This is a common question new traders ask, but there is no single simple answer. This post details how to know exactly how long you need to test your trading system.
As a general rule, you should backtest your trading system as long as you feel confident trading it with real money.
In other words, it depends on several different factors.
Let’s take a closer look at what you need to know about backtesting and how long you should backtest.
It all starts with defining terms…
Definition of “long”
The first thing we need to clarify is what is meant by “long”.
There are many different interpretations of this word, so here are the definitions that we will be discussing in this post.
- time spent on backtesting
- Amount of historical data used
- Number of trades backtested
Looking at the three definitions, we can see that the word “long” has very different meanings.
So, to clear up any confusion, I’ll explain each one and show you which ones are really important.
time spent on backtesting
As with many things in life, the amount of time you spend backtesting is not an indicator of the quality of the work done during that time.
So the idea that you have to spend a minimum amount of time on backtesting is useless.
As an extreme example, you can backtest for 4 months, but only have 2 backtested trades.
It doesn’t tell you anything about how profitable your trading strategy is.
So, when you are looking for minimal backtesting to prove your trading strategy, you should abandon the idea that you have to do the act of backtesting for a period of time.
Amount of historical data used
Another way to measure the amount of backtesting a trader does is to track the amount of historical data used in testing.
This means you can backtest your system with 3 or 30 years of historical data.
Clearly, more data is better.
When you have a lot of historical data, you can see how your trading strategy has performed in different market cycles.
This is very important in finding out if you have a strong trading strategy or if your trading strategy works only in certain market conditions.
For example, if your trading strategy wasn’t tested during the 2007 financial crisis, you never know what might happen in a highly volatile market.
So while the amount of historical data you use to backtest is an important factor in determining valid testing, there is one more thing to consider…
How many times should I backtest my trading strategy?
That brings me to the final backtesting measurements…
The most useful metric to watch is the total number of backtested trades.
Testing a trading strategy with 300+ trades is much more reliable than testing the strategy with 10 trades.
It’s clear.
But less obvious is the minimum number of trades to make you feel confident in the system you are testing.
There can be gray areas where you are not sure if you have enough trades in your test.
When in doubt, zoom out and consider the big picture.
Consider a day trading strategy…
If you are only testing 100 trades, you may only have 2-3 months of data. It’s not enough to decide if a strategy is credible.
Even 500 trades may not be enough.
You don’t necessarily have to test every day for the last 20 years, but you should have a sample of different market conditions.
There should be a deal from:
- volatile market
- quiet market
- trend market
- news shock market
The more trades you make from each of these types of markets, the more confident you will become in your strategy.
Then consider your swing trading strategy…
A swing trading strategy may have a relatively small number of trades in the backtest.
For example, let’s say you backtested the RSI Divergence trading strategy on the EURUSD daily chart.
With this trading strategy, you may only get 25 trades in 20 years.
Does that mean that even if you make a profit, you abandon that strategy?
of course not.
However, you may not be confident enough to trade it live.
In that case, it’s important to forward test the strategy until you reach a point where you’re confident it works.
Forward testing will take a while.
But if you only have a small amount of data, this is the best way to gain confidence.
You can also test strategies on other markets and timeframes.
If we find other instances where the strategy works, we can further validate that the strategy is reliable.
100 trading myths
There are rumors on the Internet that a minimum of 100 backtests are required to prove a strategy works.
it’s simply not true.
100 trades is a good number of trades.
But again, the number of deals in our tests only tells part of the story.
Here’s what you need to know about the 100 times trading myth…
The biggest problem with setting a blanket minimum number of trades is that it does not take into account the timeframes tested and the specific market behavior.
Some statistics professors may say you need a minimum of 30 trades to prove a trading strategy works.
Others will say 500 transactions.
To be fair, this is a good place to start from a strictly mathematical point of view.
But in reality there is a lot of volatility between markets, market cycles and trading systems.
What works in one market or cycle will probably not work in another.
That’s why you need to be aware of all factors, not just the number of trades.
final thoughts
That’s what you need to know about how long you should backtest.
There is no single simple answer to this question.
On the surface, you should make as many backtested trades as possible over the longest period of time in the past.
However, nuances should be taken into account.
You should consider your trading strategy, your goals and your level of confidence in testing.
However, once you start backtesting, you will know what is reliable and what is not.
let’s start.
Learn more about backtesting in these posts.