I have never seen anything have such a big impact on success, as this single phenomenon.

I'm not some important, old, wise person telling their wits at the end of the road. Yet I've grown a habit of reflecting on the past, on the events and narratives, emotions and predictions, and weighing them against what the future actually held. I'm reconciling the noise, trying to figure out what makes something, or someone, right. What is right anyway, right? I've lived through several radically different political systems, in several different countries. I've seen them evolve, crash, burn, turn around, prosper. I've always asked myself "why", and I've picked a lot of reading material to help answer this.

What is right?

How do we know if something is good, correct, etc? What makes something objective rather than another opinion? This question can be driven up a very long chain, but most people would agree that it's about outcome: usefulness in giving instruments to control our lives. For most people this means money, for a lot of people this means, not exclusively, social capital, family, etc. To have sustainable control over our fates, we need knowledge. And knowledge, is that key instrument of the mind, which makes us able to predict successfully with the greatest possible rate. Whether it's about predicting how someone is going to react to a certain action of yours, or whether changing your career is going to pay off, or whether the approval of a new medical therapy will help people. It helps us navigate this world without making mistakes at every step, and accumulate the benefits of each such successful choice.

These instruments are called "models". Some of them are hardwired, while others need to be learned. Models allow us to explain and rationalize the real world in more simple ways, that our minds can actually comprehend. And the quality of a model to be right, is whether it works. If a model, applied in the right situations, is helping reach a conclusion that ends up correct more often than wrong, then it's likely a good model. And if that model helps you make choices that are a benefit to you, then it is, in fact, a good, right model.

Maybe a couple of examples will help:

  • The most straightforward model we use every day is the model of gravity. It doesn't matter if you know Newton's definition of it, or Einstein's. All of us have this model in us. We all understand that things fall to the ground. It has a very high predictive capability (amoke, airplanes and birds can't be explained with just this model, but most things can). And it helps us make correct choices, like avoiding heights, and keeping precious things safe from falls.
  • Evolution by natural selection - in any system where traits get to replicate, the traits that make replication ability better will be manifested everywhere. This is not just in biology. This model can be translated to societies and leaders too. Leaders that are conducive towards prosperity will attract copycats who want to replicate this prosperity. Hey! I know that this adaptation isn't great. This model alone is bad for understanding societies, but it helps understand parts of it, and in conjunction with other models, better abstractions and understanding can be reach, that makes the totality of these models in particular situations right.

So far it's pretty straight forward. Everybody knows that understanding is great. The problem is in how you know that you understand.

Richard Feynman's famous quote goes like "Never fool yourself, and remember that you're the easiest person to fool".

The best case of understanding

In a perfect scenario, a situation is examined by a wise person, based on pure facts, and these facts would use various models from different disciplines to establish some preliminary understanding.

Next, the wise person would apply critical thinking, by literally following some form of a checklist:

  • What if the reverse were true? What would it take for your reasoning to be completely false? Is it a lot or is it very little?
  • How plausible is it that you're seeing the wrong evidence, or not all of the evidence (ex. survivor bias)?
  • Do you want a specific conclusion or am somewhat objective in this analysis?
  • Are there any other conclusions that you threw out? Why? Could it be that there's more than 1 probable conclusion? The future is never binary.
  • How are you factoring in new evidence, how does it affect the conclusion? Are you honest about it?

Next, borrow a trick from the science world, and peer review your analysis. Talk to a friend who is also wise and has a habit of telling you that an idea is stupid, and can explain why.

Next, follow-up, learn, and discuss with your friend. Measure your thought process, your models. Explain what you did wrong, even in cases where the conclusion ended up correct. There are always mistakes. Own them, understand why you made them, and be aware of your lapses next time.

This way, you follow the track directly unidirectionally from pure evidence, with honesty towards yourself, towards the conclusion(s).

What is wrong?

In the real world, the complete opposite happens.

  • Evidence gets distorted, inconvenient parts are omitted, and the remaining ones are embelished
  • The conclusion comes first. Evidence is massaged to fit with the conclusion
  • The reverse is not only not considered, it is relegated to the realm of "unthinkable", because the conclusion has been firmly made already
  • Questioning this process is tantamount to heresy, and grounds for social isolation or worse
  • There can only be one conclusion, and it's always 100% or more
  • Friends always agree with everything, because that's what friends do - you've chosen them this way

Sounds familiar? No? Let me give a real example:

Election predictions

The infamous predictions for the US presidential elections in 2016 and 2024. Before the elections, it was predicted to be a landslide victory for Hillary Clinton. Multiple pollsters and predictors, that the markets follow and use their conclusions to make important choices, basically unanimously handed the victory to Hillary Clinton. It was unthinkable that the reverse could happen. And sure, the conclusions weren't 100% certain, but it was close enough. And they were very wrong the first time. Did anyone consider why? Of course not. In 2024 the exact same phenomenon happened.

So what happened?

All of the above. The conclusion was made, and it was bad to go against the widely accepted conclusion, so data and models were massaged to fit the conclusion.

This isn't the only example, by far, but one of the most public example, with objective failures, and very well established reality of not addressing the methodology flaws. Even saying "methodology flaws" in this context is similar to saying that Napoleon's organs are failing in 2026.

The one thing

How did so many smart people get to be this wrong, so many times, with an interval of 8 years? What makes smart people act so dumb? What's the one thing that should change for this to not be so horrificly wrong?

Peer pressure

Predictors, pollsters - they acted this way because of peer pressure. They couldn't go against established conclusions. This isn't new. Heliocentrism, Germ Theory - they taught us that going against established conclusions is bad. People punish those who are different and challenge the Status Quo.

But we're not in the 19th century any more. We've learned a lot, and science is taught in schools. So how is this still such a prevalent phenomenon?

Social media

We can't fix peer pressure, because that's a human trait. But social media is what drives it to insane heights. By modern definitions, social media today are a set of deep, established, dangerous cults.

I made a few experiments on my own, and for the last couple of years I stopped considering to established opinions on capital markets. Not the facts, no - I doubled down on facts. But I identified the fears that were stemming from the outside, that can't be well quantified or exaplined, and the fears coming from myself, from my analysis. I put my money where my mouth is, and this has been the most successful part of my portfolio. My only regret was not believing in myself more, and listening to the loud noise from cults.

I still made mistakes, of course. They are inevitable. But I was right far more often.

This is a "trust me bro", I know, but I have no reason to lie. I will never de-anonymize myself willingly. But for the last decades, almost every time I sat down, picked up a spreadsheet, and did even the simplest projections on freely available data, I was right. I was right not just about the outcome, but also about the timing of such outcomes (some more than others), while the consensus around me was that this was all silly. Even this rudimentary method was sufficient to counter the vast majority of the loudest, most influential voices.

The best thing that I ever did, was not deferring to listening to others in favor of a 1-2 hour analysis done using the simple principles highlighted above. This shouldn't have been the case, yet here we are.

Social media cults are the most dangerous phenomenon in society. This has the potential to be the undoing of ourselves, not too different from what happened during the collapse of the Bronze Age in the Mediterranean. We are firmly betraying the core principles that have stood at the base of every major leap in the civilization that created everything from modern medicine to supercomputers that can fit on your wrist.

20 years ago I saw this guy on TV, who was an expert working for the government, and he started coming up to these smart interviews asking his expert opinions. Well, back then, he said something that to me was ludicrous. I didn't know the models that I operate with now, but having an education in this domain, and a tendency to not put the cart before the horse, I called it as complete BS, and that his prediction will work out in the complete reverse. It did. He still came up on TV, he never had any retrospectives on past predictions, he was never called out on it. He's still doing it. He's still making BS predictions, that usually turn out wrong. Similar to Jim Cramer I guess, but on macroeconomics and financial / fiscal policies rather than "stock go brrr". This never changes - the worst predictors get the loudest voices. Because they operate on emotion, and emotion is powerful.

Takeaway

If you are to take away one thing from all of this, it's to learn to make conclusions on your own, and not be overwhelmingly influenced by anyone. It takes knowledge, that you can get from books like the ones in this list. It takes some work to find and analyze the data. But if it's important enough for you to stake your money, reputation or sanity on it, then make sure to do this - it's worth it.