Ignore the noise in investing and finance
First of all what is noise? This is a concept borrowed from signal processing. In signal processing — the study of signals, as applied to voice recognition — technology behind your Alexa/Google home for example -, satellite communications — your cell phone signal…-, medicine — your ECG exams…-, you always try to increase your signal to noise ratio. You try to identify and interpret signal, which has real value, and to ignore noise, which is worthless. You remember last time you went to your cardiologist and he/she looked over your ECG? That’s typical — your cardiologist is trained to identify signals in that exam, which indicate with accuracy certain cardiovascular conditions, and to ignore the noise which is meaningless, and which is most of your ECG exam data. There are areas, like genetics, which are famous for their low signal to noise ratio — most of the data we collect in those fields is meaningless and it is incredibly difficult to identify the meaningful signals, the part of the data that has meaning and value.
We should all strive to find meaningful signals in our lives — data/facts/information that has meaningful value. And to avoid and ignore noise — data/facts/information that are useless in practice.
And investing and finance are full of noise. Whatever blog post, company case study, financial article you read yesterday, most of it was probably noise. Useless information with no power to predict anything, nor to explain anything. The question then becomes why is there so much noise in finance? Why are there so many useless facts, illogical explanations, and spurious correlations thrown at us in this field?
I see three specific reasons and one general reason for this:
1. Abundance of data
Financial data is absolutely abundant. In finance we record everything — literally everything that happens on stock markets, second by second, millisecond by millisecond, every single day. You can go on internet and find, for free, financial data for the entire past century. It is plentiful. Financial data is actually an incredible industry onto itself, worth billions if not dozens of billions of dollars.
There are few industries in which we record, and publish widely, what happens every second all around the planet. If you want to know what happened in your local hospital a few minutes ago it will be impossible for you to know. But if you want to know what happened to the price of a Vietnamese company at market open, you can go on Yahoo finance or Google finance or Marketwatch or dozens of other websites, and the data is there, free, at your fingertip.
As any data scientist will tell you, the problem with capturing large amounts of data is that usually they are full of noise — this means most of it is worthless. This is a common problem in the world of “big data.” And the same applies in finance of course. Financial data is notorious for having a very low signal to noise ratio, which is why predicting in our field is so difficult, even for the smartest scientists in the world. I worked with some of the people who designed the satellite communication systems that fly you around the world, and they unanimously concluded that the signal to noise ratio in finance is so low for prediction to be an almost impossible task even for them.
Yes, I do think general practitioners like us should avoid trying to predict short term movements in financial markets, and activities like day trading, unless someone knows what they are doing — go to a casino, you have better odds there for sure.
2. Analytical bend of practitioners
Finance is an industry overflowing with smart university graduates from good universities, with mostly analytical backgrounds. Graduates in “finance” “accounting” “business” and more and more hard sciences, make up a large part of the financial labor force. Look at every job description in finance, and you’ll often see the words “analytical skills required.” Just like financial data is a massive industry onto itself, the analysis of financial data is yet another very large global industry. As such, there is no shortage of “financial analysts” out there, working for commercial banks, investment banks, financial media — able and willing to publish their analysis of financial data. If you take other industries such as medicine, law, engineering — I would struggle to find entire sub industries within them, whose job is to produce analytical reports on the industry data.
You don’t believe me? We’ve all read the Financial Times, the Wall Street Journal, a broker report from a bank like UBS, Seeking Alpha. When is the last time you read The Legal Times? What about the Medical Street Journal?
The problem is that this army of well-meaning financial analysts are subject to one of the main “laws” of data analysis: garbage in, garbage out. And as we saw above, the raw data they work with is mostly noise and signal is incredibly difficult to separate from this sea of financial noise.
p.s. Nassim Taleb, the public intellectual, professor of probabilities, and fund manager, is known to berate financial articles that try to explain what happened yesterday on public markets. His very accurate point is that hundreds, maybe thousands, maybe more, factors, impacted public financial markets yesterday, making it impossible to ascribe yesterday’s markets behavior to any one reason. And yet on any given day you’ll find plenty of articles and talk show hosts doing just this.
3. Highly developed financial media
We have, globally, a well-developed financial media. Unlike many other fields and industries, finance has its own national, and global, media. The Wall Street Journal, the Financial Times, CNBC programs, websites such as Seeking Alpha. You will struggle to find other professional industries with such an established media ecosystem. These media of course relay regularly, repeat, and spread, the analysis produced above.
If the data is noisy, the analysis will be sketchy, and the media will simply spread meaningless noise.
4. Human desire to know and understand the world
Money and finance touch all of us. And we are all driven by the human desire to understand our world and what is happening to us. As such, we are suckers for stories that explain what happened to us. We need a narrative around what happens to us. We want to know why our Apple stock fell 5% yesterday. We need a narrative for why our tech stocks are doing well and will keep doing well — at any point in time, whichever of these topics you want to explore, you will find dozens if not hundreds of analyses on internet, for free, about why Apple “is still a good buy,” or why “this time is different.” I won’t count the number of articles I’ve seen on financial blogs, explaining why they were wrong in their previous article and actually buying Apple wasn’t such a good idea. Honest, but ultimately meaningless and money losing, mea culpas.
And here you have the full spectrum: large amounts of financial data being produced around the world every day, armies of smart and well paid analysts whose jobs is to try to explain this data for us, a global media to repeat and spread these findings, and an audience who are naturally suckers for stories and explanations.
