The zombie effect: Do retail traders actually bother with backtesting? – Op Ed
Accessibility to diversified marketplaces has never been so empowering for retail FX traders. There are not only a litany of very high quality retail focused brokerages with massive capital bases, public listings on prestigious exchanges, and their own proprietary trading systems which are fully comprehensive and access thousands of listed derivatives and OTC asset classes […]
Accessibility to diversified marketplaces has never been so empowering for retail FX traders.
There are not only a litany of very high quality retail focused brokerages with massive capital bases, public listings on prestigious exchanges, and their own proprietary trading systems which are fully comprehensive and access thousands of listed derivatives and OTC asset classes which are tradable at very low cost from anywhere at all by members of the global public.
The innovative leaps forward that have been made by the upper echelons of the retail electronic trading sector over the past decade have created an environment which allows traders who occupy themselves with online investing as a secondary cursor to their everyday life to almost emulate the professional and institutional traders of Chicago, New York and London.
Behind such technology and market access are prime of prime brokerages which distribute Tier 1 bank liquidity via their own direct relationships with the world’s largest interbank FX desks – that in itself not an easy structure to maintain – that bring the listed derivatives venues of Chicago or the investment banking market makers of Canary Wharf into domestic houses from Toronto to Timbuktu.
If this is the case, then surely the traders themselves would begin to behave in a similar fashion to the professional and institutional traders en masse.
It is patently evident that whilst there are no doubt a percentage of retail traders who are extremely sophisticated, understanding and in sometimes creating their own mathematical algorithms which measure historical data, or who use machine learning technology to aggregate market sentiment data and then plan ahead to, in some cases, be able to predict future behavior of markets and traders so that they can build their strategy around it, however this, at best, is 10% of all global traders, most of whom are loyal customers of the giants of Britain and North America.
In my opinion, one of the reasons that Interactive Brokers, the most highly capitalized retail FX brokerage in North America, began to angle itself more toward Eligible Contract Participant (ECP) business is because the company’s client base was at such a high level that it could stipulate that its core business should be hassle-free and stable ECP clients, whilst then raising the minimum deposit for retail traders to $10,000 globally.
The company continues to build specialist tools for traders, largely focused on analytics and data, boding well with the firm’s ethos of looking after ECP business. An eligible contract participant (ECP) is a group or individual allowed to engage in financial transactions not open to retail customers.
In the United States, the Commodity Exchange Act outlines the requirements for eligibility, stating that those seeking to become eligible contract participants must have sufficient regulated status or a specified amount of assets. ECPs are usually corporations, partnerships, proprietorships, organizations, trusts, or other entities that have total assets exceeding $10 million. This, combined with high net worth retail clients means that there is demand for high quality proprietary tools which work in conjunction with the firm’s platform.
Commercial decisions such as this demonstrate that some of the large firms realize that approximately 80% of retail traders are not utilizing any form of data analysis, and are simply trading on third party platforms (usually MetaTrader 4) on a completely speculative basis.
Smaller brokers know this very well indeed, hence the widespread adherence to a lead buying, lead conversion and customer churning mentality, coupled to lack of investment in proprietary trading software.
The prevalence of MetaTrader, the outsourcing of client bases to MetaQuotes by almost 85% of all brokerages worldwide, and the lead conversion and traffic buying priorities rather than engagement with experienced traders and operating as a fintech-driven financial services company has to a large extent ‘educated’ retail traders to be targets for sales calls, bonuses, and small-sum speculation due to the affiliate marketing background and mentality that is behind smaller Cyprus and offshore brokerages.
Thus, analytical traders are concentrated among a handful of companies, whilst others are viewed not as traders but as leads and a conversion funnel rather as they would have been by the gaming or affiliate networks that gave rise to many MetaTrader brokers in Cyprus or offshore.
One such strategic analysis that has only been mastered by a small number of traders is backtesting.
In a trading strategy, investment strategy or risk modeling, backtesting seeks to estimate the performance of a strategy or model if it had been employed during a past period. This requires simulating past conditions with sufficient detail, making one limitation of backtesting the need for detailed historical data. A second limitation is the inability to model strategies that would affect historic prices. Backtesting, like other modeling, is limited by potential overfitting. That is, it is often possible to find a strategy that would have worked well in the past, but will not work well in the future.
Despite these limitations, backtesting provides information not available when models and strategies are tested on synthetic data.
Backtesting has historically only been performed by large institutions and professional money managers due to the expense of obtaining and using detailed datasets. However, backtrading is increasingly used on a wider basis, and independent web-based backtesting platforms have emerged. Although the technique is widely used, it is prone to weaknesses. Basel financial regulations require large financial institutions to backtest certain risk models hence all institutional and professional traders would be very au fait with it, however it is completely available to retail traders.
Many analytics companies exist which provide the facility for backtesting to retail traders, however many small retail brokerages overlook this and continue to provide the same MetaTrader platform as over 1,000 others and adhere to the same sales techniques, concentrating on conversion and CPA deals, rather than operating as a genuine brokerage. By giving their clients the chance to backtest strategies and understand historical data, whilst also being able to use market sentiment analytics, the need to continue to churn clients would be less due to a longer client lifetime value and the ability to attract genuine traders with abilities and analytical skills, thus elevating the status of the brokerage.
There are traders who are very much engaged with online communities and forums, who can be reached out to by brokerages. We spoke to a retail trader today who said “I want to be able to play around with data in Python and run statistical tests, download it locally for back testing and use it for live trading. So far I’ve been using Interactive Brokers, but I am looking for something more user friendly.”
Preferring a Linux operating system, this particular trader had been testing the Dukascopy Tickstory.com solution which he considerd to look good but had concluded that due to its Windows-only environment compatibility, it would not be usable long term. An associate had written a PHP script that downloaded historical data, which adapted the data from Dukascopy’s system to be used independently hence therefore can work with a Linux system.
This level of technological understanding is indeed prevalent among a small number of traders, and attracting these traders is worthwhile for brokerages.
Some brokers have begun to offer their own backtesting software, but it is in many cases not used by sales staff to attract customers or engage existing ones. There are also many independent backtesting systems available, and have been since the early days of retail FX trading, with independent systems such as Forex Tester 2 having been available in the latter part of last decade.
MetaTrader 4 had been designed with its own backtesting system, which would make it very easy for brokerages to use that as a retention tool and as a means of attracting higher levels of traders, however the very ethos of many firms using MetaTrader is that they do not want to involve themselves in trading systems, an ethos which is at odds with operating an FX brokerage, as by nature, such entities are intrinsically linked to technology.
Attracting and retaining clients is now one of the biggest challenges for brokers as the FX industry matures and as CPA (Cost Per Acquisition) which is a major factor within an online advertising model grow highers, averaging $1000 per funded client in western countries.
To tackle the issue, some of the smaller retail brokers to whom cost of acquisition and retention is extremely important, have been ever more active in developing their differentiation angle via various means, ranging from promoting the social trading experience to offering deposit bonuses and cash rebates, as well as loyalty programs and interactive marketing campaigns, while automating data intelligence for the sales teams.
On that note, trading platforms such as MetaTrader 4 may pose an issue given that the ubiquitous trading platform acts both as a key facet with which to onboard clients and as a retention problem, becoming one of the main criteria when choosing a broker to open an account with, which also means those brokers risk being easily substituted following a minor incident or an aggressive marketing campaign by a competitor.
This may well appear as though it is a complex lesson in mathematics; a matter in which accountants must balance the cost of leasing trading infrastructure from a third party platform supplier against the cost of buying media and generating leads, then subsequently factoring in the cost of sales staff to call the leads and then add up the profit post-conversion.
The reality is that in today’s retail FX industry, where the end users have become ever more sophisticated and can absolutely tell if a broker is surviving from their deposited capital instead of connecting to a live market, which has exponentially improved the standards in terms of execution of trades within many medium sized retail brokerages, however it has left a conundrum, that being the completely outmoded marketing model which is not compatible with the commission business that brokerages offering genuine market execution now provide.
The brokers that came to market in the middle of the last decade with a profit and loss model (living from client losses) were never able to gain market share among quality jurisdictions where the established competition was already operating a good quality trading environment for clients – and rightly so, nobody laments the passing of firms which do not have their customers’ best interests in mind – instead coming from the affiliate marketing and media buying sector.
Now, many of those entities have adopted proper execution, facilitated by the market infrastructure specialists and prime of prime brokerages that serve the retail FX industry, but their marketing techniques in many cases remain in the mid 2000s, and in the retail FX industry, the mid 2000s is akin to the dark ages.
The fact of the matter is that high CPA of today requires ambitious goals when it comes to the client base, which means having enough available funds to embark on large marketing operations and have a unique value proposition to lower the CPA.
The quest for differentiation may also lead to the in-house development and maintenance of proprietary software despite the obvious costs because in addition to promoting their own platform to their clients and increasing the rate of client retention as they grow accustomed to an environment they will not likely find elsewhere, these brokers can also reduce their costs or even make profit out of offering white labeling solutions to other firms.
Simply, buying leads and hoping for sales calls to convert them into live clients is futile.
This is because the same leads are being recycled between brokers, the same leads are being targeted by firms offering similar propositions, and the media buying cost vs the actual conversion rate is as low as 1% in the retail sector.
Hence, one-off retail clients from non-aligned countries, with very little capital and in some cases no chance of being able to satisfy compliance requirements are depositing between $500 to $1000 as a first time deposit, and the cost of acquiring the customer is now past $1500, with a lifetime value of just three months.
Whichever way that is viewed, it is a loss.
One of the reasons for the high cost and lack of efficiency is that, according to two London-based media analytics firms that have provided digital marketing data to FinanceFeeds, the potential customers are in many cases no longer potential customers because these are leads that have been to over ten brokerages, hence the same people are being bombarded by the same calls from almost identical companies, and many have lost their initial deposit with the first company, and are therefore now worn down by the repetition of calls.
By moving away from the conversion/lead buying/CPA model that formed the online affiliate marketing sector in Israel, Cyprus and for many offshore firms, and embracing the analytical, mathematical and computer science orientated world of professional traders, the retail sector would elevate itself tremendously and attract the quality end of the market, hence themselves be viewed as what they should be – genuine access points to a global market for traders in the know.
Indeed, if trading itself is a mathematical science, then the operation of brokerage should be too.