Try My EA! Need Group Feedback...

Guess the brokers are gun shy now, lol. Like maybe the SNB will say, "Fooled you, HA HA HA HA HA! We're really not stopping our peg!"

http://www.kitco.com/images/live/nyg...07787682455684

From Finfx website...

Due to the Thursday’s dramatic move on the Swiss franc by the Swiss National Bank, a number of brokers have announced that they “can no longer meet regulatory minimum capitalization requirements because of significant losses”. A considerable number of brokers went bankrupt and closed down.
FinFX adheres to strict risk management policies and has an extensive range of internal policies and procedures to ensure high standards of business processes; thus we were able to overcome unprecedented volatility. FinFX would like to inform our clients, that our business was not affected substantially. FinFX will continue to operate normally.

Good, because a lot of other brokers are going down the tubes, including FXCM, Alpari, Fxpro, IG markets, Excel NZ...etc. They did stop trading on CHF pairs for awhile, but are allowing it again now.

Not quite $1280 yet, but top of the channel it is. Didn't even bother retesting anything below $1255.

Both test accounts looking very happy.

...and now we know why... sorry dambuster

Dang, man. That's awful! So sorry.

No big deal, its only a demo, the real worry is my 140k that is with them...

What evidence do you have of FXPro going insolvent?

They issued a statement saying they werent hugely affected and have adequate capital in place.

@dambuster

I know what you mean, I have a live account with ATC, which is an IB for FXCM. Had me worried on Friday, before they got a bailout loan from Leucadia National. I guess you have a level of protection there in the UK with segregated accounts, but in the US the accounts are mingled with the broker's funds by law, and no protection is offered. The law for "our protection" from Congress, which of course does the opposite. Moral: Don't leave much at the brokerage here in the states. If you make money, take it out quick! Some multi-service brokers (like IB) will segregate any money not used in margin at the end of the day into other protected accounts , but still......

Soooo, I can't make any sense so far of the ideas I've had with the grid beyond the $10 grid EA I already put out there - new versions are not helping, only hurting so far in backtesting. Still got buggy code trying to track and selectively take out hedges with profits, lots of problems sucking up all my time lol.

That being said, I did go ahead and get v1.2 ready for the $10 grid, and fixed a couple of issues Kosmo pointed out.
- it now draws vertical lines reliably at grid initiation, and also there's an added color for a grid termination line, should it end and sit dormant before reinitiation.
- it has a micro-compounding flag if you want that. Note that if you use a fixed lot size no compounding is done and the trade amount variable is ignored.
- it tries to close the pending trades from the inside of the grid to the outside. (I haven't seen this work yet - it might still need some tweaking).
- New feature - dynamic grid: With this set to true, the initial grid built is 10 steps out from the $10 line initiiation. If the pending trades are filled so as to push the BE lines beyond the grid it will automatically fill in to the BE line on either side. If this is set to true, NoOfSteps is ignored.

If you are new here, a more complete explanation of the inputs are at post #727 and #778.

A word of caution: Grids have the potential to blow out accounts if left untended. Removing profit regularly is a good idea. The dynamic grid is guaranteed to consume all the money in the account in certain market conditions. Keep an eye on things and don't let it get out of hand! Having a fixed grid will give you time to consider whether to continue should the box become completely hedged, whereas the dynamic grid will run until it stops due to no equity or hits a target. Also, it will put a ton of pending orders on the screen, especially with low GridStep, and the more you have, the more time they take to close when the EA recycles. The BE and Target lines are counting on the pending trades to be there to be accurate, so they're necessary, but just be careful.

I will probably just work using this version as a base, because you can easily make the $10 grid EA to behave like the v2.11 BET EA - just set your sensitivity window wide open (like >500 with the PriceGrid at 10) and it will completely encompass the space between the $10 lines, making the two EA's indistinguishable. There is a grid reset feature that takes the money off the table at the $10 line if the P/L is greater than MidGridGain regardless of whether the target is hit. This is disabled if the Sensitivity window setting is >200.

So if you like the 2.11 version and want a dynamic grid, just jack up the Sensitivity and you'll have it with this version.

By the way, those pesky Global Variables are necessary, but stick around a long time. If you want to make a clean slate I've included a couple of scripts, one to help identify them and a second to wipe them out, selectively or otherwise.

About updating - if you stop the existing EA and open this one with the same unique number, it should recognize the existing grid and not reset. I haven't tried this.
That would be no. It rewrote a new grid on top of the old one. It didn't close the old grid, though, or reset. Just ignored that there were trades already there.

NOTE: Uploaded files again at 05:28 EST with code to close the grid more efficiently. There were two downloads at that point if you guys want to overwrite the previous file.

I think you will get your money back because client funds are put into a separate account and are not treated the same as the brokers own money.

50k is covered though under this scheme.

http://www.fscs.org.uk/what-we-cover...sation-limits/

However, I see there was talk of FXCM actually buying Alpari UK......

I honestly think you'll be fine but appreciate you have money tied up with them which could be put to better use with another broker!

I'd have them on re'dial asking for my money back asap.

About backtesting - with the resets at fixed intervals ($10 lines) we should be able to backtest this with a swipe at validity. One thing I noticed was that the month leading up to Christmas was especially hard on more aggressive settings, while the month (til now) after has been very rewarding for aggressiveness. One combination of settings I found made almost 500% after Christmas, but cratered the account before Christmas.

In general, the settings that do ok before Christmas are the most conservative ones. 1% target, 0.1 trade amounts, 100 or 300 buffer, 50 step. Settings like this don't yield nearly as much after Christmas (20-30% is common) as others, but they make about the same before Christmas.

All the tests I did assumed closing out before Christmas and reopening after - that gap killed about everything. The settings mentioned above yielded around 60-70% for the whole period if you shut it down, but that dropped to around 10% if you left it running.

If anyone has settings they find that work well in all seasons it would be good to know.

Whatever the situation, I truly hope it'll turn out positive for you.

If you're already looking for another broker, there's an interesting discussion going on here. Plenty of survivors to choose from, with the quality of their execution under careful scrutiny.

Good luck, dambuster!

Absolutely wonderful! Thank you for not letting go of your efforts to push things forward, dgodfrey. WOW! The dynamic grid has arrived (yes, I did read the word of caution, lol)! Also, thank you for your input on settings/seasons. Now that we have a robust tool, we can finally focus on finding ways to optimise the performance, depending on market conditions.

Thanks, Kosmo!

The trick with the original was that it really couldn't be backtested - it was completely dependent on where the first grid initiated, so the outcomes were totally random. Start in one place, you got one outcome, start a few $'s away, and it gave an entirely different result, regardless of settings.

Since the $10 grid trades have the same entries, we should be able to backtest effectively. I don't have a good history to test with, but I might look into that.

Still, what's important is that the settings we pick be robust in several market conditions, because we don't know what's coming next. I'm not so concerned with max money output as I am with settings that will take a lot of abusive market conditions and still not crater. When I stumble across settings that hold up where others blow out the account, I'm willing to take a little less should they not be optimum for a trend like the one we've been in this month.

~dgodfrey...thanks for the updated EA...I will probably start testing with it, if it runs ok without bugs for everyone. Good luck on your broker situation!

Thanks, I think I'm ok with the broker thing. FXCM is talking about buying out Alpari - how's them apples? But I bet there's some risk analysis folks out looking for jobs today.

I'm hoping the code works well too. I might have to go back to the inside out closing of pending orders and do some more, that's kind of tricky. It would seem to be just FIFO, but for some reason it doesn't do that when you just reverse the order of closing that was there. Still, that won't keep it from working.

How not to love pure and simple PA! Isn't that a perfect place to short? Let's see if we can get back to $1250...
NOTE: Gold closing in one hour today.

Happy MLK day...

Check out this video on Fibs that will blow your mind...scroll down on the story to the Vimeo.

http://artstyle.sfglobe.com/2015/01/...e_fb_new_32233

Sure hope so, a nice $30 shot would be cool

Was doing some backtesting and for the period from July til Christmas, had one setting that yielded consistently high results, as much as 1000%. But along the way it showed a negative account balance at one point, so that sounds pretty flaky. Oddly enough, the equity during that period was flat, what does that mean?



I'm wondering if I can code this so that it regularly makes withdrawals. That would be cool to backtest.

So far, gold is starting this year identical to last year. Last year it was russia threatening ukraine. This year it's a hedge against the falling DOW and other stocks, as oil stays low. (I filled up my car for $1.75\gal) .

~kosmo...are you still using the same settings as before on your $10 entry point version? Or did you expand the buffer? Your's is doing really good. The good spread helps, too.