Having a quantifiable strategy (Backtesting Data) that can be tested over real, prior market movements can prove to be an invaluable tool. Possessing the ability to see how a strategy may have behaved for a specific time period or for a particular instrument can provide an avenue for iterative upgrades in a fast-changing trading ecosystem.
However, back-testing a strategy does not necessarily take into account all of the variables that are involved in trading. What may seem like small differences or minute discrepancies in how a strategy is handled can produce vastly different performance results. While back-testing results certainly provide a clearer picture of a strategy’s executions, one should never assume that they will see the same results in live market conditions.
First and foremost, market conditions are constantly changing and shifting, potentially invalidating strategies that rely on unique conditions that are only present for short periods of time. There are many inefficiencies in the marketplace that are never taken advantage of simply because they are not present long enough to be discovered. This can also mean that a strategy that performed admirably during the early 90’s can be identical to a strategy that shows poor results in today’s market conditions.
Another point to consider is the actual back-testing engine itself and whether you fully understand the limitations of the software as well as how the results are calculated. This will avoid wasting time back-testing a system that is not calculating correctly in the first place due to unforeseen bugs and/or quirks that skew results one way or the other.
Commissions, execution speed, and volume are also factors that can significantly affect trading results. Imagine a strategy that was profitable when executed at the VWAP (Volume-weighted average price), but produced mediocre results when executed at a standard deviation below or above VWAP. The depth available for different market prices can be a primary factor during system implementation.
The purpose of this article is not to undermine the important of backtesting, but rather to educate you on all the variables and inconsistencies when reviewing hypothetical results. Forward-testing, trial period testing, period reviews of results, and a complete understanding of the strategy components and how they behave on an on-going basis can provide peace of mind when managing a portfolio.
Trading futures and options involves substantial risk of loss and is not suitable for all investors. Past performance is not indicative of future results.
This matter is intended as a solicitation to trade futures. VFM 29662
If you wish to examine trading results that combine all the variables that automated trading should take into consideration, we can help you choose a strategy based on your trading capital, markets and type of trading