Cornflower EA

thanks

Thanks for the advice.... also thanks for creating this EA. I am testing a few and so far the CornFlower is my favorite one.

Tony

Hi Ari... maybe this was also stupid answer. here...lol, if you use 'new' demo account for backtesting.. are you already download your tested pairs data from history center ? at least for tf 1m of any pair that you want to use it..in strategy tester.. best regards...

A reminder from your friendly system author: I never intended for Cornflower to be used on 5M charts. In fact, I warned early on that it is NOT to be used on such a short timeframe. I don't mind the effort to test it on the 5M, but I can tell you right now it's not going to work as well as the hourly does. I have other systems for shorter timeframes that I documented in the cornflower thread.

That said, I commend the efforts here. I don't intend to test this EA, but I encourage others to demo it in an effort to devise modifications that might allow Cornflower to be autotraded.

Thanks Lever for the support.

I take this shot to reccommend anyone trying my EA, not to use it live, its just too simple right now for taking risks on the real market. Hopefully with time and feedback i may make this EA good enough for the real world.

The market giveth ... and the market taketh away.

The last time I posted my results, I had posted a profit of +1080 pips of a combined 10 currencies I have been trading. Last week's trading results have been a complete opposite of the previous week (and then some!). Here are last week's results:

Pip Total = -1610
Won = 27
Loss = 103
Win Percentage = 20.77%

As you can see, not a good day at the office last week. This is why we need to continue demo testing different ways to trade the EA and not jump in by trading live. Lucky for me I'm trading demo and not for real yet.

As I mentioned before, I think the best way to use this EA - or most any EA - is to make the best estimate of the direction of the trend and set the EA to only trade in that direction. So far in my demo testing I have it set to take both "Long" and "Short" signals. I think it would also be a good idea to pick just a few currencies to trade and to focus only those, rather than a large number of currencies. As you can see, I traded a fairly large number of currency pairs. Yes it has the potential to make a lot of pips, but also has the potential to lose a lot of pips. However, I believe the reason I lost so many pips last week has more to do with my EA setting of taking all trades, rather than just trading with the trend as I should have done.

This week I will focus on trading with the trend. The way I'm going to do this is by using the MACD (settings of 5, 13, 1). When I see both the Daily and 4 hr macd agreeing, then I will set the EA to trade in that direction. I will periodically check the charts 2 or 3 times a day to see if conditions have changed to cause me to either: A) disable a currency if the 4hr and Daily MACD's no longer agree and wait for the next opportunity, or B) enable a currency to begin trading if the trading conditions meet my requirements (that of agreeing MACD's). I am still using the same SL and TP that I have used before. I am also still using 10 currencies. However, I only enable the a currency to trade when my requirements are met. I then disable a currency when it is not in agreement with my rules, and then wait for the conditions to be right again. So it won't be very likely that all 10 will be constantly in a trade as was the case before.

I know my MACD method isn't the best way to determine a trend, but it is something I am experimenting with. My goal is to find a way to make it as simple as possible (in regards to finding the simpliest way to find a trend by just using the MACD's).

I wholeheartedly suggest everyone to try out different settings of the EA. Try different SL's, different TP's, different MA Triggers, etc. Don't be discouraged by my bad results last week. I've only been testing and demo trading this EA for 2 weeks now.

Remember the words of the great John Paul Jones: "I have just begun to test!" (or something like that. lol.)


Topgun

Lever, you are correct. The 5 min timeframe is not the best timeframe to use with the Cornflower system. I am no longer testing the 5 min charts. I believe the 1 hr and 4 hr timeframes are the best timeframes for this system.

I just want to say that the Conflower System itself works. Whether it can be successfully traded using an EA is remained to be seen. That's why this thread was created.

Lever, thanks for visiting this CF EA thread. Having the actual Cornflower creator post here is like being visited by royalty!


Topgun

Topgun, While I agree that price is very important I also have found that the element of time is equally important. Could you share if last week you were running Cornflower 100% hands free or were you switching it off and on throughout the week? This might help us to nail down best times. Thank you for your time. Fritz

Hi Fritz,

Yes, last week I ran the Conflower EA around the clock ... 24 hrs a day from the time the market opened on Sunday to the closing at the end of the week on Friday. However, there was one day during that week in which I had to restart my computer, and then afterwards I somehow forgotten to get my charts logged back on. It wasn't until 3-5 hrs later that I realize my charts were not on. So I ended up missing between 3-5 hours of trading possibilities. But other than that, I had been running the EA "hands free" as you say for that entire week.

I still have the EA running 24 hrs a day. It's just that now I am checking in every 4 hrs (except when I'm sleeping) to see if any potential trades are meeting my new requirements in which I talked about 3 posts up.

I agree that the time of day that you trade could affect the results positively or negatively. Also just as important (if not more important) are the settings of the EA. That is why everyone needs to test different settings to see what works best.


Topgun

Postrock, You have asked for ideas, here is a simple one that may improve performance and limit losses.

1. Each time a signal is given have a user option to straddle the bar after it closes. Buystop 10 pips above the high and a Sellstop 10 pips below the low of the bar.

2. User would need a "pipsaway" setting so that the buystop/sellstop pending order could be adjusted pipswise.

3. The intent of this straddle is not to hedge the pair but instead I have noticed after examing charts for the Cornflower EA that often a trade goes in the opposite direction of the signal so this simple straddle will catch the price as it heads that way.

4. In order to stop both a buy and a sell to be put on at the same time there should be a user option to cancel the unopened pending order when the other pending order gets opened.

5. The NFA has now instituted a no hedging rule so that is why #4 would be necessary unless of course you are trading two accounts.

Thank you for your time and consideration for this modification. Fritz

Hi fritz..you have a couple good points. I'm currently coding some of what u say.
Hopefully i'll post v0.2 later.

edit: sorry i cant finish v0.2 today, i didnt consider that having those new functions would make the EA not work on multiple charts..stupid me..i have to rethink some of the new logic.
No big deal, but not for today.

Hi everyone,

There is a difficulty for me to modify the ea as you (or I) wish. On
Ibfx, the EA is black/white color which means we cannot make any changes.

Is there something I can do to make changes???

thanks

trademeister

Hello TopGun,

I've just arrived here, I put CF on the EuroUsd, strat tester, I am getting an ordersend error 130..

Any ideas?

Tks.

Hi Postrock, As soon as you are finished with the above modification I am ready for testing and additional chart analysis with the new mod. ....no rush. Thank you for your time. Fritz

Hi guys,

Glad that you have made an EA for the wonderful cornflower method.
Is anyone still there???? Hello? Please come back.

trademeister

sorry everybody for my disappearance.

I'm hard at work on a EA for myself..wich is quite complex for me.

I havent chacked wich error is the 130 but i guess it has to do with order opening, since this EA hedges and now hedging isnt allowed any longer. I have to change it but i'm so imvolved in my own project that it takes my coding practise time all away.

Pls have patience. I'll try to give some time to the Cornflower EA as soon as possible.

Postrock,

Hope your EA works out for you, not that I am keen to use the cornflower ea I have still been thinking about it...

Maybe if you take a bit longer I might come up with something add!

I have followed the tread. Did you ever get Cornflower V.2 up and going? As I understand the thread, V.1 will not run at all, because of hedging?

I hope there is progress in the works.

Murph

Hello to all. Newbie here. I hope this thread is still alive. I have downloaded the ea and already testing it on a demo account and I will post it at the end of the week. Hope it will once again stir up people's interest again.

As promised, this is my test result. Fyi, i am not trading on Friday because it is bank holiday, plus NFP is due. Crazy spike here and there, so no point.

First, lets start with the setting.
Broker: FX Open
Timeframe: 1 hour
Trading time: Asia and UK session only.
# of currencies: 11
Bar: 0
MA Trigger: 7
TP & SL: 30 & 50
Trading Direction: Long & Short

Total pips: 830
Total trade: 57
Win: 46 (80.7%)
Lose: 11 (19.3%)

Below are pips accumulated according to currencies:
EURUSD: -90
USDCHF: -40
GBPUSD: 240
USDJPY: 120
EURJPY: 160
GBPJPY: 110
CADJPY: 150
AUDJPY: 70
AUDUSD: 30
USDCAD: 120
AUDCAD: -40

And that is all for this week, not bad if I do say so myself. Next week, will conduct another testing with another setting for MA trigger.