We’re constantly analysing our trading performance on a short term basis, we can clearly see it in our rolling profit and loss statement, or at the end of the day (EOD). Over a short period of time, after our introduction to forex trading, we’ll also become proficient at using the back testing mechanisms on our platforms as part of our overall analysis, to establish if a potential strategy has what we’d term “positive expectancy”. However, once that back test moves from demo to live with our actual money at risk, we tend to abandon one crucial aspect of the back test protocol that encouraged us to move ahead with our initial live test; the timescale over which we’ve judged our back test.
Once we move through the gears; from demo back test, to demo forward test, to live trading, we should still (initially and in theory) be in test mode; risking less than we’d normally risk during our normal live trading. For example; rather than trade full lots we should trade mini or micro lots, until such time we have enough confidence and evidence in our strategy that it’s working. However, once we establish that the strategy does in fact work, then we tend to avoid placing a time scale on the outcome of our live testing. We’ll have back tested in demo going back a few months if it’s a day trading strategy, or if it’s a swing trading strategy, we’ll back test over recent years. And yet during actual live testing, we can be guilty of failing to attach a reasonable period of time over which to judge our strategy. We tend to become completely invested and focused on the bottom line only.
Once we move to live trading we must attach the same metrics to judge our strategy’s performance as we would in back testing, despite this representing a significant challenge if you’re a swing trader. If we’re a day trader we may take three trades a day, therefore we could analyse our live strategy over an approximate six – seven week period; fifteen trades per week there’s our 100 forex trade analysis. During this time we should have experienced many of the regular high impact economic calendar events and perhaps some of the political issues, which can dominate the behaviour of many currencies and their relationship with their peers. Technical issues may have also become apparent as our chosen currency pair’s (or pairs) price may have reached certain technical points, such as the 100 DMA and 200 DMA critical handles (round numbers). What we can’t do is forward test a swing strategy over years, it’s simply not possible.
We may be able to back test/analyse our swing trading strategy over past years, but with forward testing we have to take a punt on the projected performance. As we begin to learn quickly, markets range more than they trend, some analysis suggest ranging accounts for 70% of forex movements. With regards to swings we may only experience a few each year, therefore, as an example, we may be back testing only twelve swings over the past three year period. Is this enough to judge a swing trading strategy’s performance? Opinions differ, but it’s important that we recognise this critical difference, versus the efficiency of back testing scalping, or day trading strategies.
Back testing is only as good as forward testing. We can (relatively) quickly forward test in live market conditions our 100 forex trades if we’re a scalper or day trader and perhaps that’s where we should concentrate our efforts, when using backtesting and forward testing.