How Reliable is Back Testing in Predicting Gains?

Ever wonder if past investment strategies can predict future results? Learn how back testing over multiple decades provides insight. Could this method offer you more confidence in your investment choices?

This article is a website version of our weekly FREE Best Ideas Newsletter sent on 23.04.2024. Sign up here to get it in your inbox every Tuesday.
Sign up here to get the emails delivered to your inbox.


I had a really insightful e-mail exchange with a thoughtful and smart investor based on the article Is Back Testing Lowering Your Returns? I shared this with you recently.


This is a summary of what he wrote:

When you test an investment strategy just once, it's hard to know if the results are reliable.

A better way is to test the same strategy several times over the same length of time, like 5 or 10 years. For example, you could look at how the strategy did from 2000 to 2010, then from 2001 to 2011, and so on. This way, you can see a range of possible outcomes—what's typical, what's the best case, and what's the worst case.

Understanding these results helps you know what to expect if you start using the strategy today.


He raises a lot of great points that make a lot of sense.

The longer you test the strategy over different 5-to-10-year test periods the more you have a feel of what the strategy will do.

It is something I am sure you have also thought of.


I Also Tested Everything

After doing thousands of backtests, I realized that all the testing in the world is not going to help me find the best or even the most likely returns of a strategy.

Let me give you an example.

If you did a lot of backtesting and found that:

  • from 1990 to 2000 buying undervalued companies performed greatly,
  • from 2000 to 2010 momentum (Companies going up in price the most) performed great and
  • from 2010 to 2023 the largest big tech stocks completely dominated returns.

Now you would have 33 years of data, and different periods you tested but it still won't tell you anything as to what strategy is going to be the best-performing strategy over the next 10 years.

And, if you tested only ONE strategy, returns would have been all over the place. In some periods it performed well and others not so.


What Strategy Is Best?


So, what strategy will give you the best returns, you may be thinking?

As you already expected you just do not know.

No one does, not even that fund manager note you read last night!

If you had all those 10-year period returns, would they give you a clear idea of what to invest in the next 10 years?

The simple answer is no.

All that testing would not have given you a strategy or actions you can take right now that will give you the best future returns.



What I Do and How It Can Help You


The more I tested and invested (over 37 years already) the more I realized there is simply no such thing as the “best” strategy.

The more back testing I have done the more I find that testing is only good to make sure if a strategy performed “relatively” well in the past. This means it generally worked, sometimes outperforming the market and sometimes not. But it generally worked.



What I Do in My Portfolio


In my portfolio, I use a few different strategies.

I have momentum, deep value, value and momentum, trend following, and a small interest-bearing part.

And I follow two simple but effective rules to keep me in the game.

The two rules protect my capital and protect me emotionally so that I can stick to investing over the long term.

It is the exact same rules we follow with the Quant Value and Shareholder Yield investment newsletters.

  • Rule #1 - I get rid of losing positions fast. We follow a strict trailing stop-loss strategy of 20%
  • Rule #2 - I stop buying when markets are falling. We stop buying if a market is below its 200-day simple moving average.



Summary and Conclusion


You have most likely heard me say this a lot.

Investing is not hard but it’s more difficult than it looks because of things you have no idea about when you get started, for example emotions, doubts, and stress.

If you have done it long enough, I am sure you have realized investment success boils down to:

  • Using a time-tested strategy that fits your nature
  • Following systems that keep losses low
  • Implementing these in a way that lets you sleep comfortably at night and keeps you investing.




Quant Value newsletter update


As you can imagine performance last week was mostly negative, but there were positive highlights, mainly due to good results, for example:

  • Severfield PLC +24% on good results
  • Renold PLC +13%
  • AlzChem Group AG +7% (Recommended this month)


Subscribers are still sitting on the following solid gains:


If these ideas sound interesting, you can get more information here: Your Treasure Map to Europe, Asia, and North America's Hidden Gems!

Want to have a look before you subscribe? Simply reply to this email with “Quant Value trial issue” in the email and we will send you a recent issue to review.



Shareholder Yield Letter update


Since May when we started the 44 ideas have already paid an average dividend of 2% and I am sitting on an average return of 10.9% after the pull back in the markets last week.

Dividends keep in rolling in and will increase substantially soon when the yearly dividend paying companies start paying. This makes it a great portfolio if you are looking for income ideas.

As things stand today the portfolio has an average historical dividend yield of 4.7% and  bought back 4.2% of their stock last year. This gives you an average Shareholder Yield of 8.9%!

If this sounds like the kind of companies, you would like to invest in you can find more information here: Invest big, win bigger with our market beating large-cap strategy!

Want to have a look before you subscribe? Simply reply to this email with “Shareholder Yield trial issue” in the email and we will send you a recent issue to review.



Your, helping you find your best strategy analyst