Backtesting is one of those trading terms that we talk about in slightly hushed voices. If you asked 100 traders to pick the worst part of trading, many would say backtesting, but almost every trader understands how important it is. Whenever I talk to new traders about learning or developing a strategy, the conversation always includes backtesting. And one of the most common questions is, ‘Well, how long is this going to take me?’
Every trader wants to see results; that’s no secret, and the sooner you can get started, the better. That’s where any kind of testing becomes a pain because you don’t want to spend days checking past results when you could be live trading and making money, right? But on the other hand, you want to make sure your strategy works or you’re never going to make money. So where’s the sweet spot?
I’ll spoil the article slightly here; there is no perfect solution to how long you should spend backtesting. It’s not a ‘one size fits all’ approach, but there are some guidelines you can follow. In the next part, I’m going to talk through what I teach our students when it comes to backtesting, and hopefully, this will help to answer that age-old trading question.
Let’s get started!
In this article, we’ll cover
- What backtesting involves
- How many results do you need for a backtest
- How many backtests should you do to check a strategy?
- How long should this process take
Before we start – every trader is uniqueand learns at a different pace. We’ve coached thousands of traders in the past five years, and we know the value of working with you as an individual trader. This article is a general guide to backtesting and the number of results needed, and if you’d like to find out more about how we could help you, there’s a link at the end to get in touch with our team.
Contents
What Is Backtesting?
Let’s start with a bit of background info. Backtesting is a technique for assessing a trading strategy’s performance by applying it to historical data. It’s an essential step in verifying that your strategy has the potential to be profitable in real market conditions. Put simply, you check through old chart data to see how your rules perform, which can give you some confidence that your strategy works. Of course, past performance isn’t a guarantee of future results, but it definitely helps to see previous market behavior.
How Much Backtesting Data Do You Need?
To understand how long we need to backtest, we can look at the number of results. The amount of data required for backtesting can vary greatly depending on your trading style and the robustness you seek. For day traders, a few months of minute-by-minute data could be enough, while long-term investors might look at several years of data to understand how the strategy performs across different market conditions.
What this means is you need to tailor your backtest to the strategy you are testing. If you only get one entry per week, then one month of data isn’t enough. At the same time, if you’re getting five or six entries per day, then gathering data for the past five years is slightly excessive for a first backtest.
Is 100 Trades Enough for a Backtest?
Ideally, 100 trades could provide a glimpse into a strategy’s effectiveness, but it may not be enough to validate a strategy with high confidence. More trades mean a larger sample size, which can lead to more statistically significant results. It’s important to make sure diverse market conditions are included within those trades to see how a strategy works under different conditions.
Is One Year of Data Enough for a Backtest?
Using one year of data for backtesting can provide insights into how a strategy would have performed over the past year, but it may not account for varying market cycles or unusual events. Extending the backtesting period to cover multiple years can help capture a broader range of market scenarios and improve the reliability of the backtest results.
How Long To Backtest For Day-trading
Let’s look at a backtest for a day-trading strategy. This means we’ll be seeing one or multiple entries on most days, potentially across a few different markets. If that’s your kind of strategy, here’s what I would do to backtest it.
First Test – Three Months
To begin, my first backtest of any new day-trading strategy is a three-month check. This first test is all about making sure that the strategy is easy to follow, shows promising results, and understands if there’s anything I need to change at that stage. This might be adding new rules on specific entry times if there’s a strong correlation between markets I’m trading or other items. If that test comes up with a positive R number, then I’ll move on to stage two.
This test is usually completed in less than one day, with some time to analyze the results. A good trader should be able to backtest this in a few hours with a bit of experience. It’s always best to take your time and make sure you don’t make mistakes though, so don’t rush a backtest if you’re new to this process.
Second Test – One Year
This second part is a more comprehensive test, looking at a full year of data, often on multiple markets. At this point, I’ll be gathering more info per trade, too, which will focus on items like the MAE and MFE numbers, the impact of using different entry points, and whether to use a trailing or fixed stop loss, to name a few examples.
This process usually takes me a week to complete, including checking. This can vary, though, particularly if I spot anything while backtesting that I want to double-check. Overall, this is the backtest where I’ll decide if the strategy is worth forward testing or if it needs more refinement
Third Test – Forward testing for 4-6 weeks
The next stage isn’t really a backtest, but I wanted to mention it anyway. If you’re happy with the strategy performance over a year, then you need to test it on a live market to check everything works in real-time. This forward test (on a demo account preferably) will give you a chance to see how easy it is to set up and enter the trades, and if you give yourself 4-6 weeks, it’s enough time to see if the results mirror your backtest.
What’s next?
If you want to find out more about backtesting, we have a couple of guides on our YouTube channel to help:
How we can help you with strategy backtesting
Let’s be honest – even with a plan and structure, backtesting isn’t easy. A great trader needs the patience to collect data and the experience to check for mistakes and improvements. That’s a lot of work, and while the results can be well worth the effort, it’s tricky to get started.
This is one of the reasons why we talk about data collection and provide our backtest results in our trading program. Over the years, I’ve completed over 50 long-term backtests of WB strategies, my own strategies, and also checking work from many of our traders.
If you’re looking to get support with backtesting your strategy – or you’d like to skip the backtesting altogether and learn a strategy that’s been tested and traded for years already – we can help in both cases. If you’d like to learn more abouthow we trade and if we can help you with your trading, you can get started with the join button below. Alternatively, check out our articles on learning to trade different markets, linked at the bottom of this page
