. I copied the graphic from the MetaTrader but I have no way of pasting it. I even went and saved it as .JPEG in MS Paint. Do I still then need to use the "Attachment? icon to up load dit? I now if I go that route it will be as a seperate attachement andnot comein the Discussion asyours.

On seconthought couldyou just do a run of the mill systems test in the Metatrader, the copy the graph here in a discussion along with the steps you took from copying to displaying.

Much appreciated.


Bless!

You can always do a [right-click] and save the graphic then upload it BUT it wont be dynamic and update. Also thats NOT me! ( Its a pic I saw on FF but it cracks me up every time I see it)

Try > THIS <

It not one of the worst performing but consistently losing
Nice backtest is posted there with all details!
I am seeking 48hrs/day.... so not working in that direction!
if you find anything let us know

EA attached!

Hello ForexSaint,
I will look into this and let you know what I find.

Oh, I for got, I responded to the PMbefore I say your contribution here.

I also see that you know how to post the raphis as well. could you attach a liitle "How to post MT4 graphics in an FF Forum discussion"?

That would help me very much and make me better able to share results.

Bless!

[right-click] the graphic in the report and save it to your hard drive. When you create a new message or repy just click 'Attachments' and select the saved graphic, click 'Upload' then 'Close' the window. Done. (If your adding to a prev written message click 'Go Advanced' to see the 'Attachments' option again.)

Lets See how directible I am.

Much Thanks to all those who assisted:
TRK, ForexSAint, FullTime 247 and Here ais how it was done:

 

1. Deisred Image to MS Paint
2. Saved to Desktop as .JPEG
3. Open Discussion or reply to Discussion
4. Choose Attachment icon (Near Smiley face in message window Tnx TRK)
5. Locate '.JPEG' File
6. Click Ok
7. Click Up Load
8. Revew post
9. Save Changes/Post Discussion

Viola!

Thankx abundle and I hope you get visited by the LIVE Pips fairy
Oh the joys of a singular accomplishment. Hopefully this will assit others as it did me.

Bless!

ForexSaint,
I am only now getting the full context of your post. When I visited your "TryThis" link, I saw your test results having

profit factor of 0.84
Win Loss Ration of 1047/1663(tot 2710) = 0.6296

Thus this is an excellent start.

I have tried to optimze the EA you attached here (se my PM) but even by trying to vary the 10 parameters indicated, I have gotten no results back.

I I had any luck with optimizing (even using time frames between 1 minute and 60 minutes), I would be aiming for

# a Gross Profit(Win)/Gross Loss(Loss) ratio of 45% (0.45) 0r less?
and
# a Winning trades/Loosing trade ratio of 30%

From your other post I have seen that you fully versed at optimization. I hope your schedule will allow for you to get these parameters. Try doing it on a M1 char over a full week Sunday to Saturday or over two weeks. (most recent full/completed week) and see what happens or ho close you can arrive at the suggested parameter. With that We should be well on our way.

Eagerly awaiting your next post or visit.

In the mean time, do you have any other EA's that require less inut parameters, a working "SlingShot" for example. The ones (slingshots) I have come across I have not bee able to fit for the purpose.

Bless!

You required to reverse an EA trades..right? From the feel of the thread!
why r u then trying to optimize it then?

Hello ForexSaint,
as Always the contribution is most welcome.

We optimize the settings to get a desired quality from the Expert advisor. From the entire distribution of settings, we can divide them in to three sections if you consider the normal distribution of possible settings. Those in the middle they will merely oscillate, win today and loose tomorrow, with an alarming regularity. to this extent such settings for a properly designed, EA that is based on a sound system, for simplicity sake, MA crosses, Bollinger Breakout systems with MA, Ema; Slingshots or some other technique viz neural networks, with the regular requirements, of Stop Losses, Take Profit Money Management Risk Rewards etc. So we have a reasonably designed system.

These settings in the middle of the bell curve will win with #Wins/# losses of around 50% but, typically fall in the range of 40% to 60% winners. These will perform cyclically.

then you have those that will Either win or Lose 70% of the time.

The thought I have is based on the latter group, if we can isolate the settings for a reasonably designed EA (replicating the actions of a modest manual trader such as myself) it is easier to Isolate the Loosing settings than the winning ones.

To get the winning EA's you have to have excellent programming skills a as well as a grasp of the fundamentals of ForEx trading. That (IMHO)is why the Consistently Profitable EA's are rarely shared or appear freely in public domain.

With an EA and settings that Have been shown (via optimization, to be losers from the onset) then provided said EA settings meet sepcific criteria such as Profit Factor (Gross Win/Gross Loss) & #Wins/#losses of about 30% then such settings in my humble opinion would be for an EA that can be successfully reversed to produce a decent return and net positive Net profit and Available Free margin to make even the most Greedy Trader thank Than you.

So those are my reasoning. Since your contributed EA was at

Profit Factor (Grossprofit/GrossLoss) = 0.84 (target 0.333)
Win Loss Ratio of 1047/1663(tot 2710) = 0.6296 (Target 0.3 or less)
If we can get these key values to be halved (individually) then we would be well on our way (IMHO).

Form your post ForExSaint you have more than the capability and willingness to contribute such settings and since I have not been able to get the MT4 to "optimize" the settings for Level_trading_123 EA.mq4 , I pray you would.

Your suggestion on the default settings was tried. the MT4 strategy tester started, but stopped shortly thereafter, even after waiting 2 + hours in faith.

The one i posted, works a lot easier and recall that I know zilch about programming, just manual trading and an appreciation of an EA that has a reasonable profit potential.

So ForExSait, I hope this request to share the aforementioned worse settings will challenge you (And other FF members) sufficiently. With such, it will be worth the effort, deal, you produce the settings requested and we'll see that we were both closer to reverse signaling that we initially thought.

I implore you, please give it a try, if not then the joke will be on me (in a manner of writing).

Once again, I thank you for the interests and I pray this will soon spark even more spirited participation from the greater FF family, even programmers and then we can take it where it leads.


All I have now are the ideas, but the solution lies in the combined consciousness of this collective

Bless

@pathumracled

Sorry ..i am not able to understand clearly..
Do you want me to Optimize it and then share the bad settings with you?
So that you can reach below that target of 0.30 in profit factor and win loss ratio?
Is it?

If so...then just tell me
for how long period should i optimize it?
on which pair?

AMEN

Path you dont need to open MSPaint (thats why your graphic has the extra whitespace) - Just [right-click] on the graphic IN THE REPORT and save it then upload it.

Fulltime247

Thank you kindly. When I get the next set of worse settings, I will use that tip.

On another subject Fulltime 247, do you have any EA's and or Worse settings that you can contibute to the project? It would be nice to have you on board.

Bless!