Since January 2000, SP Trader management has developed a computer online trading system which was back-tested and continuously improved using 30 years of historical data. The proprietary system has a track record from Jan 2000 of actual historical returns trading futures (see Monthly % to Date) and is capable of trading hundreds of futures contracts in dozens of markets - equities, currencies, commodities, energies and treasuries. The systems constantly monitor all associated micro and macro risk factors and through technical analysis and online trading electronically initiates buy and sell orders, therefore investment decisions are not dependent on fundamental data. Should there be a major market reversal due to new factors in the market; the program reacts instantly to adjust to the new environment, thereby minimizing losses while increasing profit potential.

From the standpoint of the portfolio, it is important to remember that the standard deviation of the individual strategies should be non-correlated to provide a lower standard deviation of the portfolio as a whole.

 

 

At the core of many of our software trading systems are relational strategies between symbols, such as Crude Oil (CL.P) to the S&P 500 Futures (ES). In our view trading systems should be based on fundamental truths such as the relationships between certain commodities, the indices and the dollar. These relational studies work in the background to affect the trading data and cause entries (or exits) at the closing or opening of certain bars, depending on the time frame of the study and the strategy. Over time, these types of fundamentally based strategies prove to be the most robust and are not simply form-fitted to some past filtered historical returns and data. In fact the rules are so fundamental, that little or no optimization is necessary and successful blind walks of a year or more are suggestive of their forecasted futures trading behavior. In our utilization of these strategies, actual live trades have confirmed that the percent profitable trades, the ratio of average winning to losing trades, monthly standard deviation and maximum adverse excursion percent (and absolute) have conformed to their past historical behavior as well as to their blind walks. (These strategies are the intellectual property of SP Trade Investments Capital Ltd. and we do not license them to third parties.) Critical elements of the online trading campaign include risk management and asset allocation procedures. The strategies are non-discretionary and use clearly defined 'stop-losses' and profit targets upon entering the market. The systematic approach is designed to capitalize on the relationships between dual symbols as well as on the deviations from randomness that market price series exhibit.

On average, the systems position the SP Trader Investment Futures Growth Fund and other SP Trader Products with 10-12 Futures trades per month. It does not scalp trade. It takes positions of 1-3 day swings based on technical criterion which is programmed into our proprietary system. The criteria include proprietary formulas for determining short term oversold and overbought market conditions and is based on a well-defined trading methodology utilizing a large number of rules. It has traded profitably since January 2000 as you can see by the historical returns: Monthly % to Date. It was designed to find large winners, while suffering small losses and yet be highly accurate. The program was designed to eliminate many of the outside influences or emotions and other mistakes investors and traders tend to make. It has no emotions about whether the market goes up or down. Long and short trades are split almost evenly.

The systems are currently set for a short term to midterm time horizon. One key to its success is to limit draw downs by daily maintenance of stop market orders. In this way, if a market reverses, the loss is severely limited, while if trend continues, profits are protected and winning trades are optimized. This boosts substantially the winners to losers’ value ratio. Risk and asset management plays a key role in our financial strategies. Our proprietary program limits the risk per trade to a maximum of X% of total portfolio assets applied. In addition the system continuously screens market volatility and adjusts portfolio exposure accordingly. Every newly opened position is allocated a market stop loss which sets the point at which the online trading position will be closed. This market stop loss is continually and automatically controlled. We do not do option trading but only futures trading which permits us to control stops and profit takes precisely.

 

System development occurs at two levels: 1) individual trading strategies, and 2) system integration. Individual trading strategies utilize a variety of technical approaches that identify recurrent patterns, trends, countertrends, correlations, etc. at daily and/or intraday time frames. Stringent back testing procedures are used to quantify confidence limits on performance statistics considering the back test duration and number of system parameters to avoid over optimization. All systems are symmetrical with respect to long and short trades.

System integration involves the selection of individual strategies to be utilized in the final trading methodology and allocation of capital among the strategies. Money management rules are explicitly incorporated into actual historical trades but also utilizing 30 year back testing in such a manner that draw downs can be computed that correspond to start-trade amounts (i.e., the worst drawdown is assumed to occur at the outset of trading such that it is not attenuated by prior gains). Pooled system mean monthly and annualized returns, percentage start-trade draw downs and drawdown durations are computed for various risk allocation strategies using the following approaches:

Composite Method - This method computes month-by-month composite performance over the back test period as the position size-weighted sum of individual system monthly performance. This method has the advantage of naturally accounting for actual system performance correlations, but it does not provide any information on the probability of more adverse draw downs than those encountered over the test period.

Monte Carlo Method - 10,000 year Monte Carlo simulations were performed using composite system statistics (mean, SD and skew) from Composite Method results yielding a probability distribution for integrated system performance. For each of the above methods, two scenarios were considered: 1) using best estimates of individual system performance from the raw back test results. The development approach and 2) using adjusted back test results corresponding to 95% probability lower confidence limits of mean monthly returns (i.e., there is a 95% probability that future mean returns will be higher than this) considering the duration of the back test period and the reduction in degrees of freedom due to system parameterization.

The integrated trading system is designed such that best-estimates of start-trade draw downs with a 1-in-100 year probability of occurrence are less than 20% and values computed using lower confidence limits of returns are less than 30% for a 1X leverage account, while the ratio of mean annualized return to drawdown is maximized. The development approach results in a trading system that maximizes returns while explicitly controlling downside risk at a given level of capitalization. The foregoing methods will be used to monitor system performance and to make adjustments (e.g., add/delete individual systems and/or change risk allocations) as capitalization changes.

Following is a description of one kind of non-relational, straight-forward Bollinger band based system which we utilize (amongst others) to trade S&P 500 Futures contracts:
Look at the 30" Chart above. If you click on Trades 2004 in a different browser, you will note that trade number 77 had an Entry Long at 1078.00 on May 12, 2004 at 13:00 with an Exit trade on the same day at 16:00 at 1092 or a gain of 14 S&P points. This is the first trade with the blue entry point and blue line on the chart above on this page. The trade was signaled because the system recognized that, among other things, on a short term basis the market was very oversold at an important support level, price divergence was high, the tick count divergence was extreme and our displaced moving average, which contains 97% of all historical S&P price action, was triggered. On the next leg up (trade 78 in Trades 2004), the system reversed because, among other things, the simple moving average was breached and a short signal was triggered which turned out to be a 5.50 Point loss. On this unusual day in which 3 trades were signaled, the gain was only 2.25 points because two of the trades were stopped out. The first because the system entered the long too early at the displaced moving average and the second because the trend continued more strongly then the system anticipated. This was an unusual and not very good day but one which we present to demonstrate one extreme of the system at its less than perfect. Nevertheless, it did make a profit for the day.

A more typical day would have been the next day, May 13, trade number 79 on Trades 2004. The market opened gap lower but strongly trended higher for the morning reaching an important resistance at 1102.50 before reversing for the rest of the day. The system triggered a short signal at 12:00 at 1097.75 and an exit the next morning at 1089.50 for an overnight gain of 8.25 Points. Overall, we find that overnight positions from extreme overbought or oversold conditions are worth their weight in gold.

The last trade on our chart above was very similar to the one above, with a lower opening on May 14 followed by a move up and then a fall from difficult resistance near 1100. We entered at 1099, near the top of the market and held overnight and exited on May 15 for a gain of 16 points.

During the week of May 12, 5 trades were signaled, three were profitable and two were not. This was a less than average ratio of % winners to loss trades as our Historical Returns is closer to 70% profitable or closer to 2.5 to 1. Nevertheless, it was a profitable week in which we gained nearly 21 S&P points in a few days.

For more information about Moving Averages, Resistances and other tools that we utilize, click Fundamentals.

For more in-depth information about numbers, performance and graphs, visit the navigation bar to the right - Why SP Trader or Historical Returns - which compiles an in-depth rigorous analysis of the summary performances, trade analysis, monthly analysis, performance graphs and trade graphs since 2000. For more information about the SP Trader Investment Futures Growth Fund, click Questions. For how the Fund ranks worldwide, click Ranking.

If you have any queries, comments or suggestions, please don't hesitate to Email us.

To Subscribe to the SP Trader weekly Free Newsletter, go to the right nav bar above, enter your email and click "go". To see our Newsletter Archive click Newsletter.

To subscribe to the Fund, please click Open an Account.