It is estimated that more than 70% of trading is done by algorithms, and most fundamental-based traders are not entirely discretionary. Instead, they approach their trading with the assistance of support and resistance levels and technical as well as fundamental analysis.
Many invest significant time and money in developing quantitative analysis. Whether or not they pull the trigger themselves, there is usually some systematic element to their decision. However, the markets have a quality of being complex and dynamic. A change in one or more key variables could cause indicators to brea down. Trading systems and technical indicators work until they don’t; it is difficult to quickly adjust faltering systems. This raises the question: What approaches can a trader use to detect if a system or indicator starts deteriorating?
How healthy is your trading system? When you go to the doctor for our annual visit, the doctor doesn’t tell you how healthy you are. Instead, key metrics are measured: Heart rate, blood pressure, cholesterol levels, etc. The results provide a rough map of your physical performance and whether it is in a statistically normal range. We can apply this concept to evaluating the health of trading systems.
There are established categories for evaluating trading system performance. “The Evaluation and Optimization of Trading Strategies,” by Robert Pardo (see “Trader’s report card,” right) provides several metrics to analyze a trading system.
We use various metrics to evaluate system performance to attempt to answer: How fit is the trading system? Fitness goes beyond profitability. We need to examine each trading signal in the context of the price excursion during the trade.
A crucial metric to help identify where to improve trading systems or rules is to quantify unrealized profits. If you go long EUR/USD at 1.2100 and close it at 1.2150 the result is a 50-pip profit. Most would consider this to be a good trade. But could that trade have been more profitable? Maximum Favorable Excursion (MFE) measures the highest value path taken by price between the opening and closing of the trade. In the 50-pip winning example, if the EUR/USD had a maximum value of 1.2190, the trader or system left money on the table. Diagnostic questions arise: Why did the system miss reaching the Maximum Favorable price point? How often does that happen?
Conversely, consider a trade where a loss occurred. The trader goes long EUR/USD at 1.2100 but closes it at 1.2075. The trade loses 25 pips. When the price during the period of the trade actually went further down to 1.2050, the Maximum Adverse Excursion (MAE), which is the lowest price point achieved between the open and close period. In this case the system was deft enough to not allow the loss to go to its maximum unrealized extent. This shows that winning and losing percentage does not tell all. Many successful traders believe the key to success is the ability to take small losses.
Trading systems could improve their performance by reducing unrealized gains and increasing unrealized losses. A system that shows a large amount of unrealized gains is most likely in need of a tune-up.
One of the most important measures of a trading system’s fitness is whether profits are turned to a loss--a cardinal sin of trading. Reducing the frequency of profits turning to a loss could involve better setting of limits as well as stops.
No system is perfect. Some depend on having a significantly higher percentage of winners to losers for success. Others may produce as many losing trades as winnings trades, but the winners typically are significantly greater than the losers. In either case, it is important to monitor metrics to ensure the system is maintaining its edge.
Traders need to keep improving their evaluation of their trading approaches and systems. Beyond current standards of evaluation, optimizing system performance should include the following price excursion measures:
- The amount of unrealized profits;
- The amount of unrealized losses; and
- The percentage of trades where profits turned to losses.
These are important because every system will go through hot and cold streaks. The key to success is limiting the losses during cold streaks and maximizing the profits in hot streaks. Sometimes a system is performing well during a drawdown or failing to fully exploit winners. Metrics give traders a better picture of what is going on so they don’t mess with a system that is performing well during a cold streak or fail to notice it underperforming during a hot streak.