Systems, Lies, and Emotions

You have now a system to trade or invest. Does that solve the emotive approach to markets problem? Actually, it does not. Anyone who tells you that is lying but lies are common in finance and especially trading.

I wrote a little book two years ago “15 Lies About Trading And Investing” that I offer for free to anyone who subscribes to my book “Fooled by Technical Analysis: The perils of charting, backtesting and data-mining.”  In that book one of the 15 lies is that automated systems remove emotions from trading (and investing.)

Below is an excerpt from Chapter 12 of the book which now allows free access for limited-time.

There were reports that hedge fund managers interfered with system signals in 2018 after the sudden increase in volatility. These actions, driven by emotions mainly because no one knows the future, may or may not result in performance improvement in the short-term. The worst case, which is counter-intuitive, is when performance improvement is realized. This often reinforces a bias in favor of interfering with the system and the longer-term results are disastrous. Humans are very bad in forecasting market moves due to higher order nonlinearities that cannot be accounted for with linear thinking.

As I wrote above, finance and especially trading are full of lies and “gurus” with little or no skin-in-the-game that relentlessly discover new narratives. Below is the list of narratives as I have lived through them in my long trading career.

1. Technical analysis
2. Backtesting
3. Automated systems
4. Passive investing
5. Robo-advisors

1. Technical analysis was a naive attempt to come up with trading rules (as I explain in this article, a lossy compression scheme.) It essentially acted as a wealth-transfer mechanism from naive random traders to market professionals.

2. Backtesting was presented in late 80s as the way out of subjective technical analysis by introducing evidence-based analysis. There are numerous problems with this idea and one of the most serious is data-mining bias. There are some counter-intuitive facts about backtesting; one of them is that the less you backtest, the better it is. The optimal backtesting is doing it once, for a single unique idea, and then never doing it again. There are several free articles that discuss these problems under “Perils of backtesting” in our Trader Education section.

3. Automated systems and associated narratives emerged after traders complained that even after using backtesting they were still unprofitable. “But you did not follow the system!”, “You need discipline!” were some of the responses of an industry that was basically serving market makers. “Systems remove emotions from trading”, which is Lie 12 of the book I mentioned in the beginning of the article.

4. Passive investing was served as the ultimate escape from analysis and the guarantee of longer-term profits since markets go up in the longer-term. This is a case of using facts to distort reality, which is a curious behavior that only humans are good at. Of course markets go up in the longer-term but many portfolios do not go up despite that. See this article for more details.

5. Robo-advisors is the latest combination of 2 and 3 above to offer a sense that the human is taken out of the loop but in reality humans develops the robo-advisor algorithms so this is only a good argument for those that are not aware of the tricks and narratives.

What will be the next narrative(s)?

I suspect an IPO frenzy although this already occurred in ICO and crypto with countless victims. The idea is that you can only profit from IPOs although this is not new. This is a wealth-transfer mechanism but lies take advantage of survival biases and memes like “Amazon made 45000% return since IPO” that I first exposed in social media.

Then I suspect things like social media sentiment and natural language processing will pick up steam and will be presented as solutions to the failed 1 – 5 above. We will see, maybe something totally new will emerge to salvage an industry of lies.

Just to clarify, backtesting and analysis are very useful but as I have written in my books since 1999, these are not methods for anyone but for few that posses the skills and knowledge of how to use them properly. These methods cannot be presented as solutions to problems because they are the problems to solve in the first place.

If you found this article interesting, I invite you follow this blog via any of the methods below.

Subscribe via RSS or Email, or follow us on Twitter

If you have any questions or comments, happy to connect on Twitter: @mikeharrisNY