Forex Manual Back Testing – Candlestick
Manual back-testing is the most readily available form of back-testing available. Most traders understand it is possible to manually back-test their trading systems, but most traders choose not to manually back-test. Many charting packages make it easy to back-test. The vast majority of charts will allow the historical price, so that you may advance price slowly and “trade” the chart as it unfolds. For example, in Meta Trader 4 you simply hit the F12 key on your keyboard to advance the charts one price bar at a time. This is all that is needed for manual back-testing.
C A N D L E S T I C K
The candlestick is a popular chart that displays the opening price, the closing price, the high price and the low price for a market during a given time period. Each candlestick clearly represents the important market activity for the given time period.
To manually back-test, you simply scroll back in time and record your trades, the trades you would have taken had you been trading the chart in live market conditions. You can advance the candlesticks slowly, one at a time, record your entry price, the number of lots traded, your stop loss, and your profit target. You may be concerned that manual back-testing involves a lot of notes, spreadsheets, and record keeping. It is a meticulous, involved, and laborious method of back-testing. It is also extremely powerful.
If you aspire to trade by looking at a chart and making a trading decision, you are nearly duplicating the trading process with this form of back-testing. Most traders employ discretionary trading systems, so it follows that manual back-testing is the most appropriate form of back-testing for most traders.
Obviously manual back-testing will take some time, and it is sometimes difficult to avoid “cheating” with this type of back-testing. However, you must take care to avoid going forward on the chart and then reversing back in time to take a trade you would have taken. The key is to trade as if you are in that moment in time, with no information of the future. The experience gained with manual back-testing is priceless. You can quickly accumulate experience with your trading system if you back-test correctly.
Manual back-testing will yield statistics to help you understand the nature of your trading system. These statistics, as we will see later in this book, will become invaluable for determining how you should trade as they may be utilized to project your results into the future. A note about manual back-testing—it may seem tedious, it will take some time, you may want to give up when it progresses slowly, particularly when you are looking to accumulate hundreds of trades. Resist the temptation to avoid manual back-testing. Make progress in small chunks, an hour of testing per day can give you a huge advantage. The payoff is great and your experience testing your system will allow you to gain “experience” trading the system through different market conditions. Each trade during manual back-testing will march you closer to expertise. This expertise may be the difference between you abandoning your trading system during the inevitable drawdown, and maintaining confidence in your trading system through a drawdown. However, there are pitfalls associated with manual back-testing.
The most common pitfall traders fall into with this testing is engaging in future trading by advancing the chart, either by accident or intention, and then deciding that a trade would have been taken. Traders at times have difficulty discerning when future price data creeps into manual backtesting. Traders who are strict about not allowing trades after future data is viewed will avoid contaminating back-testing data. By trading only from rent candlestick on the chart, serious back-testers avoid going forward in time.
Hindsight bias is a critical killer that may creep into your manual backtesting if you allow it to happen. A good rule of thumb is to advance the charts slowly during your testing, and if you accidentally advance the chart too quickly, you must keep your trades to the right-hand edge of the chart. If you advance the chart and then go back to take a trade, you are leaving yourself open to the hindsight bias.
H I N D S I G H T B I A S
The tendency to overestimate the predictability of the market after the future outcome is known.
Even if you only take trades during testing by advancing the chart one candlestick at a time, hindsight bias may still sneak into your testing. If you are testing historical data on the EUR/USD in 2008, and you had experience trading in 2008, you may have a problem. Your subconscious may be pulling up the 2008 EUR/USD chart and you are obviously not even aware of it. To fight hindsight bias, take trades on the right-hand edge of the chart, do not go forward on the chart and then reverse back, and if you do accidentally go forward, resign yourself to skipping over any signals that pop up and stick to trading on the right-hand edge of the chart.