System Trading Blog

Make Your Money Work Harder

Archive for the ‘’ Category

DIY System Trading

leave a comment »

Can a person, having full time occupation, do the system trading by himself? The short answer is “Yes, but too risky”.

Why? By far the hardest, the most time consuming work that requires certain qualifications as well is system development itself. And this is not an issue. Companies like Covestor, Collective2 or Zulutrade have done a good job to offer sufficient amount of back-tested and well-maintained trading systems to choose from. To acquire some basic working knowledge on trading system analysis shouldn’t be a major problem either.

The key issue with DIY system trading is still far from 100% reliable routing of the trade from the system vendor to live account. The hard fact is that automated system trading from to time results in mismatched positions in live account. Mismatched positions, if not corrected in timely manner, is one of the costliest mistakes one can make in system trading.


Written by A.S.

January 5, 2012 at 5:59 pm

What happened to our real-time experiment?

with one comment

On June 1, 2010 we started a real-time experiment to illustrate how the system trading works and the profit potential of this type of investing.

We followed through for about 7 weeks before we diverted the resources to an income-producing work at hand. During those 7 weeks the results were mixed.

Overall, pure currency systems (Alpari PAMM, Tradency, Zulutrade) performed better than stock (Covestor) or commodity systems (Collective2). Since then, Covestor has redesigned its web site and the service offer looks more compelling than ever. The performance of Covestor models (i.e. strategies or trading systems) have recovered noticeably from last June. As at the time of writing, top-performing Covestor models are back on track with up to 55% returns during the last 3-month period (apparently, in line with the general market upward move).

Tradency has redesigned web site and now have the product offer directly from their web site rather than kept it obscured somewhere in brokers´ web sites. Collective2 and Zulutrade have kept their service offer basically unchanged, save for some non-essential features (such as additional languages and too-simple-to-be-useful analytic tools).

The basic premise of system trading is that, theoretically, a good trading system should perform well irrespectively of general market direction. In practice, most trading systems depend on certain market conditions, their performance is limited in time and it is critical to recognize and scale out systems whose ¨production stage¨ is over.

One of the conclusions to draw from the experiment is clear: constant monitoring and adjustment of the system portfolio is a must. Hardly any of the trading systems, originally included in the portfolio, pass the test 5-6 months later.


On Collective2, out of 11 systems, 9 are down, 2 are up. The average decline is about 10%.

On Zulutrade, out of 15 systems, 4 systems are not identifiable, 6 are down and 5 are up. Average up is about 8% (not counting the unidentifiable four).

On Tradency, out of 6 systems, 2 are not identifiable, 3 are down and 1 is marginally up.

So, out of 32 systems, only 8 or 25% are good half-a-year later. The other 24 systems would have been removed from the system mix in the due course of the ongoing selection process.

System trading portfolio down 1.1% in June

leave a comment »

Value of the sample portfolio, run as a part of the system trading gig, declines by 1.1% in June.

Value of three portfolios (with trading systems sourced from Alpari PAMM, Tradency and Zulutrade accordingly) is up while value of two portfolios (Collective2 and Covestor) are down.

The best performer with 6.3% gain is Alpari PAMM portfolio. The worst performers are Covestor (6.4% loss) and Collective2 (3.6% loss).Value of S&P index during the same period has declined by 2.45%

Written by A.S.

July 4, 2010 at 5:54 pm

Sources of Trading Systems

with one comment

The ecosystem of automated trading develops fast. Capital allocated to system trading through industry leaders, such as Covestor or Collective2, reportedly grows by 50% annually.

There is still a plenty of room for growth as the investors ¨chop off the old block¨, i.e. disallocate the funds from chronically non-performing and increasingly risky traditional investments in stocks and bonds.

I here list 8 sources for you to choose as a provider of trade signals for your system trading account. Those sources are the ones likely to be used by both emerging system developers as well as investors.

Collective2 or C2 is an internet-based company. C2 collects performance data of various trading systems, independently verifies such performance data and makes the systems available to subscribers to trade in real-life trading accounts. At the time of writing there are about 7’000 trading systems registered on C2.
The way it works is — you go to C2 web site, register as a user, find a trading system which, as an example, shows 88% annual return, make a few clicks to set the system up to send the trades to your brokerage account and, if the systems lives up to the expectations, you have just arranged for an investment with 88% annual return.
C2 promotes itself as think of Collective2 as „do-it-yourself” hedge fund.
C2 works only with hypothetical data, although it checks all the signals in real-time and provides both estimates (slippage, realism factor) and tools for investors to evaluate what performance in live accounts would look like.

Covestor (and
Covestor is one of the best-managed and innovative trading system aggregators.
Covestor displays performance data from live trading accounts and is limited to stocks only.

Zulutrade deals with hypothetical forex signals only.
Zulutrade is probably the fastest growing of all system aggregators. It takes only a couple of minutes to sign up as a system vendor. Zulutrade is equally easy for investors as account opening procedure is straigthforward and easy with any of the selected brokers.
There are no fixed fees for neither system vendors, no investors. Systems can be added and removed from trading in live accounts with just a few clicks and at no cost. Fee for the services is pay-as-you go in form or bid-ask spread markup. Such spread mark-up (about 2 pips round-turn for the most liquid currency pairs) is then shared between Zulutrade and the system vendor.

Tradency service is available through selected forex brokers, including several major retail brokers, such as FXCM.
Tradency system performance data is hypothetical rather than based on live account trade logs.

If you´re looking for system aggregator service for forex signals based on live accounts, then is for you.

Myfxbook positions itself more than a community of forex traders rather than system aggregator service provider.
Forex traders allow data from live forex accounts to be explained on Myfxbook website. Potential investors can review the data and contact the trader directly for signal delivery terms and conditions.

AlphaClone compiles data about major equity positions from 200 major hedge funds and institutional investors from their public filings.
There is a time-delay involved as such public filings with SEC are made quarterly. If you feel comfortable to ¨clone¨ positions of, say Warren Buffet, this is the service to use.

Alpari PAMM
Alpari is a Russian-owned forex broker with operations in several countries, including UK and UAE.
Alpari customers may enable their live account performance to be available for public disclosure. If done so, the live account performance is ranked and any investor may open PAMM account to copy the trades of the most profitable traders.
The fees related to PAMM account are structured as performance fee. Under this arrangement, investor pays a share of profits as a performance fee and such fee is shared between the broker and the trader.

Source: free e-book Be Your Own Hedge Fund Manager.

Looking for more trading system sources

leave a comment »

I´m looking for more trading system sources, apart from, and

So far I´ve spotted and marked for further research the following ones:

1. FXCM and Alpari
Both are forex brokers. Brokers, generally, are well-placed to identify the best traders and trading systems. They keep customer funds and have all the trade information recorded on their servers.  At at the time of writing, FXCM  has 2077 systems listed, Alpari – 499.

2. AlphaClone allows investing in ¨the stock ideas of top managers at the time their stock ideas become public¨. There is time delay involved, of course, as funds disclose top holdings quarterly. Basically, information from 200+ funds ¨cloned¨ into 15´000+ investment portfolios.

Performance [Misre]Presentation on Zulutrade

with one comment

Looks like total geeks or rogue marketers run Zulutrade. I tend to believe that they’re geeks since the site has some whistles and blows and a very good business concept.
When it comes to presentation of the content (i.e. trading systems), it does a good job to display junk prominently and gives no other tools but to dig trough the list manuallly.

This is the recent e-mail from Zulutrade (click thumbnail on the lef). The yellow markings are mine. The system is ranked #7. The average profit per trade are 2.8 pips. 2.8 pips, modelled on 2 pip spread?! Empirically, the spread is about 10. 487 trades. Adjust for fees and slippage and you’re likely to get a deep-loss system. This is a typical representation.

Very likely another system (click the thumbnail for a snapshot) with 35 average profit per trade is profitable, but somehow it’s digged in as #343 after all the “a couple of pips systems”.

As well, the “advanced filter” is of little help as it doesn’t let you filter any basic parameter like return, drawdown, average profit.

Currently Zulutrade lacks pedigree. Something you can’t say about and, particularly, But then, Zulutrade seems to be successful in building customer base and, hopefully, one day they’ll fix the shortcomings.

Written by A.S.

May 20, 2010 at 7:10 pm

Zulutrade shortfalls are know, but are they intentional?

leave a comment »

I´ve posted a number of times, that Zulutrade has critical shortfalls and its performance rank table is virtually useless.

Conceptually, Zulutrade is great (so is Collective2 and Covestor), but … so far it´s hardly designed to identify good trading systems and make money for customers. Worst of all, it may be intentionally so.

The key shortfalls are:

– Performance tables are logged on trade close-price basis rather on tick-by-tick basis.
This shouldn´t be a problem as long as reasonable stoploss levels are obserserved consistenly. Unfortunately, this defect allows to, effectively, fake performance data and by far most top-ranked systems´ performance looks good only due to this shortfall. This defect obscures real drawdown levels.

As an example. A system vendor opens a position, keeps it open for 3 weeks and closes with 10 pip profit. On the performance chart it is logged as clean 10-pip-profit trade and looks great. What is NOT logged that this trade at some point was in MINUS 700 pips. The performance chart doesn´t display the drawdown of 700 pips.

– No adjustment for leverage
Performance stats display only average pips for trade. The data in such form is about as informative as “average temperature in hospital”. It´s a huge difference if, say 30, pips per trade is achieved by trading 1 lot at a time or 10, 20 or even 100 lots at a time. The data would become usable and informative, if presented in ¨average pips per each lot traded¨.

Alternatively, there may be 10-20-30 or more positions open at the same time. Often in the same direction. If it works, the pip count goes up nicely. However, the problem is that it is unlikely that any end-user of such service would allow so many trades from any single system vendor. He may capture only a small fraction, say 10%, of the action and performance would look much different in comparision with performance posted on zulutrade.

Others write about the problem here and here.

Worst of all are not the above mentioned shortfalls per se, but it´s the fact that any system vendor who wishes to reach the top of the performance ranks HAS TO EXPLOIT THE SHORTFALLS, in fact defects, of the system at the expense of trading system quality.

Since the aforementioned defects are technically easy to fix (a bit more computing power to log performance on tick-by-tick basis and a minor change in reporting algorithms is all it takes), I suspect that Zulutrade intentionally keeps the reporting system the way it is since it is more profitable for them rather than have clean, efficient performance representation. Why? First of all, it generates a lot of nice-looking performance curves to attract new customers looking for easy, outsized profits. Secondly, the smooth operators who use the loopholes left for them to exploit, are usually very high frequency traders.

Profitable system which trades infrequently and finds trades with great risk-to-reward ratio is what customer needs. However, it´s not necessary what zulutrade needs. For zulutrade, a break-even or high-risk-low-profit-lots-of-trades system is much better since zulutrade receives it´s income from markups and amount of such income is a direct function of trade volume. There is a clear conflict of interest between the customer on one side and intermediary (zulutrade) and system vendor on other side. I´ve seen zulutrade business plan, it may well be that it states that they are in entertainment business (the same as casino) or that they should encourage churning as a business model … or may be it´s just lack of management vision, or funding or whatever …

The bottom line is that it´s ok to use zulutrade as an interface for routing trades from system vendors to customers´ brokerage accounts, but performance stats (and ranking, in particular) is nothing one can trust. As the shortfalls become common knowledge, at some point it´d be profitable for them to sort out the mess …

Written by A.S.

January 24, 2010 at 1:41 pm