RSJ 50% Liffe volume: The hedgie with a 99.9% success rate

Discussion in 'Wall St. News' started by just21, Jan 10, 2017.

  1. just21

    just21

    The hedgie with a 99.9% success rate
    Harry Wilson, City Editor


    January 10 2017, 12:01am, The Times

    [​IMG]
    Despite being one of the most influential firms in the London markets RSJ, co-founded by Karel Janecek, is barely knownKAREL JANECEK
    Share
    Save
    As darkness fell on the Sydney skyline on New Year’s Eve, a 43-year-old Czech billionaire stood barefoot in a yellow polo top and black shorts, an arm extended out towards a party in progress just below his terrace.

    With a crop of blond hair and a fresh rash of stubble, Karel Janecek does not look like a typical billionaire hedge fund manager in the post on his Twitter account. Indeed, little about the demeanour of the maths-genius-turned-financial-programmer would hint at the central role he has played in financial markets for more than a decade, in which RSJ, the firm he founded 23 years ago in Plzen, the home of pilsner beer, has become one of the biggest players.

    Even in the Square Mile, RSJ is barely known, despite being one of the most influential firms in London financial markets. On a daily basis RSJ makes markets in thousands of financial products and is one of the leading, if not the biggest trader, of short sterling and long gilt futures contracts, meaning it has a crucial role in determining the price of products at the heart of the British financial system.

    Even more extraordinary is just how quickly RSJ has gained such dominance. Within two years of placing its first derivatives trade on Liffe, now called ICE Futures Europe, in 2002 the firm became the biggest player on the exchange and 15 years on remains the one to beat, turning Mr Janecek into the Czech Republic’s first self-made billionaire.

    Mr Janecek, who owns about 40 per cent of the company according to official documents, has made so much money from RSJ that he retired a couple of years ago to focus on pet projects, such as rooting out corruption in the Czech government and promoting his idea for “democracy 2.1”, which involves using computerised voting to make the electoral system more efficient.

    Explaining his ambitions to students three years ago, he said: “People who are constrained by living expenses cannot do everything and I wanted to have full freedom to be able to decide what I wanted to do . . . and with the success, with the money, finally came the thought: ‘What is the most important thing to do next?’ ”

    Three years ago, Bloomberg reported that the Prague-based business traded products worth a notional $106 trillion on an annual basis. To put that figure in context, each week this firm, employing about 30 traders, buys and sells contracts worth roughly the same as the annual output of Italy.

    Despite its size, and there is a good case for arguing that RSJ is larger than American rivals such as Virtu, Citadel and Jump Trading, it is nearly six years since anything has been written about the firm in the mainstream press.

    RSJ’s size has given it a direct hand in helping to shape the regulation of the markets it trades on. When in 2007 the then NYSE-Liffe proposed a new formula for allotting orders, RSJ lobbied against the move. The firm claimed to be responsible for one-in-ten trades on the exchange, giving it a considerable voice in its running.

    On Eurex, the Deutsche Börse-owned derivatives exchange, Bronislav Kandrik, chief trader at RSJ and one of its ten shareholders, is a member of its council that meets to discuss how it is run. RSJ is thought to be one of the biggest traders on the Frankfurt-based exchange.

    In a barren year for hedge funds, RSJ’s in-house proprietary algorithmic investment fund, the heart of its sophisticated high-frequency trading systems, made a return of more than 90 per cent.

    We couldn’t get their books and records if we thought they did something untoward . . . I mean, they are from Czechoslovakia, they’re going to flip us the birdBart Chilton, US Commodity Futures Trading Commission
    Whatever shocked the markets last year, RSJ’s money machine ploughed on regardless. Six days after the Brexit vote, RSJ Prop reported a net asset value per share of €791, up 5.5 per cent month-on-month, according to a filing with the Mauritius Stock Exchange, where the fund is listed. Donald Trump’s shock election victory had no impact and RSJ reported a net asset value of €1,096 per share at the end of November, a rise of 7 per cent on October.

    In the three years since RSJ Prop shares were listed, the fund’s NAV has risen by 2,632 per cent. A spokesman for RSJ said that the three-year figure might overstate the performance because of changes in the number of underlying shares in the fund as investors withdrew money.

    Even taking these withdrawals into account, the returns are astounding. To put RSJ’s performance in context, the average managed futures hedge fund lost its investors 1.5 per cent of their money last year, while macro funds eked out an average return of 1.2 per cent. Even Renaissance Technologies’s Medallion fund, regarded by many as the best algorithmic trader in the world, was on course to achieve a return of just over 20 per cent for 2016.

    The size of the fund is not known and the spokesman for RSJ said that the firm did not provide information about its assets under management. However, with a minimum investment of $100,000 and a sophisticated understanding of financial markets a prerequisite to joining, this is a fund that excludes all but the wealthiest and savviest of investors.

    Indeed, RSJ Prop’s performance is not monitored by any of the leading hedge fund index trackers, meaning it is unlikely that the fund attracts many investors outside the firm’s immediate circle of present and former staff.

    Although it has generally flown below the radar of most market-watchers, RSJ’s success and influence have attracted criticism. Four years ago, Bart Chilton, then a commissioner of the US Commodity Futures Trading Commission, the regulator that led the Libor-rigging investigation, warned how little the authorities knew about RSJ. “We couldn’t get their books and records if we thought they did something untoward . . . I mean, they are from Czechoslovakia [sic], they’re going to flip us the bird. They couldn’t care less what we want,” Mr Chilton said.

    His concerns apply as much in the UK. Mr Janecek’s website boasts that RSJ, from which he has largely withdrawn from day-to-day business, is “the biggest trader on the London derivatives market”. However, the firm operates in the UK through its Czech financial passport, meaning that none of its staff have to be verified by the Bank of England or the Financial Conduct Authority.

    This means in effect that a Prague-based firm, supervised by the Czech National Bank, is responsible for a signficant part of the market-making in the world’s three largest derivatives markets: Chicago, Frankfurt and London. This has upset small traders, who say they are being squeezed out by RSJ and other algorithmic traders.

    Andrew Bailey, chief executive of the FCA, warned last year that algorithmic traders were “very important” and that the regulator would crack down on those who put their system to “inappropriate use”.

    However, Megan Butler, the FCA’s director of supervision, admitted that no inspection had taken place of the computer code used by traders such as RSJ. Even if the FCA asked for such information, its staff would have no direct jurisdiction.

    The extent of RSJ’s presence in Britain is its computers nestled as close as possible to the servers that run the world’s largest derivatives markets, a practice called co-location.

    While some high-frequency traders have hired staff in the UK, and other big markets in which they operate, RSJ’s sole physical connection with London consists of expensive computers sitting in Basildon, the home of ICE Futures Europe’s data centre. Inside these servers are what RSJ describes as its “homemade” algorithms, built on models honed over 15 years that enable it to best even the world’s most sophisticated investors.

    How RSJ achieves such incredible returns is a closely guarded secret. In common with other algorithmic traders, the firm jealously guards its trading models, but in pages now no longer on its website it explained the principle of how its mathematicians use advanced statistics and the law of large numbers to beat the market.

    RSJ explained that its investment strategy came down to the same principle as betting on the toss of a coin. However, rather than odds of 50-50, RSJ’s “long-term observation” had achieved the financial market equivalent of discovering a 50.5 per cent bias in favour of tails.

    Toss the coin again and again and this small advantage in their favour builds into an almost certain money-making scheme.

    As the firm put it: “With each subsequent game, the probability of you making a profit will get higher. If you toss the coin 100 times, this probability will be approximately 54 per cent. If you toss the coin 1,000 times, this probability will rise to approximately 62 per cent.

    “If you toss the coin 100,000 times, this probability will be approximately 99.9 per cent. There is, nevertheless, always risk. We may have got it wrong, the probability of the coin landing tails-up is lower than 50 per cent and we have lost our bet.”

    In RSJ’s case, replace the toss of a coin with millions upon millions of “buy” and “sell” orders and the essence of its strategy is displayed. And it looks like the coin toss is very much in RSJ’s favour.
     
    TraDaToR likes this.
  2. wintergasp

    wintergasp

    Market maker, not hedge fund.
     
    lawrence-lugar likes this.
  3. algofy

    algofy

    Who's the tits in all your profile pics?
     
  4. They're not a hedge fund. But yes, RSJ is massive.
     
  5. haroldg

    haroldg

    I'm sure the 'fund' the article mentions was set up as a tax-efficient way for the principals, not the public, to invest.
     
  6. JackRab

    JackRab

    Hmm... fun to read, but I must say I find it a crappy written article.

    "Donald Trump’s shock election victory had no impact and RSJ reported a net asset value of €1,096 per share at the end of November, a rise of 7 per cent on October"

    Wut? No impact? I'm pretty sure I saw ES down 5%, European markets down 5% at opening... and ended in the plus.... If you call that no impact... you probably don't understand trading/hedgefunds and how to make money, mr City Editor.

    Also, if they don't release their AUM... who says they are big? Sure 100 trillion is definitely a lot.... but anybody here want to look at how much they turned over last year? That number goes up really really quickly my friends, especially in futures and leveraged currency trading....

    I strongly doubt they are bigger than the likes of Citadel... but again, what is big... Turnover? AUM? NetLiq? Net profits?

    Anyway, thumbs up for Karel! Wish I had a billion....
     
    Last edited: Jan 10, 2017
    TraDaToR likes this.
  7. wintergasp

    wintergasp

    They're not bigger than citadel. But you don't need to in order to make 1Bn of net worth.
     
  8. The point of the article was that the US election had no impact on RSJ. That's hardly surprising, since RSJ is a high-freq market-maker.

    And yes, how does one define "size" for a shop like RSJ? It's certainly not AUM, since they have no need for ever larger amounts of capital. Turnover statistics, which they make public, are sorta useful, but, as always, it's the risk-adjusted return on capital that matters. Given that RSJ is a private shop domiciled in the Czech Republic, it's not likely you'll see these numbers.
     
  9. haroldg

    haroldg

    HFTs sometimes quote Sharpe ratios -- typically in the 10- 20 range -- but the denominator is squishy, because marketmaking requires essentially zero capital.
     
  10. Well, yes, but those are numbers which are either not available for peeps like RSJ, who don't publicize a lot of things. Alternatively, if they're published, one has to take them with a large pinch of salt. The only numbers where you can infer something meaningful come from the infrequent official filings (e.g. Virtu).
     
    #10     Jan 11, 2017
    haroldg likes this.