From Attain Capital – MANAGED FUTURES SPOTLIGHT: ROE CAPITAL MANAGEMENT

    This post is redistributed from the original newsletter at Attain Capital Management (www.attaincapital.com).  To see the original material, please click here.

    Managed Futures Spotlight: Roe Capital Management Monticello Spreads Program

    By John Cummings

    We hear requests from clients and readers of our newsletter quite often to highlight a managed futures program which is doing something different than most others, and/or highlight an up and coming program which is not generally known or listed elsewhere.  We are happy to oblige when we can, but we usually run into problems finding a unique enough program (most are doing trend following or option selling) and a program with a long enough track record.

    Today, we highlight the Roe Capital Management, which has two programs which fit both the unique requirement and emerging requirement (having less than $10 MM under management). The Roe programs are called the Monticello Spreads and Jefferson Index programs, which for all you history buffs out there are named after the third President of the United States – Thomas Jefferson.  Manager John Roe picked Jeffersonian names for the programs given their contrarian nature and his view that Thomas Jefferson was a political contrarian.

    Who Is The Manager?

    If there is one thing we have learned about managed futures managers over the years, it is that they come in all shapes and sizes.  For every quant or rocket scientist who can rattle off probabilities of a market finishing the day in a certain direction but must look up the difference between Chicago and Kansas City Wheat; there is an old school pit trader who wouldn’t know a fractal equation if it hit him on the head but has honed his or her craft by observing and trading the markets from the floor of the exchange.

    Today’s manager, Mr. John Roe, falls somewhere in between as he started out in the famous trading pits of Chicago, but quickly realized that they future of trading was in the electronic markets.  Originally recruited to the trading floor for his tech skills (he was manning a Globex terminal), Mr. Roe eventually found himself in the middle of a changing landscape that saw seasoned pit traders heading off the floor and upstairs to trade electronically on “the screen” as they call it.

    Mr. Roe got his start in the financial markets as clerk in the Eurodollar pit at the Chicago Mercantile Exchange, which at the time was the largest (nearly the size of a football field) and most active pit on the trading floor. Over 1500 traders and clerks came to work every day on what was then known as the CME’s upper trading floor.  Years later, only a couple hundred people remain in the Eurodollar pit as electronic trading now dominates the landscape.

    While on the floor, John worked for of the largest institutional spread brokerage groups in the Eurodollar pit, where his role was to help brokers create markets on volatile back month Eurodollar spreads by executing unfilled legs of spread orders on the CME’s GLOBEX electronic exchange.  During this time, Mr. Roe also worked with local independent traders who had moved off the floor and onto the trading screen.  His service to them was to provide “color” of what was going on in the pit (who was buying, who was selling, etc.) to the upstairs traders who no longer had a feel for the momentum in the pit.

    In August 2005, Mr. Roe left the trading floor to manage a trading desk specializing in automated electronic system execution for high net worth clients, where he was formally introduced to the world of algorithmic trading.  This experience exposed John to the benefits of system trading and it wasn’t long before Mr. Roe was researching and developing trading models of his own.  Roe Capital Management was launched in October of 2007 and began trading for clients in June of 2008, right in the teeth of the now infamous credit crisis.

    Outside of work, Mr. Roe is active in philanthropy and politics.  For five years, John served on the board of directors of the Awassa Children’s Project, a Chicago based non-profit group, which operates two non-governmental organizations for children orphaned by AIDS in Awassa, Ethiopia.  John has since resigned from the board of directors to focus on Roe Capital, but he continues to remain active with the organization as a volunteer.   Mr. Roe also has a keen interest in politics and has managed several state congressional campaigns and served as a strategic advisor to a few US congressional campaigns.  In his limited free time, John likes to ski, sail, and is such a rabid Chicago Cubs fan that he named his dog Wrigley.

    How Does the Program Work?

    Both the Roe Capital Monticello Spreads Program and the Roe Capital Jefferson Index Program focus exclusively on US stock index futures markets, which are amongst the most liquid and high volume markets in the world. The programs are short term in nature, with an average trade hold time of approximately 1 to 2 days.

    But instead of using volatility breakout or momentum based strategies like the bulk of stock index traders (especially the systematic ones), the core strategy behind both the Roe Capital Monticello Spreads Program and the Roe Capital Jefferson Index Program is a contrarian strategy which looks to benefit from the fact that markets often take two steps forward, then one step back, and vice versa.

    Mr. Roe believes his models can identify periods of what he calls momentum exhaustion, where the prevailing short term trend has run out of steam and will reverse course over the next few days.  The underlying models consist of algorithms that use a variety of proprietary calculations (we tried our best to get more out of Mr. Roe, but he guards his secret sauce recipe dearly – and was only willing to give us the vague proprietary calculations) to forecast the probability that e-mini SP 500 prices will reverse direction.  Once his models calculate that a trading day’s action has signaled a shift in the probability that a move has run its course, the program looks to enter into a trade on the market close in hopes that the next day’s price action will reverse course.

    Mr Roe believes that short term price movement is more apt to see short term reversals due to basic forces such as profit taking, changes in sentiment, economic releases, and so on; which are usually contrary to the prevailing short term trend. Anyone who has ever seen the market bounce back the next day, or sell off the next day after the earnings report or merger that drove prices higher is reviewed in further detail knows how that can work.

    Trade Example: Monticello Equity Spreads Program:

    As the ‘spread’ in its name suggests, the Monticello Spread Program will enter a trade via an inter-market spread. Spread Trading is the simultaneous purchase and sale of two usually similar futures contracts in the hopes that the one you purchase rises more (or falls less) than the one you sell  Most spread traders usually buy one contract month of a market while selling another contract month of the same market. For example, buying July 2010 Corn futures and selling December 2010 Corn futures is a spread trade. But in the case of an inter-market spread, there are two separate markets involved.

    In the case of the Monticello program, those two markets are the e-mini S&P 500 futures and e-mini Nasdaq 100 futures.  The emini S&P is always the ‘leader’ in the trade, where Roe will buy the SP and sell the Nasdaq when a long trade is signaled, and sell the SP and buy the Nasdaq when a short trade is signaled. The program will typically look to enter the position with a ratio of five long ES contracts to one Short NQ contract, but trades can also be entered at a ratio of 5/4 and 7/4.

    This SP/Nasdaq inter-market spread is what Mr. Roe calls his core position, and it is established at the close of the markets (3:15 CST) on days in which the internal models signal a high probability of a reversal the next day. This core position is held overnight, and the following day the systems will begin to evaluate market conditions on the open of the SP Futures (8:30 CST) and look to insert a stop in the market on the swing high/low of the first half hour of trading.

    This stop level will be used as the exit point for the core position if market conditions go against the trade. (Disclaimer: stop orders can no guarantee an order is filled at the desire price).  If the trade were to be stopped out, the position would be liquidated by selling the 5 ES and buying 1 NQ simultaneously (if long, vice versa if short).

    Here’s where it gets interesting – should the market move in the Core positions favor during the day, contrarian day trade models are employed that will look to trade around the core position, trying to take profits, then enter back into the core position at better levels than the previously executed day trade. These day trade models provide an extra opportunity for profit, but more importantly, they provide flexibility to the strategy and allow the manager to hedge his core position throughout the day.  The downside to it is that the market may not give the day trade model a chance to get back in line with the core position, leaving the program flat while the market moves in the direction previously held by the core position.

    The Jefferson Index Program trades the exact same models, as the Monticello Spreads program with the only difference being that Jefferson will only take the long or short ES futures position on each trade leaving the system 100% biased in the direction of the systems signal.  It does not include the Nasdaq hedge positions the Monticello program does. Because of this, the Jefferson Index Program can be considered the more aggressive of these two strategies.

    While 90% systematic, both programs also include discretionary risk filters designed to limit market exposure around major economic reports (FOMC, Unemployment, etc.) and over the weekend.  These were implemented in August of 2009.  Before major announcements the manager may choose exit the core position or cut back on the open position to cut down on exposure to temporary spikes in the marketplace that often occur in the first few minutes after an announcement on interest rates or otherwise.  The second filter is that Mr. Roe will force the program to close out all positions on the close on Friday and remain flat over the weekend.  The goal of this filter is to eliminate exposure to news based market moves in the Friday to Sunday news cycle (Friday to Monday if Monday is a holiday).

    With trades on either side of the market, its short term holding period, and day trades around the core position, both the Monticello and Jefferson programs trade a nit more frequently than your typical CTA, at about 7500 to 10,000 RTs per million.  On average both programs have winning trade percentage of approximately 60% with a ratio of 1.25 points made for every 1 point lost.  Therefore limiting slippage plays a big role in this systems success.

    From the risk perspective, the average risk per trade is 0.25% to 0.50% although max risk per trade is in theory unlimited, as the core position will be held overnight without a stop.  Mr. Roe uses fixed portfolio sizing and will trade 1 contract per $20k in the Monticello Program and 1 contract per $25k in the Jefferson Program.  There is a floor 5 ES / 1 NQ for Monticello and 2 ES for Jefferson.

    Attain Comments:

    The list of strategy types which have seen success over the past 16 months include short term traders (Dominion), spread traders (Emil Van Essen), contrarian [through short volatility] traders (HB Capital), and stock index traders (Paskewitz).  The Roe programs, being a short term stock index spread trading contrarian model, combine aspects of all of these varying strategies which have enjoyed success in the difficult trading conditions of the past 16 months, making it no mystery why they have been able to perform so well.

    But it is the fact that they were also able to perform in the volatile 2008 market environment which shows us there is something good going on here.  This is likely due to the manager’s philosophy to take what the market gives him, while also proactively managing risk.  Unlike other short-term traders who can be stuck in a losing position as the markets move against them day after day, the ROE programs will look to trade around positions and limit losses by employing intra-day stops (and via the Nasdaq hedge in the Monticello program).  The advantage of having a spread trade on is that it can help mitigate risk (especially overnight) and the manager can be more flexible with is trade size.

    Another feather in Mr. Roe’s cap is that he has been executing stock index systems professionally for the last 5 years, giving him a leg up (in our opinion) on what it takes to ensure that trades are filled efficiently with a limited amount of slippage.  He has experience working directly with GLOBEX and has access to best of breed order entry technology, which should work to his advantage given the number of trades put on each year by the programs.

    One thing that we are interested to see is how performance compares across the two programs moving forward.  In our conversations with Mr. Roe, he has mentioned that in testing Monticello has outperformed Jefferson on both winning trades and losing trades (7% to 5% on the upside, -4% to -5% for losing trades) as well as overall risk adjusted results in the back testing.  However, the Jefferson program has outperformed Monticello on both counts in real time trading.

    According to Mr. Roe, this is because the Nasdaq has been acting stronger than it normally does during the rally off the March 2009 lows, cutting back on the profitability of the Monticello program.  Mr. Roe expects that over time as the economy improves the spread between the ES and NQ will move back to a more normal distribution.  However, we could also argue that this phenomenon would also help the Monticello program during a drawdown phase, as the NQ would provide a better hedge on a losing SP trade.  We suspect that the performance difference between testing and real life could also be the result of higher than expected trading costs (slippage & commission) as the extra positions in the Monticello program are cutting into performance when compared to Jefferson.

    Items to take note of include Roe tending to have greater success during low volatility periods and potentially struggling in periods of expanding market volatility.  For example, both Roe programs outperformed their peers in the 1st quarter of 2010 when stocks drifted higher at a slow and steady pace. However, during the most recent increase in volatility the Roe programs remained near breakeven when many short-term traders put up big numbers.

    Another item is the discretionary filters Mr. Roe is using to limit market exposure. Any time discretion is used; there are concerns about it being used consistently. We understand Mr. Roe’s theory of going flat over the weekend and before major economic reports as it helps limit risk and maybe improve the investors psyche. However, coming from a systematic background we also know that this type of scenario can be very hard to test and there is a chance that exiting trades early and before the weekend is doing more harm than good. At the very least, we wish that the filters being used were more quantifiable.

    Lastly, as with most emerging managers – there are risks with Roe concerning his being a small operation at this point.  When Roe is the name is on the door, what happens if Mr. Roe is sick, goes on vacation, and so on? These are often the tradeoffs between an accessible program, however; and one with a $5 Million minimum.  And Mr. Roe assures us he will look to put money back into the company and build up his staff and back office as his assets under management expand.

    Overall, we are excited to follow along and work with Mr. Roe as he grows his CTA.  Thus far, he has proven to be an adept manager in some of the toughest markets conditions conceivable for a short-term stock index CTA.  Growing pains are a frequent problem for emerging managers; but in seeing Mr. Roe’s eagerness to learn of potential problems before that happen and work with firms like Attain to iron any possible issues out – we are confident that Roe Capital will make sure all decisions have their customer’s best interests in mind as they continue to grow their business.

    IMPORTANT RISK DISCLOSURE
    Futures based investments are often complex and can carry the risk of substantial losses. They are intended for sophisticated investors and are not suitable for everyone. The ability to withstand losses and to adhere to a particular trading program in spite of trading losses are material points which can adversely affect investor returns.

    Chart of the Week : Roe Capital Performance Summary

    Futures based investments are often complex and can carry the risk of substantial losses. They are intended for sophisticated investors and are not suitable for everyone. The ability to withstand losses and to adhere to a particular trading program in spite of trading losses are material points which can adversely affect investor returns.


    Past performance is not necessarily indicative of future results. The performance data for the various Commodity Trading Advisor (“CTA”) and Managed Forex programs listed above are compiled from various sources, including Barclay Hedge, Attain Capital Management, LLC’s (“Attain”) own estimates of performance based on account managed by advisors on its books, and reports directly from the advisors. These performance figures should not be relied on independent of the individual advisor’s disclosure document, which has important information regarding the method of calculation used, whether or not the performance includes proprietary results, and other important footnotes on the advisor’s track record.