Great EA in backtest! - page 99
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
In answer to Aragon. Is it desirable? Yes it is if you consider if we were able to reverse decision during 8,9,10,12 we would have picked up $2k over the last week. In 3 days Ea performed 40 trades. of which 27 were losing trades and 23 were in profit. That equates to 67% loss rate. Equal to the win rate trading during hours allowed. Difference being we climb to $2k profit from 500 a whole lot quicker when trade reversed during these 4 hrs.
I gave tried by adding this...
if(Decision == DECISION_SELL)
{
Decision = DECISION_BUY ;
Print("Solution - to sell: reversed based on trade hours to BUY: ", DecisionValue);
// BlockSell = false;
BlockBuy = false;
return(0);
}
if(Decision == DECISION_BUY)
{
Decision = DECISION_SELL;
Print("Solution - to buy: reversed based on trade hours to SELL: ", DecisionValue);
BlockSell = false;
// BlockBuy = false;
return(0);
}
//////////////////end of reversing decision during specific hours test//////////
It didn't appear to make any difference in a backtest at all. Same result with the code turned on as turned off. I'm just a novice coder. For some reason I don't understand this code printed the line it was told to but it didn't open any trades from the reversed decision. I don't know why not. The attached EA has the -rh on it meaning 'reverse hours'.
Like everyone else today I won a couple trades and lost one which took all the gains of those two wins away. Its some big news event tomorrow. I'm scaling back my risk may even turn it off early tomorrow...let the news blow over. I don't know what to make of that gap today. Maybe someone who is better with code can fix what I started.
Ok I have a couple thoughts here...
1- Watching this EA work I sometimes wonder about it's closing logic and how it could be improved. What can it look at to assist it in saving profits.
Case in point: Last night while I slept it took and won two positions totaling about $9.50. I was happy to see two consecutive win at the risk level I had set. I noticed that compared to the backtest at this level the win amounts were a little small. In the tester wins were anywhere from $5 to $8 on average and sometimes more. The losers were ranging from $11 to $14.50 on average. Two average wins is more than one average loss that way. Life is good.
So l get up to see what is the situation this morning...two wins, good good, and it's also into a third position which it is down $5.40. Looking at the stop loss it could lose about $16 if it goes all the way to stop out.
So now I'm faced with the decision and the question is? Do I allow the system to run on it's own without intervention and trust it's win/loss ratios being true to the system's logic or do I manually exit the trade?
I watched the price move in favor of the position all the way to break even on the third trade. I thought, well this bodes well. it looks like a gradual trend in favor of the position and maybe it won't t/p right now but give it time and it will. I might have to endure a couple more retracements but eventually it will t/p and the wins will be 3 an 0.
So I decided to let it play out on it's own with the EA's logic watching over the closing.
I returned a hour later to see that the pair had gapped up just enough to trip the s/l and sure enough it took $16 drawdown.
Now I wonder how this sort of scenario could be avoided.
2- This brings up a whole issue about trading discipline and either keeping hands off or going hands on. I had the thought that the third position was untennable and it looked like about a 50/50 chance to me of it going either way. I don't consider 50/50 really good odds.
On the other hand if I don't trust the system I'm trading then I've got no business trading it either. So where do I find the balance in these two? I realize I probably can't win every trade but I still want to.
Lastly I observe that the backtester gives me a large panoramic picture of the performance whereas watching each trade at the rate of only a couple trades a day, 2 or 3...well it's hard to see with such a microcosom view day by day how they fit into the larger picture which looks so good on the backtester.
Can I conclude that I just got unlucky in that the wins were two small ones right up against one large loss (relatively speaking)? Or do I conclude that the broker is messing with the data feed and manipulating it to work against the system intentionally? I don't know what evidence suggests or really how to interpret the results.
For now I will let it keep trading and keep tabs on it. I wonder what discipline this is developing? I don't like not knowing what to do that is within reason to do and the fact that the evidence is gained so slowly. I'm impatient for answers and I don't have enough information to really say what's going on.
The only thing that kept me from turning it off is that Dave and other's have posted information saying that they are seeing that the forward tests are indeed following the backtest patterns if allowed to run long enough. Patience was never my forte.
I'll give it another day.
Hi Aragorn, your more persistant than me on CT thats for sure, I gave up on this a bit, had to get back to real trading, I already lost about $500 live on this system, it would win really good and then lose it all.. over and over and kept creeping down instead of up, would have been cheaper to buy the Pro version and save the headache... Cant deny this EA has some ingenious logic in it and its somewhat addicting to try and make it work... but I was thinking that because its so ingenious maybe the developer meant for it that way so you'd be lured into getting the Pro version and stop hassling for months wondering if it will ever really pan out. I'd buy the pro version in a heart beat but for the life on me I can't get one person to say they have it and its working or not...
You mention disclipline, I dont think you should have to wait months to see small profits, if thats the case then this EA is not performing like we thought it might. Look at USDJPY, its been mellow and easy with reasonable trend, CT should have been able to profit very well from this over the last 3-4 weeks if it could run as backtested, but it has given everything back even when excluding the volitile hours....
I thought Daves last comment was that CT gave back his profits so he went back to his old thread..
sooo I continue trading as usual and playing with EA's on the side and the one of interest thats doing very well right now is Phoenix2007 also Sashken, these 2 are in the top on the Championship contest and my live instances are making exact trades within 1-2 pips.
I thought Daves last comment was that CT gave back his profits so he went back to his old thread..
.No you missunderstood. I have small profits from it. (small)
I stopped posting settings and test results because I found whats working for me. It is still in the 70+% win rate.
But I see the same problem that argorn is talking about. I see the profits are getting pulled out befor the usual 8 pips the back test show. I don't know if its cause the ea thinks the price move is slowing or what, but I have seen it take profits out at 5 pips and then wach the price continue on with out me.
You seem pretty good at adding things to the EA Argorn. Can you try to get it to be able to set a hard tp that it sticks to?
On the other hand I saw it take 12 pips profit 1 time. HUH?
Mabe its market conditions or something. I dunno.
Started this on a IBFX demo on sunday night.
Here is my statement so far this week.
Using daves settings...
Jack
I was playing with 1.88 to see if could get it to honor the min pip setting. I even changed code to state best pip profit = 15 It didn't seem to work right.. And when playing with that pip timer thingy, the results didn't change no matter what setting I used. (30,60,120,240)
But when turning pip timer to false I did make a few more pips than with it on. runing over 1 month it seemed to take the same trades though.
But thats just with my tests
Also in all OS versions. In the settings window it has profit setting option but I found nothing in the code where it refered to this extern setting.
I just wonder how much might be left out thats in the pro version.
EDIT: DudeWorks....yes it is addicting....as you can see I cann't get away for testing even though I got satisfying results. (so far)
testing CT is like a drug, I cann't get enough.
Hi Aragorn, your more persistant than me on CT thats for sure, I gave up on this a bit, had to get back to real trading, I already lost about $500 live on this system, it would win really good and then lose it all.. over and over and kept creeping down instead of up, would have been cheaper to buy the Pro version and save the headache... Cant deny this EA has some ingenious logic in it and its somewhat addicting to try and make it work... but I was thinking that because its so ingenious maybe the developer meant for it that way so you'd be lured into getting the Pro version and stop hassling for months wondering if it will ever really pan out. I'd buy the pro version in a heart beat but for the life on me I can't get one person to say they have it and its working or not...
You mention disclipline, I dont think you should have to wait months to see small profits, if thats the case then this EA is not performing like we thought it might. Look at USDJPY, its been mellow and easy with reasonable trend, CT should have been able to profit very well from this over the last 3-4 weeks if it could run as backtested, but it has given everything back even when excluding the volitile hours....
I thought Daves last comment was that CT gave back his profits so he went back to his old thread..
sooo I continue trading as usual and playing with EA's on the side and the one of interest thats doing very well right now is Phoenix2007 also Sashken, these 2 are in the top on the Championship contest and my live instances are making exact trades within 1-2 pips.By all means please share the links to these other EA's or where I can download and test them. The only reason I'm hanging on here is that I don't know anything better.
Dave has some slightly different settings than i do I believe. I for example have used the reverse index=8 to generate more trades and then used my new filters to sort out the noise. I don't know that what I'm doing works any better/worse than his method of running the reverse index=3.82 I kinow I hate to see it go a whole day without taking a single position.
Odd thing this morning...I woke to find that it had won two positions, one on the EURUSD and the other on the USDCAD. when I looked at the experts tab to see where it was in it's decision value it was saying that it's solution to sell was -.0016! yet it wasn't executing an order. I found this especially wierd since my DV filter allows anything to execute over and absolute value of .0009 here it was having just won a trade and with a solution to sell of .0016 and still not executing it's next position!!!!
So lemming that I am I followed it and opened an order manually. I took a 1 lot position but then immediately regretted it. I had just made $6 and the next bar the solution to sell went to .0001 in the expert tab. I decided to take my spread loss and exit rather than let it take everything back. I was still stinging from the loss the previous day. A few minutes later I see that it would have given me the 5 pips I was asking for. But hindsight is always perfect eh?
Anyway I'm still mystified by how this EA thinks. It obviously does take and win trades. I just thought it did so based on it's solution Decision Values. Apparently there's more to it than that or it would have entered the position long before it was when I saw it. I wonder what kept it out of the position when it's DV was so high?
Another thing I find interesting is that my tests have shown it does better on the EURUSD now than on the jpy which I focused alot of testing on earlier...that spread on the gbpjpy is insane. If I can get this to perform on the eurusd instead with a spread of 2 most of the time I'm thinkin that is gonna reduce the risk too.
thanks for the encouragements, persistence I can manage I just hope I'm not being stubborn.
No you missunderstood. I have small profits from it. (small)
I stopped posting settings and test results because I found whats working for me. It is still in the 70+% win rate.
But I see the same problem that argorn is talking about. I see the profits are getting pulled out befor the usual 8 pips the back test show. I don't know if its cause the ea thinks the price move is slowing or what, but I have seen it take profits out at 5 pips and then wach the price continue on with out me.
You seem pretty good at adding things to the EA Argorn. Can you try to get it to be able to set a hard tp that it sticks to?
On the other hand I saw it take 12 pips profit 1 time. HUH?
Mabe its market conditions or something. I dunno.I seriously have considered this. I'll look into it if by no means I get distracted . Distractions abound in my world...lol I don't get much time to focus on programming really, I have to try to fit it in on the side. In my back tests I discovered that t/p on one trade all the way to 98 pips once. Actually here is the data series of t/p's that I pulled out of the results....this was from August 1 to current on the EURUSD, these are just the t/ps of the trades it made...
if we were to hardcode a t/p into it where would we set it?
0.0076
0.0047
0.0047
0.0025
0.0025
0.0025
0.0024
0.0021
0.0021
0.002
0.0019
0.0019
0.0019
0.0018
0.0018
0.0018
0.0017
0.0017
0.0017
0.0017
0.0017
0.0016
0.0016
0.0016
0.0016
0.0016
0.0015
0.0015
0.0015
0.0015
0.0015
0.0015
0.0014
0.0014
0.0014
0.0014
0.0014
0.0014
0.0014
0.0014
0.0014
0.0014
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0012
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.0011
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0008
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0007
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0006
0.0005
0.0005
0.0005
0.0005
another curiosity, notice that it only did t/p 5 pips 4 times in two months in the backtest yet in my live forward test it did it once already and the other trade was only 6??? something is still fishy here.