Premium Market Analysis, Trader Education, Software, and Trading Strategies

Risk Management

Forecasting is Far From Enough

Forecasting is far from enough for market success. Knowing who you are up against is far more important than being able to work with data and forecast market price moves.

The idea of forecasting the future is intriguing. But for market success, forecasting is far from enough.

You work hard to forecast the future price level if you are a technical analyst, the future return if you are a quant analyst, or the trend if you are a macro analyst, using the most advanced tools at your disposal.

You then post a message on social media with your forecast, and you get some likes.

Someone posts a random forecast and gets 100 times the number of likes you got. In the thread, people clap, elevating the random forecast to a prediction.

If you are brave enough to take the challenge, you get back irony and sarcasm: “Look at the number of likes first and then reply.”

The forecast, or its quality and significance, is not enough. Who you are up against is way more important.

If your forecast says crude oil will fall but the government suddenly starts releasing reserves, then mathematically, you are right, but in reality, you are wrong.

There is a significant gap between mathematics and reality.

There is also a huge difference between 10 likes and 1000 likes on social media posts.

Who are you up against? This is a more difficult question to answer, and social media sentiments have not been able to provide a good answer, in my opinion.

If you are up against collective madness, then the most accurate forecast based on past data will prove to be garbage.

People cannot affect weather forecasts by their actions.

But market participants can affect forecasts in any direction due to reflexivity.

Market forecasting is orders of magnitude more difficult than weather forecasting.

Someone comes up with an impeccable forecast based on mathematics, and an influential economist appears on national television the next day and takes the opposite view.

Whom are the mom-and-pop investors going to listen to? The person with 10 followers on social media, or the celebrated economist on television?

The media cares about the value of narratives in creating advertising revenue. They do not care about the “truth”. At any rate, in a reflexive sense, the “truth” is created by the media.

Therefore, you have to be very careful in the markets. There is no “truth”. There is only “reflexive truth”.

If the technical analyst with thousands of followers says, as did yesterday, that the end of the US dollar is coming because of a “MACD divergence”, whatever that means, they get 900 likes, and this is the actual number.

The person who challenges this with 25 years of skin-in-the-game gets mocked. “You don’t understand technical analysis”.

The markets have changed in the social media era. There is a significant regime change. For now, long-term trends are not affected (yet), but short-term trends are being affected constantly.

Knowing who you are up against and risk management is way more important than forecasting.

By the way, forecasting as a subject has not progressed much in the last 20 years. Some optimization algorithms found their way into some math algorithms and gave rise to machine learning, but beyond that, there has been no major theoretical advance.

This is because forecasting is both an art and a science.

Various actors claim forecasting competitions raise the understanding of the subject and its sophistication.

The only understanding I have is that if you have a thousand participants in a forecasting competition, someone with a random model will win due to luck. It is no different than those forex trading competitions. There can be no proof that a win means a breakthrough in the field of forecasting.

Forecasts are tools, but in most cases, they are random, and even if they are significant, they can be invalidated by market participants.

Every time a trader enters a position based on an expert forecast, they should ask: “Who am I up against? Am I up against collective madness?”

Forecasts cannot provide an answer to this question.

Disclaimer:  No part of the analysis in this blog constitutes a trade recommendation. The past performance of any trading system or methodology is not necessarily indicative of future results. Read the full disclaimer here.

Charting and backtesting program: Amibroker. Data provider: Norgate Data