Gann SQ9 EA

[AG2012]

SERIOUS WARNING

Most Forex traders lose all their money.
Using the robot posted here in trading Forex does not guarantee success.
Trading this robot could lead to serious financial loss.
Trading this robot without understanding its underlying trading strategies guarantees traders will lose their money.
This is not a set-and-forget ea; there is no such thing and anyone who tries to claim there is, is either stupid or lying. This ea requires frequent manual intervention.
At best, a trading robot is only 90% as good as the manual strategy it trades. At best. At worst, it can be much less effective. If the strategy is rubbish, so is the robot.
To trade this robot, you have to understand:

How to use EA's.
The RSI indicator
The Gann SQ 9 Indicator
The old "Jor" ea



Hello Steve and the rest of my friends.
Recently while at another forum I had stumbled onto a thread started by a gent named "Sachmo", requesting if anyone could code an EA for his Gann Square of 9 indicator (which he very graciously shared). After noticing that there was no response after a time, I accepted the challenge.

The Indi itself is an interesting little beast. It changes very frequently sometimes following the price.
Initially "Sachmo" indicated to me that he was using this thing on the M1 charts. It does seem to work there, but I believe that it's much more accurate on H1 or higher time frames IMO.

I've tried various other entry confirmation "add-on's"; ADX, BB's, ATR, 2 MA/5 MA crosses, and it all seemed like trying to split hairs IMO.

So one day I decided to put up a Stoch on my (H1) chart with these settings: K:50, D:2, Slowing:9.
It does seem to line up nicely with the Gann indi most of the time. This now is the trade strategy I've adopted for the system.

Please bear in mind that I consider my MQL abilities similar to a chimpanzee up in a space capsule wondering what would happen if he were to press the big "shiny red button" sometimes, and that's why I use my old EA generator software to help create my EA's.

Entries are as follows:
For a "buy": Stoch below 20, triggered by Gann "zero level" < Gann "zero level" 1 candle ago.
For a "sell" : Stoch above 80, triggered by Gann "zero level" > Gann "zero level" 1 candle ago.
(Yes I've tried other "candles ago" variations but this seems to work out the best for me).

The latest version I'm putting up is SQ9 EA v 4.2 CAFA. "CAFA" stands for Close order At Fixed Amount (in dollars).
Due to the way that the software renders the EA, there are some duplicate fields IE:
BuyLots, SellLots, etc. Note these two fields in the EA settings, Amount34 and Amount37.
This is the "CAFA" amounts for buy and sell.

I have been running this live on Alpari US and HotForex (E/J H1 and recently U/C H1) for about 2 weeks now, and it does seem to be profitable. Although the Indi can and does get "faked out" from time to time. I'm keeping a VERY close eye on it and, I'm setting the Trailing Stop to 20 and the Trailing Gap to 15 in an effort to at least "lock in" S/L at entry price early to avoid a larger loss in case of reversal.

Update: 04/10/12:
I have temporarily stopped testing the EA. It seems that there still are a lot of false entries IMO.
I will be addressing this and will post a new version here when ready. Thanks for your patience.

Update 04/12/12:
New version SQ9 EA v. 4.3 CAFA is ready. I've gone back to using a 2 MA, and a 14 RSI for "steering".
I've run a short 1 month B/T on E/J H1. Results are posted below along with the EA in MQ4 format.
Try it with "ootb" settings and see how it runs. Enjoy!

StrategyTester_v4.3EURJPY.gif

Update 04/14/12:
Added yet another version: SQ9 EA v. 4.3 CAFA RSI. It's almost exactly like the 4.3 but I've added "CloseBuyRSILevel" and "CloseSellRSILevel" input fields, as a sort of "stealth" S/L in case of reversal at the extremes.

Update 04/15/12:
Version 4.3.1 CAFA RSI now has the RSI periods selectable. Enjoy.

Please B/T first, although it seems that v. 4.2 may be the way to go, according to the feedback I've gotten so far. Although that version seems to be very vulnerable to reversals. We'll just have to keep testing and see.

Update 05/11/12:
Version 4.4 RSI (Upgraded)

Hi friends. I've taken bits from the 4.2 and the 4.3 and combined them to make what I think is a solid EA now. This is a monster of an EA. It's actually Steve's old "Jor" Bot modified to run with the new Gann_SQ9_A6 indicator (that you'll need to load into your Empty4), that I had to modify as well. If someone can help to increase it's performance, by all means please do so and share.

I've left the original Jor "factory items" in place in case you wanted to use the bot as "Jor", and added some "extras" that are probably useless, but left them in just for grins.

I STRONGLY suggest that you leave all the "default" settings as they are, except maybe for lot size. The stop loss and take profit inputs are actually way down in the settings at the bottom...
They are the "Hidden" S/L and T/P settings.

Please use those instead of the the inputs at the very top, for I quote Steve's remarks inside the code; "Only included as an options. Users should not set this as it overrides A's calculations".


You will still get some DD, sorry that's unavoidable. So as always back test it first, to get a feel for the EA on your preferred pairs. I'll be working on a "Sachmo" version next... Enjoy.
Here's some background on the Jor EA : http://www.forexfactory.com/showthread.php?t=313656

StrategyTester_v4.4_3MoUSDCAD.gif

Update 05/14/12:
Version 4.4 SE

This version uses the original Gann SQ9 A2 indi. It seems to give better performance when the "close on opposite signal" is enabled. Now you won't have to wait out a trade for (sometimes weeks) to play out with big DD, although you will get an occasional loss from time to time.

StrategyTester_v4.4SE_3MoEURUSD.gif

Cheers
AG

[AG2012]

SQ9_4.1_Cafa_D_USCAD.PNG

In the screenshot above, the actual EA that placed that order on U/C last week was my last version (4.1 CAFA-"D"). The "D" meaning Directional. The "CAFA" set to $30 did it's job.
But there were a couple of drawbacks to that version.

1. You cannot back test that version because of a "check last order type" filter that I had included in the EA to satisfy a couple of people that asked for that function. For example the EA checks your order history (for the pair it's put onto) to see what was the last closed order type (a buy or a sell).
There does not seem to be an order history available for the visual B/T in the Meta Traitor, so the EA won't place any trades.

2. If the last order type was a sell for example, then the EA would wait for a BUY entry setup for it's next order.
After seeing 2 "buy" setups appear one after the other, I decided that this version was leaving opportunity on the table.

Back to the screenshot above, the way I think this is going to play out, is that when the Stoch goes above 80, then the price should move up past the (now) Gann "zero level" of 0.99843.
When that happens, the current "zero level" should disappear. When a NEW "zero level" reappears, is when the EA will trigger a sell order in this case.

I'll just have to wait and see now.

[jb68]

And in zip there is an exe.

- No back-testing .. hmmm this is weird. You can easy see what was the last order type by using a variable.
How about some forward testing statements?

Why don't you post the MQL version, there are people here qualified enough to change and back-test it, if there is something promising then we can have a really good EA. Till then probably nobody will run an exe.

[AG2012]

Hi jb68.
Thanks for posting. I've uploaded the MQ4 version on the first post. Sorry, I must seem a little mysterious. I think I must be a little "road weary".
Here's a link for those who want some more background on this EA.
http://www.worldwide-invest.org/threads/9052-EA-for-Gann_SQ9_A
I'll be putting up some back test results if you want me to. In the mean time you can view my live stats using the links above on post #1.
Here's the 4.1 "directional" version for you if you care to back test it.

SQ9 EA v 4.1 CAFA-D.mq4

[AG2012]

Well guys.
Since everyone is so paranoid, then I guess I'll have to ask Steve to delete this thread, since it's probably a waste of time anyway.

[AG2012]

P.S.

I've just uploaded the Indi itself on first post for those of you who like to do extra work (and don't trust .exe's).
For crying out loud, I'm trying to build my reputation - NOT destroy it...
I was also going to upload a version that I'm testing now, but like I said, it's probably a waste of time.

[pmacel]

I have put this on demo 8 charts. EURUSD, USDCAD, EURJPY, USDCHF, GBPUSD etc
So far one closed order EURJPY nice profit. 5 more open orders. We will see how it goes. Please keep working on this.

[AG2012]

Thank you for the kind words pmacel.
Guys, please forgive the "mini rant" earlier. I can't blame you for being apprehensive about running an exe file. I have been following Steve and the others at FF for a while, and have the utmost respect for them. I would not willingly post something harmful here. I just felt that my goodwill was "frowned upon". It's been a long year for me, can't you tell?

Well with that out of the way, let's get down to it shall we?
I have temporarily stopped testing the EA.
After testing and testing and re-testing, time and time again, I find that there are many false entries caused by the Indicator. And adding the other supporting confirmation " Indi add-on's" probably limit trade opportunities even further.

Don't get me wrong, I truly believe that this Indi/EA combo has great potential. I have reverted back to an earlier version EA that just runs off the Indi, but is prone to the False entries / orphaned orders.

I think that there's a repaint issue here. I also think that the EA would run great using just the Gann Indi, if we could just get around the repaint issue.
I'm not looking to re-engineer the Indi either.

I've noticed that to get the most out of this thing, the best entry is on the first candle formed when the "zero line" or "base line" of the Indi stays on the chart.

I've just had an idea - what if (as a confirmation) we could make the EA wait x number of minutes or whatever, to confirm that the "zero line" or "trigger" is still present before opening the order?
In actuality, I've seen the price reverse after a few candles, go against me and through the "zero line", causing it to disappear again (maybe that's where the S/L comes in handy LOL?).

The way I have it coded now is: (for buy) If Gann Indi (current candle) is lower than Gann Indi 1 candle ago, open buy order. (for sell) If Gann Indi (current candle) is higher than Gann Indi 1 candle ago, open sell. Any suggestions I guess I'll continue testing....

[AG2012]

Now hopefully, version 4.3 will run as nicely on live forward test as it did on my B/T.
Version 4.3 is on the first post of the thread.

[jb68]

Hi,
I did look at your indicator and is based on zigzag. Unfortunately it is not possible to make a zigzag indicator without repainting. You can play with it on 1 minute and will be obvious. Zigzags works decent on ranging markets.

Usually when yo get the first indication on a triangle coincide with the stochastic or williams. The diff is when zigzag will repaint stochastic willl just stay overbought.

The only thing you can do to increase your probability is to apply signalling theory and use a Fourier approx like to take a trade when more stochastic will coincide, preferably for periods of n, 2n of 4n.

Another thing is to take more then 1 trade (martingale) so for example if indicator will repaint and price go against you with another n pips to take another trade. I would add a filter and check what are the best hours.

What you say about waiting for a candle to go higher/lower was already done on zigandzag indicator. You can take a look.
http://codebase.mql4.com/3702