Such a nice post…
Thanks for sharing with us…
This system works best on a fast move but the size of the move is unpredictable and varies with each pair, so cutting losses after a win is the best way to increase profits. There will be some trades were the SL will be hit and the entire group will lose and there isn’t much we could do about that but when 1 wins there is a good chance that it will keep going in the same direction without going back to BE, if we want a bigger cushion we could wait until the second TP is hit to move SL to BE and that will give us 20 pips worth of breathing room like the original SL.
Another possibility is to move the SL 10 pips after each TP is hit that way when the first TP is hit losses are cut in half the second TP brings all remaining trades to BE eliminating all losses after the third TP is hit all trades are winners, I think this would work better than an active trailing stop loss.
Here are the results from this week so far on the 30min TF with no trailing stop, no MM, and 10 trades ,the yen crosses and the euro made a killing, Cable did ok and the AUD/USD took a beating, but overall the batch made a good profit. I broke the results down to pips and marked all trades that should have closed at BE.
Ok, back to work today…
Yes that does look strange. Do you still have the expert log from when that bug occured? It should be located the the [metatrader directory]/experts/logs folder in a file named with the date in question.
It’s a bit difficult to work out what’s going on just from the trade history. If there was any errors when the SLs should have been modified, then they should be in the log.
Thanks.
I attached the logs from this week, other than this problem it’s been working great since you told me how to avoid the duplication issue.
EA logs.zip (13 KB)
I was just looking at the logs and every time it was suppose to change these trades the log has a huge gap, every time it happened, like it knows it screwed up and doesn’t want anyone to know. Maybe I’m reading them wrong but looking at the times the gaps in the log match the trades that should have been closed at BE. Why would there be gabs in the log MT4 and my computer have been on the whole time, It runs slow but I expect that when I’m running 5 pairs. BTW I’m using OandA if that makes a difference.
I just updated my MT4 maybe that will fix the problem.
kingkevbo, I think I know what is happening, the only transactions that are being recorded are the stop orders when they are placed and deleted when the swing_zz changes, it doesn’t log when the stop orders are filled, it only shows up in history, so the EA has no idea when the first TP is hit because it happens on the brokers system since the TP and SL are set when the stops are placed. I don’t know how to fix this problem without having the EA manually place each trade in real time, unless there is a way to use the history file to tell the EA that a TP was hit.
If anyone has any ideas on how to fix this problem please share.
BTW, I still have no idea why it works fine in the strategy tester but fails in forward testing.
Here are the results of my forward testing for this month so far IF the EA was working the way it should. I also attached a spreadsheet showing the trades which were made with the latest version of the EA. I modified the profit to reflect what the results would be if TP was moved to BE after the first TP is hit and I also deleted duplicate trades and the stop orders.
usd/jpy $1597.23
eur/jpy $(2207.53)
aud/usd $641.37
eur/usd $772.99
gbp/usd $1612.72
March results.zip (28.8 KB)
Encouraging results. I’ve had some other work commitments to deal with over the last couple of weeks, so sorry about not updating the EA, i’ve not forgotten about it, and I’m fairly sure I know what’s causing the bugs. I will get around to to it, just not sure how soon.
What settings are you using to get those results? Are they the same across all the currency pairs you tested?
They are all set to 30 min TF, Initial TP 100, TP interval 50, 10 orders, and .72 risk % TP and trail 200, MM and trail false, running on a $100,000 demo account.
I’ve gotten some great results on my 3 year Euro backtests, 2010-2012 with the same settings above except I used 10% risk. Here are the backtest results.
Statement.zip (558 KB)
It does have some big ups and downs but I think it will be a great long term system.
Just wanted to let you know I have been following this thread - except for the last few days (so not really up to date) Cool strategy though! I have tried the EA out some on my demo account - some with nice wins other times some bad losses… Thanks for sharing!! I will continue to follow and let you know how it goes… 
Cheers!
Here are the final system results for last month with an error corrected spreadsheet, same corrections and settings as my last update. Even with eur/jpy losing big it still managed to turn a $4935 profit running on a $100,000 account. The dollar amounts are probably quite a bit lower than they should be due to the errors causing losses which would lower the lot size. These results reflect a 2931 pip gain across the 5 pairs.
With these settings it looks like the eur/usd , usd/jpy and aud/usd are the best pairs.
AUD/USD $1714.55, 449.7 pips
eur/usd $3577.81, 1668.3 pips
gbp/usd $478.99, 110 pips
eur/jpy $(2860.07), -902.9 pips
usd/jpy $2024.42, 1605.8 pips
Final March results.zip (63 KB)
Impressive. It looks like EUR/JPY has been performing poorly consistently. Maybe it would work better if used only on the 6 majors?
I’ve been looking at the code for the EA trying to figure out what it’s doing wrong, but my knowledge of mql4 (and time to look at it) are pretty limited. Have you made any progress on it kingkevbo?
I added US/CAD and US/CHF to my forward testing to see how they do but just looking at the charts it looks like the most viable pairs will be EUR/USD, USD/CAD, GBP/USD, AUD/USD, AND USD/JPY. The lower volatility on the others doesn’t allow the wins to make up for the losses.
Last Tuesday I added USD/CAD and USD/CHF and at that time I changed the % risk settings and added funds to the account making dollar amounts meaningless so I’m just going to post the pip counts. Here are the current results.
AUD/USD 1108.2
EUR/JPY -550.9
EUR/USD -782.8
GBP/USD 1161.2
USD/CAD 43.5
USD/CHF -173.0
USD/JPY -426.6
Total so far for April 379 pips
Intresting, I like basket trading, maybe you can try these 5 - “AUDJPY” “NZDUSD” “EURJPY” “GBPJPY” “GBPUSD” or 8 - “GBPUSD”“EURJPY” “AUDUSD” “NZDJPY” “EURUSD” “GBPJPY” “NZDUSD” “AUDJPY”.
Here are the results for April so far. I can’t wait till the EA is finished, this system could make some decent money over time. I’d rate them best to worst EUR/USD, GBP/USD, AUD/USD, USD/JPY, USD/CAD, USD/CHF, EUR/JPY.
AUD/USD 890.5
EUR/JPY -728.9
EUR/USD -289.4
GBP/USD 1915.8
USD/CAD -235
USD/CHF -683.0
USD/JPY -616.6
Total so far for April 253 pips
nice whorl guys simple strategy, are you all trading manually or using the EA?
I am foward testing using the EA but it’s not working right, I take the results and delete the loses that wouldn’t be there is it was working right. If you want to know how it should work run a backtest, it works perfectly, but in forward trading it won’t set the secondary trades SL to be after the primary hits TP.
If you set it to 10 trades and the SL at 20 after the first TP is hit, it should bring the other 9 trades SL to BE, so you get all the benefit without any risk after the first TP is hit with the drawback of 10 losses if the first TP isn’t hit, but the benefits greatly outweigh the drawbacks. 10 loses= -200 pips 10 wins=+325 pips but in my backtests my 50:1 margin was exceeded and it didn’t open all 10 trades which gave me very interesting results.
[B]Amazing backtest results set to 30 min TF, Initial TP 100, TP interval 50, 10 orders, and 10 risk % SL and trail 200, MM and trail false. [/B] 50:1 margin
Hi All
Just joined and very interesting thread.
I tool the liberty of running some history (tick data) with the EA and have attached the results. The spread is
from an ECN broker so they are very tight. 0.3 to 0.6.
Looks promising, will run some more extensive tests and can start optimizing it on the later versions.
cheers
If you like those results try these settings: TimeFrame=30; OrderOneTakeProfit=100; TakeProfitInterval=50; NumberOrders=10; TrailingOrder=false; TrailingStop=200; StopLoss=200; RiskPercent=10; UseMM=false; I backtested the Eur/USD with these settings, 50:1 leverage and 90% model quality and the results were amazing.