I get a lot of traders that ask me how our trading strategies do in back testing, so I thought I would write an article that discusses back testing results on price action trading strategies.  My thoughts on back testing may be different than what you have heard in the past, but I’m fairly adamant about my stance on back testing, so I’ll try and explain why in this article.  For whatever reason, there may be some traders that do not understand the term back testing, so I’ll try and explain what it is before going any further.  Back testing is the process of testing a trading strategy against previous time and price periods.  In order to do this, you must have a set of trading variables that you can test over past price action.

Back Testing
Back Testing Price Action Strategies
(piphut.com)

As an example, let’s say you have a trading plan that says you will buy every time the 15 bar ema moves above the 50 bar ema.  Back testing will also require some additional rules such as profit targets and stop placement.  For our example, let’s assume the profit target is 5 points and our stop is 5 points.  We then run these variables against the previous price action, assume the past 30 days, to see how many winners versus losing trades we will encounter.  It’s fairly straight forward if your trading strategy can be written down with standard entry and exit rules.  The idea is that hopefully this back testing will show a nice profit over the long haul if it is a good trading strategy.

Now, let me ask you a very important question about back testing.  How many professional traders do you know that talk about back testing or even say that they have ever back tested their trading strategies?  I personally know quite a few professional traders, and none of them back test their trading ideas, at least not in order test their ideas for profitability before they trade them.  Have you ever heard or read about fund managers, portfolio managers or even commercial traders talking about back testing?  I haven’t, and if back testing is so important, then why don’t professional traders back test?  The only time I have ever heard about professional traders conducting back testing was in order to produce hypothetical trading results in order to help sell their services.  I can assure you that you never want to invest in a trading fund that is trying to lure your business in by hyping their back testing results.  You don’t want hypothetical trading results, which is all back testing is really.  Hypothetical results are worthless, so you only want to know their actual trading results and preferably what their real results were over several years of trading.

What I am telling you if I haven’t already gotten it across to you is that I think back testing is bunk!  You can not back test our price action trading strategies, or any other trading strategy and get any real results.  There are simply too many variables, but even if you could work through all of the variables, it still wouldn’t tell you anything of real value.  You will get false positives and false negatives in back testing any strategy, so you are simply wasting your time and efforts by trying to back test a trading strategy.  I know that there are many of you that are probably ready to argue against my line of thinking, but that’s OK, because I know you have been programmed to think this way by the misinformed teaching types.  The very ones that are teaching you the importance of back testing have likely never had a profitable year in trading.  They are the same ones that are trying to sell you their holy grail of trading ideas for many thousands of dollars.  Forget all of that and throw it all out, because the only way to truly test your trading idea is to get on the simulator and trade it in real time and see how it does over time.

If you think I’m the only trader out there that does not believe in back testing, then think again.  Below is an excerpt from an article that I read recently at www.danielstrading.com and it had the following to say about back testing.

I often speak with traders that want to back test their strategy and trading ideas before going live. They spend days and weeks pouring over data attempting to simulate every market condition to see how their “program” works. The goal is to find the ideal situation where a trader’s ideas compare with past market conditions to produce profitable (hypothetical) results. A considerable amount of time, effort, energy, and capital are tied into back testing-but does it work?

Before delving into back testing we must first identify how traders approach the markets. There isn’t a trader alive who isn’t searching for a new idea, indicator, or edge. The goal of any good trader is to maximize profits by assuming the least amount of risk and minimizing losses. To accomplish this, traders are constantly refining their ideas, studying charts, researching data, and searching for market inefficiencies. A logical step in this process would seem to be to back test strategies. Before risking capital, a trader wants to make sure that the strategy has worked in the past.

However, the fact is that professional traders don’t really back test. I personally know hundreds of traders from prop traders, to hedge funds, to local market makers and not a single one has spent time back testing.  Professional traders don’t back test their strategies because it doesn’t really tell them how their ideas perform or operate under live conditions and present market activity. What seems like a logical first step to trading is really just testing how the market traded in the past.  You can read the rest of the original article here.

If you throw out all of the BS and get right down to it, the only way to make money day trading is to have a good understanding of price action with the ability to read a price chart.  Even if they say differently, if a trader is making money on a consistent basis, they are able to read a price chart efficiently, and it is a skill that can be both taught and learned.  Chart reading skills are nothing more than what we call price action trading, and it’s the ability to look at a price chart and understand what prices are doing, and also why they are doing it.  This chart reading ability allows you to predict, with high probability, what prices are likely to do next in the near term.  There is no way to transfer the skill of chart reading to a formula that can be back tested, so attempting to do is is both futile and a waste of your time.

If you would like to learn how to read a price chart and how to make money as a day trader, we can show you how, and that is by learning our price action trading strategies.  Don’t waste your time back testing results on price action trading strategies, or any other trading strategy.  Your time is better spent in trading the simulator with a proven trading strategy that is proven in real time trading only.  Any time spent on back testing is simply a waste of valuable time and efforts that you could be applying to learning real trading techniques and getting real trading experience.  There is no substitute for real time experience, so get on the simulator and start working on it today.  If you would like to learn more about our price action trading techniques, you can find more information at http://priceactiontradingsystem.com/pats-price-action-trading-manual.

Leave a Reply to antonio machado Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Comments

Lluis November 6, 2015 at

Hi Mack, First of all, I want to apologize for my bad english. By the way, I think your birthday is the same day as mine!

I’m thinking how to learn about price action and I would like to know your opinion: I have the platform ForexTester and I can view charts since 2001, to train with the price action this platform could be validated? Exist an option where I can see bar by bar and observe the best entries, then after I try to apply it to simulator real-time. I find interesting what you mean in the post, but in this case I think would be beneficial to learn. What do you think?

Thanks a lot for your time Mack!

Lluis from Spain

Reply

    Mack November 6, 2015 at

    If you are asking me about trading on a forex simulator using historical data, then yes, it will be good practice. I do not like trading forex, but for simulating trading, a price chart is a price chart, and price action works on any of them.

    Reply

antonio machado November 27, 2017 at

I’m developing a fully automated robot trading software, that trades all the best opportunities from the stocks Dax market.

At this stage I am building the knowledge base based on facts and rules.

I am a programmer who knows nothing about trading, but with the help of experts, I am building my knowledge base.

Facts & Rules :

I trade chart patterns Bullish kicker, Bullish flag, Cup with handle, Inverse head and shoulders… (*1) like Tim Grittani, Tom Bulkowski, William O’Neal, Dan Zanger, Mark Minervini…

Experts say to use moving averages to improve performance.

My period of analysis is 17 years, I need to cover several cycles.

I work with 25 tickers from Dax, a liquid market with low commissions.

Take profit average: 3.62
Stop loss average: 3.06 (Win/Loss 1.18)

Maximum take profit: 11.5
Maximum stop loss: 6.5 (Win/Loss 1.77)

Experts say the optimal ratio of wins to losses shoulde be 2:1 !

My time stop is 7 days !

Experts say not risk per transaction more than 2% of total capital !

I found 425 Bullish Flag patterns in 17 years, for 25 tickers (*2) !

Experts say that 30 trades is a minimum to suggest statistical significance !

First result: if I trade 35 times a year, I get an average year profit of 37.82% (*3), deducting commissions from the broker!

Questions for the expert:

(*1) Is this a profitable strategy ? Do you work with another strategy more profitable ?
This is profitable but you work with another pattern ?

(*2) Is this value statistically significant ?

(*3) Taking into account the results of professional trader, is this first result a low, medium ou high value ?

Thank you,
Antonio
AI programming student

Reply

thevmackwin August 20, 2019 at

When you say “Fact & Rules”, are you using the Prolog language to implement that knowledge base?

Reply

Leave a Reply to antonio machado Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.