7 Mistakes Traders Make When Backtesting a Day Trading Pattern

7 Mistakes Traders Make When Backtesting a Day Trading Pattern

7 Mistakes Traders Make When Backtesting a Day Trading Pattern 1024 546 Steven Dux

Intro.

What is backtesting, and as a day trader, do you really need to know about it? Do you need to do it? Do you need to have it become part of your trading strategy? The simple answer is YES!!

And in this article, I’ll explain why (as well as how you can!)

In one of my previous articles, we discussed how important it is for every trader to backtest their day trading pattern and strategies.

It’s the moment of truth every trader waits for. Whatever the result, it’s a good way of determining how your day trading pattern will fare in the live market (to a certain extent). But like anything in the trading world, no method is capable of predicting a future pattern. Running these simulations can be quite helpful but if not implemented correctly, it can hinder more than help. You need to approach your backtesting with utmost care and logic.

There are some mistakes that even the most seasoned traders are prone to, so if you can make smart decisions while being aware of these common mistakes, backtesting can be a rewarding research tool.

What is Backtesting

To put it simply, backtesting is a method to see how well a day trading pattern or strategy could work based on historical data.

It’s an essential research tool used to stress test either current trading strategies or to find new ones. Backtesting can be used on Excel, but it is usually done with code, using programming languages like Python, C++, and R. This is why knowing how to code becomes an asset while trying to backtest because it gives you enough flexibility to run any kinds of tests you want. But if coding is not your strong suit, then some Backtesting software such as Stockcraft, Tradingview, TradeStation, NinjaTrader can be helpful.  

This type of software, while very advanced, isn’t really as flexible as just using libraries from a high-level coding language.

This is precisely why professionals tend to prefer them over this software. But having said that, they can still become essential for new or amateur traders. At first, Backtesting can feel like magic. But what many don’t realize is that simply running some parameters against market data and tweaking them to get a positive sloping equity curve, isn’t actually Backtesting. If it were that easy then everyone would be a pro back tester with foolproof trading strategies that could make them billionaires.  

That is why knowing how to actually Backtest with a logical yet cynical outlook while keeping the common mistakes in mind is what matters when trading in a live market.

How Does Backtesting Work

The obvious first step to start Backtesting is to define your day trading pattern or strategy.

What is the strategy you plan to use in the open market?

Is this strategy feasible for you and your lifestyle?

Do you understand the conditions and risks of your strategy?

These are some of the questions you need to ask yourself before defining a trading strategy. Once you’ve done this, you need to make sure your strategy is formulated and not just random words with a vague meaning put together. A strategy can be said to work if it leads to a win over 70% of the time (with at least 20-25% profit returns). These are numbers I have arrived at after immense practice and experience. Figuring out this number involves the actual Backtesting. After all, you don’t know what works until you test it. As mentioned before, this can be done with code or software. But the idea in both essentially remains the same. 

Let's say you formulated a strategy that you assume will work over 65-70% of the time and want to verify that it actually will.

There are a few steps you will take to perform the test:

You begin with selecting a market and the time period you want to backtest in.

Then, you would plot the important trading tools and indicators.

Observe the setup, be aware of the potential mistakes, and record the results of the trade.

From here, you can move the chart forward and keep repeating the previous steps till you have a bunch of results.

Finally, with all the results of your backtest in your hand, you journal your trades. This includes noting down the setup and other essential details of your trade such as the price in, price out, stop loss, and others.

Why Is Backtesting Important

Backtesting is one of the most important aspects of the process of developing a trading system. If a trader interprets it properly, they can use it to improve and optimize their day trading pattern and strategies. It can also help in finding technical or theoretical flaws and give you confidence in your strategy if the results seem positive.  

Some reasons why testing strategies is important: 

1.

Check The Performance Of A Strategy

To call a strategy useful, we have to be confident of its performance. We have to estimate its profitability! A strategy has to be at least mildly profitable for it to be called useful and yield profits over 20-25% of the original amount invested. If it provides more losses than wins (consistently), the strategy is low on performance and should be discarded.  

2.

To Gain Confidence In The Strategy

Unless you’re an old trader with years of experience, you have to backtest a strategy to be confident about it. Only when you know about its performance should you feel confident using it in the real market. Before you start to trade in the real market, make sure you know how your strategy performs. Without testing, you’ll have no idea what the results could be. 

3.

To Make Sure We Get Consistent Results

Often we might think a strategy is working, use it in the real markets, but after a while, it starts to produce a series of losses. You figure the strategy isn’t working anymore, start to look for a new one until that also gives you losses and the entire cycle repeats again. Sometimes, the strategy is still working at its optimal capacity and is just going through a stream of bad trades. But more often than not, the strategy is flawed and needs to be optimized to perform effectively.  

4.

To Make Sure We Get Consistent Results

Often we might think a strategy is working, use it in the real markets, but after a while, it starts to produce a series of losses. You figure the strategy isn’t working anymore, start to look for a new one until that also gives you losses and the entire cycle repeats again. Sometimes, the strategy is still working at its optimal capacity and is just going through a stream of bad trades. But more often than not, the strategy is flawed and needs to be optimized to perform effectively. 

This can be avoided if you test your strategies thoroughly before using them.

/.

Consistent profits are the only way you can sustain yourself in the game because consistent losses can very quickly wipe your account.

To ensure consistent profits you have to use the right strategy that works consistently and in multiple situations. If you know what you’re doing, use the right strategy and know your patterns, you can make a lot of money in day trading— and that’s how you achieve financial freedom. 

What Are The

Mistakes Traders Make When Backtesting

As mentioned before, there are some common mistakes my students (and many beginners) make while Backtesting. Being aware of what these are is the first step of avoiding doing them. In my experience, there are three types of students…  

/.

The first are the ones who don’t put in an effort. They don’t put in the time to learn. That itself is a big mistake so their journey in trading ends pretty quickly.

/.

The second type of students are the ones who make an effort, discover what to do and not to do, and because of this, they can control the market. They pour in a lot of their money before they’ve had a couple of years of trading experience, leading to a massive loss. This destroys their confidence and it’s really hard to come back from.  

/.

The third kind of students are the ones who know what to do, are very disciplined, and so are consistently profitable. The big mistake they end up making is succumbing to greed.  

1.

Overfitting

Overfitting (also known as over-optimization or curve fitting) in trading is the concept where the designed trading system adapts too closely to the historical data. Meaning, your strategy is showing you that it’s profitable even when it might not be. Any results that have been overfitted will never be valid in the future, so it essentially makes the whole Backtesting process invalid.  

The important thing to remember to avoid overfitting is that the past can never predict the future in the financial world. No one can. Strategies that have been adopted too closely to the past data won’t be flexible enough to adapt to the future too.  

2.

P-hacking

P-hacking is also referred to as data dredging, data fishing, data snooping, and data butchery (it also closely relates to the previously mentioned overfitting). P-hacking is said to happen when a trader finds misleading patterns in any given data. The trader or analyst will test various patterns but then only focus on the ones that gave positive results while ignoring any negative ones.  

This is a common problem in many scientific studies too. They forget to note that correlation doesn’t equate to causation and that’s why many end up making this Backtesting mistake. A simple way to avoid this is by using out-of-sample data. This means you optimize your data and observe if a similar performance is repeated. If it does then good news, your strategy might truly be an efficient one. If not, consider making changes to some parameters that will make your strategy more robust.  

3.

Look-Ahead Bias

When your trading strategy is based on information that you’re not supposed to know (i.e. from the “future”) there’s a high chance your strategy is suffering from a look-ahead bias. 

Let’s take a look at a quick theoretical example. Say by knowing historical data, you’re aware that Amazon’s stock has been trending up during the years 2015 through 2020. Based on this, you design a strategy that backtests brilliantly. The equity curve is smooth and has a positive slope. If you were to implement this strategy in the past, you’d be rich! But the real question is, would this strategy work in the present time? The answer is no because the trader has conveniently chosen a time period that saw consistent growth and ignored the time when the market crashed in 2020.  

This basically constitutes cheating. A trader can not and should not just consider data that’s convenient to them, knowing how their strategy will fare because they already know what’s to come. Backtests with look-ahead bias do not really hold any meaning. That’s why it’s crucial to be aware of look-ahead bias creeping in while you’re designing your strategy.  

4.

Look At Financial Instrument In Isolation

This is an impractical and deadly mistake to make while Backtesting your day trading pattern or strategy because NO financial instrument (be it a stock or a commodity) can exist in an independent financial market that doesn’t get affected by any other factors. That is simply not how it works! For example, you can not pick one stock in a particular timeframe and test some trade pattern on it while being oblivious to the happenings in the rest of the financial world. It is key that a trader considers things from all aspects and goes forward with any kind of strategy with this 360-degree view on everything.  

To combat this, it’s best to use professional Backtesting software that allows you to simultaneously keep an eye on different instruments and markets, all while observing different time frames too. Only then there’s a chance of getting an effective Backtesting result.  

5.

Ignoring Commissions And Transaction Costs

For a day trader trying to backtest, it’s essential to remember transaction costs can add up and eat up a huge chunk of your profits. Similarly, it’s best not to neglect commissions. They usually have fairly standard rates but ignoring them can end up being problematic for your strategy in the long run. It might not seem like a major detail but it is.  

6.

Not Treating Markets As Living Organisms

Every trader should know that a market is not a static place where things don’t change. In reality, it is like a living organism that’s constantly changing and evolving. What worked 30, 20, or even 10 years ago might not work in the current market. The best a trader can do is test historical data for 1 to 2 years and make decisions based on that. It gives enough information to judge whether the system you have created is promising.  

7.

Focusing More On Tools Than Financial Data

There are so many fancy tools in the market right now that help traders code better. But it is essential to remember that the focus while Backtesting should always be on financial data and not these tools. Using these tools you can come up with all kinds of crazy results, but it’s important to think about whether these results would make logical sense in the real world. If the answer is yes, then only it would be best to go ahead with Backtesting using these tools.  

Tips On

Improving Backtesting

It’s best to apply your method to live markets where you’re more aware of all the factors that might make the prices move.

Understand that running simulations are just data mining while actual Backtesting involves thinking over your results logically.

Be hyper-aware of overfitting and look-ahead bias because they seem to be the most commonly made mistakes done by traders while Backtesting.

Don’t treat Backtesting as an engineering problem… rather, look at it as a way of understanding and handling financial data.

Consider various possibilities and be smart while making decisions based on your results..

No markets, timeframes, or financial instruments exist in isolation so don’t treat them like that. Test different variations of these simulations for better results.

While reviewing historical data, don’t go back too far in history. Reviewing the last 1-2 years is often enough.

Remember that your backtest results do not take into account the news from that day which might have led to a sudden spike or dive in the prices of a trade.

Try to backtest as much as possible in real-time.

It's best to apply your method to live markets where you're more aware of all the factors that might make the prices move.

1 comment

Leave a Reply

    APPLY TO THE FREEDOM CHALLENGE.

    Insert your name and email address to apply now!

    FREEDOM CHALLENGE

    FREEDOM CHALLENGE.

    If you’re new to all of this and have limited knowledge as to how the stock market works, I highly recommend you invest in The Freedom Challenge.

    APPLY NOW!

    x