What to do…

Got a spam email from FXCM about a contest on some new platform they’ve built (themselves it seems). It is called Strategy Trader and after a quick download I must say that it actually “feels” quite interesting. FXCM is of course claiming that it is much better than anything else on the market and goes on to “compare” features with MetaTrader 4. We all know that MT4 is old, so the comparisons are a bit unfair. MT5 is already out, even if it hasn’t gotten too much attention yet. I can’t say if ST is better or not, but at least it feels new and fresh. I’ve just looked around it for 15 minutes now thou.

The Strategy tester is multi-threaded so backtesting their complementary Grid trader was quick. They brag about 1 year of 1M data for free in the platform, but that doesn’t impress as we’re all hooked on tickdata from DukasCopy. A nice feature is that it is possible to test with both ask and bid prices.

The contest they’re arranging comes with a $10,000 first price, so it might be worth it to try and port some of my MT4 EAs to the Strategy Trader. Considering that I haven’t had time to do the same for the MT5 contest reality probably is that I won’t make it in time for the deadline thou…

Check it out: http://www.fxcm.co.uk/automated-forex-trading.jsp#Overview

Update: After messing with the platform for a few hours now, including downloading a few sample strategies from their forum I must say that I am impressed. I will certainly try to get a scalper strategy of mine ported from MQL4 to C# and the Strategy Trader. I’m planning to do so in steps. The first step will be to move functions from the MQL4 code to a C# written DLL. I can then verify that the converted code gives the exact same behavior as the original since I can call the functions in the DLL from a wrapper EA. I can then use the same file with function in a new C# wrapper in Strategy Trader (and call the DLL from MQL5 if I want to also).

The platform is different to work in as MT4 really been the only tool I’ve been using for a long time and gotten very used to, but everything is starting to fall into place. Some things I like (reports for example), other are still confusing. The speed is undeniable, everything is very responsive and the backtests are lightning fast.

Just woke up and excitingly checked my EAs performance during the night. Found this (on both my JadeFX and The Collective FX accounts):

That is some seriously unlucky stoplosses. Out of the 9 positions that were taken 7 just hit their stoplosses before EURGBP returned to make the remaining two positions take profit. If this had happened at a non-ECN (and just one) I would be very tempted to think about stophunting. But as it happened at all my brokers I guess I was just very very unlucky.

So, 10% lost of the JadeFX account, resetting the account to 0% growth during September.

Quick update on the performance on my main scalper account at JadeFX during August. I’ve previously posted about the excellent results during July. August was also good, but I had one day that basically all trades went to stoploss for a negative hit of 551 pips or about 15% of the account (which also corresponded to my risk-appetite right then). It was inevitable to happen of course, and it will happen again. I didn’t change any settings after the “crash” and the losses was made up the coming 10 trading days. Excellent.

Other than the massive losing day (17th of August) I had two more days with negative pips. Overall the account made 742 pips which amounted to 18%. I am very pleased with that, considering that the maximum loss scenario was realized.

Going forward I have changed the setup a little. During July and August I have been experimenting somewhat by running three different setups of the main moneymaker EA here. Now it is clear which setup I will use going forward. It is very stable (if I had only used that one on the 17th the drawdown would have been much smaller).

I have also introduced a new EA from the Funyoo forum on my live account on the 1st of September that I’ve been optimizing with Dukascopy tickdata. I’ve been trading it with great results on demo for the past two weeks, and so far it seems as good on my live account (only had two trading days yet, but the trades are identical with demo trades taken).

http://www.freelancer.com/projects/NET-C-Sharp-Programming/Prepare-Forex-Statmeents.html

Totally sick. Even crazier that people want to help the moron.

I just asked myself that question as its been a few days since I started backtesting with tickdata from Dukascopy. Even short runs take a lot of time… Optimizations of a few parameters can take weeks. I just aborted a optimization run that was supposed to  take 1800 hours (that is 11 weeks roughly).

So, naturally I started thinking that maybe it is time that my 3 year old Intel Q6600 is replaced by a faster part. I must say that it have performed absolutely excellent, especially since I’ve overclocked it to 3GHz all the time I’ve had it. It is a quad core CPU, but sadly MT4 is a single threaded application, so only one of them is used at a time. I usually solve this by running four instances of the same test with one of the parameters range cut in four, but it is a hassle. MT5 solves all this as it not only is able to use all available cores in the host system it is able to out-source processing to agents running on other computers. But reality right now is that MT4 still is the main platform, and it probably will remain so for at least another 6-12 months.

I’ve googled for benchmarks, but I can’t find any. Sure it is a niche, but I really expected at least some results for MT4. I thought surely someone had bought a new computer sometime and tested it by running the same backtest on both. I know I would, but it seems I’m alone in this. Can’t find any benchmarks for either MT4 or MT5.

Read more…

Just a quick reflection that for the past two months basically all scalpers seem to have been doing very well. I’ve re-enabled some old favorites, like FAPT50 and Forex Shocker 1.1 trading EURGBP and they’ve been doing very well. Even GBPCHF by FAPT50 have been very nice. Just watch your risk exposure, calculate the maximum you want to risk in a trade and then all EAs (possible) trades together. It is very easy to keep increasing the lot sizes in periods like this when everything is going great, leading to more risk that you actually want to take.

July at JadeFX, my main scalper account.

As you can see above July was great at JadeFX. 38% increase and just one day with negative pips (33 pips lost). The best day gave 125 pips.  All trades are by scalping EAs.

I found a great (old) article about getting optimal data for backtests in MT4. I personally usually use history data from JadeFX because that’s the broker I, mainly, use scalping strategies at, but it is full of errors and rarely gives even 90% quality.

Check it out: http://eareview.net/tick-data

I’m gonna test this as soon as I get a hour or two over… IE hopefully on Saturday.

Just a short recommendation if you like MQL4 programming or at least running long optimization runs in MT4. Join the private “elite” section of Funyoo’s forum over at tradingsystemsforex.com. It is $99 per year and gives you access to a section on the forum with over 280 different EAs. The huge number is both a great thing and the biggest problem with the forum. There are actually too many to handle. Many are made by Funyoo on request by members and never make any profit, but there are a few really good ones actually.

I’ve been a member for two months  and my poor CPU is working hard on optimization runs all the time now

Are you interested? Seems to be some openings in freelance MQL4 programming. The funny thing really is that these guys want you to do all the work; including coming up with the strategy, implementing it, writing the manual, providing snapshots of trades and even making a video of a live trade. Then they proceed and make a fantastic website with the usual bullshit and set a $99-$499 price I guess.

This one is even better I think. They just want nice backtests. No mention of that the strategy should work for live trading. Perhaps nice backtests is all you need to sell a few copies… I actually have a strategy that backtests extremely well, but can’t make money even in demo trading. Perhaps I should sell it

I check the MetaTrader / MQL4 listings at Freelancer.com now and then. I could actually see myself taking smaller assignments (say 8-20 hours of work), but so far I haven’t come across anything interesting enough to bid on. Ideally I’d like to make implementations of already manually tested strategies / systems or implementation of indicators. Generally I think the pay is low for this. If you want a decent $/hr ratio you really have to stress the implementation, it won’t be high quality. There seems to be a lot of “low cost” countries represented in the bidding thou, for a Chinese programmer $400 is a big deal of course.

I haven’t really been paying to much attention to all the hoards of EAs released this spring, especially not scalpers, but this one really caught my attention by showing extremely good performance in scalping exotic pairs (like GBPCAD, CHFJPY and others). It is called Primeval EA and is actually a collection of strategies, but personally the scalper is what I’m interested in (there’s also a hedger, gridtrader and some kind of martingale strategy. They might be alright but the scalper is on fire).

There’s tons of backtests on the site, that obviously look good (why would anyone publish them otherwise?), but there’s also verified accounts at MyFXBook (2 demos and 1 live account at FXCBS).

Here’s the live account: http://www.myfxbook.com/members/primevalea/primeval-ea-v21-live-fxcbs/34374

Yeah it has made 45%, but the drawdown is a bit to much for my taste. I would lower the lotsize to 0.1 or even lower with this small account. But still, the pips taken is impressive and the maximum loss is survivable (38 pips, the SL is set at 36, but I guess there was some slippage).

The demos are doing even better, which is sad of course, but very common. Blame the brokers for not offering fair demo environments.

http://www.myfxbook.com/members/primevalea/primeval-ea-v21-fxcbs-demo/34377

http://www.myfxbook.com/members/primevalea/primeval-ea-v21-alpari-uk/34379

What are the downsides? Well, it is a scalper. They tend to work well, bring steady profits and then brake down now and then. It has happened with all other scalpers I’ve tried. But with proper risk management this can be handled of course. A good thing with this one is that it trades so many pairs profitable, which brings down risk somewhat if one or two stops working. But like most of us know, many pairs are quite correlated and a stoploss in one pair seldom comes alone.

The other downside is the price. It is $600. Yeah, four to six times the usual $99 - $149. But, really, if it keeps performing this well, it is worth it. There is a 45 day money back guarantee (and since it is a Plimus product it is very likely that you actually get your money back if not satisfied, that has been my experience with some priors at least).

So, I’m leaning towards a buy. I’ll keep you posted of course. If anyone else takes the plunge, let me know.

Check it out. Primeval EA.

Update: Oh yeah, you can use it on 2 live accounts per license, which is a added perk of course (and suits me great, I would put them on my existing JadeFX and The Collective FX accounts.)