Summary of Antifragile by Nassim Taleb

  • Post category:Summaries
  • Post last modified:September 18, 2023

Chapter 15: History Written by the Losers

The author tells the story of an economist worried to teach traders a complicated theorem supposed to help them with options trading.

However, the economist had it backward.

Option traders began trading before the theorem was established.

Even though they may not understand how the economy worked, they could trade options the same way a child can ride a bike without knowing about the physics that make it move.

The economist believed the theorem came before practice. In reality, practice came before theorem.

We don’t put theories into practice. We create theories out of practice.

Romans, Greeks, French built monuments and cities based on rules and heuristics, not mathematics. The math we use today to build them came after these buildings had been built.

What does it have to do with antifragility?

Practice is antifragile because it has options. When you practice to reach a result, you tinker, try, and change until you reach the result.

A theory is fragile. It does not leave space for tinkering and becomes invalidated as soon as its strict application does not yield the promised results.

A good example of such is cooking. There is no theory about cooking – it’s entirely apprenticeship-based.

Technology and its development work the same way.

Nobody invented the computer and the internet with the idea to connect them together. Both of them were results of needs for tech improvement, and happened to connect together in a random way.

The same thing happened in medicine, where the search for x led to the discovery of y. These tend to happen in the realm of private research, not academia. Why? Because academics have an agenda.

If they discover y, they will disregard it as this is not x.

The Inverse Turkey Problem

For the antifragile, the unknown is likely to bring good news, but bad news for the fragile.

Eg: biotech companies make no profit. But when they do – they earn big. They are antifragile. The opposite is insurance companies. They make a lot of money most of the time, but one event can bankrupt them completely.

Let’s summarize:

  1. Look for optionality. Rank things based on that.
  2. Look for things that bring open-ended payoff (no limits).
  3. Don’t invest in ideas or plans, but in people that can tinker and try until they find what works.

Chapter 16: A Lesson In Disorder

There are two types of environment. The ludic, with its set of pre-determined rules, and the ecological, where rules are unknown, which is real life.

Seen this way, parents may be the biggest hindrance to children’s development, taking them away from real life into ludic environments, like soccer games, classrooms, etc.

We need randomness, mess, adventure, and uncertainty.

-> only the autodidacts are free.

Chapter 17: Fat Tony Debates Socrates

The idea of this chapter is that you do not need to define a word to understand what it is.

The author tells the story of a trader trading green lumber, not knowing what green lumbers were. That didn’t stop him from being a good trader.

There are two types of knowledge. The one you learn in textbooks, and the one, opaque, you learn through practice.

Socrates asked people to define the words they used, and, should they not be able to do so, told them they shouldn’t use them. Doing so, he confused people – and was put to death because of that. He infused fragility into society.

Not everything you know or use can or (has to) de facto be explained.

Socrates was all about knowledge. Fat Tony is all about results. Fat Tony focuses on the payoff of his actions, not on their nature.

The payoff, the consequence, is always the most important thing.

To quote the author.

Philosophers talk about truth and falsehood. People in life talk about payoff, exposure, and consequences (risks and rewards), hence fragility and antifragility. And sometimes philosophers and thinkers and those who study conflate Truth with risks and rewards.

True and false do not matter. Asymmetric payoffs – consequences – is what does.

If you look at your decisions, you will see you make them based on fragility, not on probability. There is almost always an asymmetric result.

Eg: people that board the plane are unlikely to be terrorists, but we are fragile to terrorist events, hence, we check everyone.

-> the Black Swan (the event) and how it impacts you are not the same things. You will not care about a Black Swan if it does not impact you -> you’re obsessed with the impact.

And yet in society, after a Black Swan happens, society chooses to reinforce its risk-assessing model, instead of creating an antifragile structure capable to handle the Black Swan.

To conclude Book IV, practice is what matters, and theory learned in academic institutions doesn’t.

Book V: The Nonlinear and the Nonlinear

Chapter 18: On the Difference Between a Large Stone and a Thousand Pebbles

The fragility of something is vulnerability to the volatility of the things that affect it.

Furthermore, the impact is non-linear.

Eg: being hit by two stones of 50g < being hit by one stone of 100g.

image 52
The relationship is non-linear.

Why is fragility non-linear?

Because if it wasn’t the case, jumping one centimeter would put as much damage on you as 1/10000 of the damage you’d receive jumping 100 meters.

So, we are wired to resist small events, but not extreme ones. There are many, many more small events than extreme ones too.

-> we are immune to cumulative effects of small events.

Jumping 100 times one centimeter ≠ jumping 1 meter.

For the fragile, the cumulative effect of small shocks is smaller than the single effect of an equivalent single large shock.

This is why it’s better to lift heavy weights at the gym a few times, than light weights a lot of times.

This also means that the bigger the Black Swan, the disproportionately large the event will be.

Non-linearities relationships are everywhere.

A bit of randomness does not have a big impact, but as soon as you have too much of it, the impact is largely non-linear, and big.

Similarly, small is much more antifragile than big due to squeezes.

A squeeze is when you have no choice but to do something right away.

Eg: if you own an elephant and water becomes 50% more expensive (squeeze), your water budget will be much bigger than if you owned a dog.

Likewise, if you have to purchase, say, 1 million pieces of bread, for 1 euro, and manage to have a 10% discount, you save €100k!

10% on one piece of bread is a mere €0.1.

Finally, ants, being small, can carry their weight hundreds of times.

Big animals absolutely cannot.

Projects and Predictions

Uncertainty in the context of a flight will extend the duration of the flight – not shorten it. When a “surprise” happens in the airline space, it’s rarely a good one.

The same can be said of projects.

More uncertainty will incur more costs, more delays.

Eg: WWI was supposed to last a few months. By the time it ended, UK and France were heavily indebted, and it had lasted…four years.

-> complexity + asymmetry = explosive errors!


Fragility in any domain, is non-linear.

Chapter 19: The Philosopher’s Stone and Its Inverse

In 2003, the author saw that Fannie Mae (a lending company) made very little money when an economic variable moved downward, and would lose a massive amount when it’d move upward.

-> it was the definition of fragility, and the author subsequently bet on its collapse, which happened in 2008.

-> measures of fragility are much better than measures of risks (which don’t work!)

Measuring fragility is easy. All you need is to check by how much the downside increases with variation.

Eg: if a city’s traffic slows down with 1000 cars by 10 min and by 30 min with 1000 more cars, you have something fragile.

So, we can classify things in three distinctions (coming back to the Triad: fragile, robust, antifragile):

  1. Things that like disturbances in the long run (evolution, scientific discovery, etc)
  2. Things that are neutral to disturbances (a book)
  3. Things that dislike disturbances (traffic)

Now, it’s time we define what’s the philosopher’s stone.

The philosopher’s stone works as follows: if you are in a situation where the asymmetries are positive and you have the option to stay in that situation or get out, you will do better in the long term in uncertain situations.

The more role optionality plays, the more you will outperform.

Book VI: Via Negativa

Via Negativa means not designing something by what it is, but what it isn’t. Likewise, the secret to antifragility sometimes is more about stopping doing things or removing them, than doing them and adding them.

Eg: when Michelangelo was asked by the Pope looking at David the secret of his genius, he said he just “removed everything that wasn’t David”.

In practice, the ones at the top act more by avoiding, than doing.

  • Chess masters win by not losing
  • Rich become rich when they avoid losing money
  • Religions are about what you cannot do
  • Life is about what to avoid
  • Good political systems focus on removing the bad guy, not on getting the good ones in.
  • Focus is not about saying yes to one thing, but no to 1000 things.

Antifragility is reached by not being a sucker.

-> we know more about what is wrong than what is right.

Negative knowledge is more robust than positive knowledge.

-> knowledge grows by subtraction more than by addition.

What is wrong is robust, and what is unknown is fragile.

Another idea linked to via negativa is less-is-more.

Less-is-more is applicable in Black Swan’s domains – Extremistan – where they can have a disproportionate effect. If you focus on profiting from a Black Swan, all you need is to be right once.

In a way, less-is-more is tied to Pareto’s law, which is a power law. 20% of people own 80% of the land, 1% of authors sell 99% of the books, etc. Likewise, a 1% change in a system can lower fragility by 99%.


  • A few homeless people cost the state a lot of money
  • A few employees annoy everybody
  • The sickest 10% of the population are responsible for 64% of healthcare cost

If you have more than one reason to do something, don’t do it. It means you are likely trying to convince yourself to do that thing, which means, you actually don’t need to.

Likewise, people should be known for one idea, and one only.

-> avoid people with big resumes that wrote about everything.

Chapter 20: Time and Fragility

What is old is superior to what’s new, because what is old has resisted time, and time crushes what is fragile.

If something survives, it means it is not fragile.

Therefore, the way to imagine the future is to look at what is fragile now, and take it away.

The future is what’s left.

What futuristic authors like Jules Verne failed to predict was that our world is much closer to theirs than they imagined.

Think about it. Shoes, restaurants, and silverware exist for millennia.

Let’s now talk about the Lindy Effect.

To Age In Reverse: The Lindy Effect

The Lindy effect separates the perishable from the non-perishable.

Humans are perishable, DNA is not. A car is perishable, the automobile industry isn’t.

The Lindy effect works as follows:

For the perishable, every additional day in its life translates into a shorter additional life expectancy. For the nonperishable, every additional day may imply a longer life expectancy.

The longer something has been existing, the longer it is expected to exist.

Lindy things age in reverse because if something has been existing for a long time, it is robust, and the longer it lives, the more robust it gets!

In a way, the future is the past. By looking at what has lived the longest, we will know what will live the longest.

Chapter 21: Medicine, Convexity, and Opacity

Only take medicine when the payoff is largely more positive than the potential harm (eg: surgery).

When the payoff appears small, you expose yourself to consequences that may not yet appear.

Eg: smoking. It took decades to find out it was actually bad.

When something is not natural, it needs to prove first it’s not harmful, and that takes time.

First Principle Of Iatrogenics (Empiricism)

Evidence of harm is not what we are looking for when testing a new drug since this will likely come in the future. Rather, evidence of no harm is what we want.

Second Principle Of Iatrogenics (Nonlinearity In Response)

We shouldn’t take risks with healthy people, but we should with those deemed in danger as it’s worth it – if they’re going to die, might as well try everything.

If your tension is a bit too high, the chance to benefit from a medicine will be 5.6%.

But if blood pressure is high, then you benefit from 26% up to 72%.

Similarly, you’re unlikely to find drugs that will make you feel healthier than already-healthy as if it existed, nature would have already found it.

The problem with medicine is that its successes have always been shown, and its failures, hidden.

Furthermore, it always takes years, if not decades, for the side effects to show up – then to be diagnosed.

When it comes to something like blood pressure, normal medicine does not take into account the variabilities. It’s not because it’s a bit high once, that it is all the time!

As a result, we medicate people that don’t need medication, and kill them in the process.

Chapter 22: To Live Long, But Not Too Long

It is often advertised that we live longer now than before, but it’s not so sure.

Most people died at birth, giving birth, or in battles in the past. Those that lived, lived long.

Then, the solutions medicine brought need to be contrasted with modern plagues, such as smoking. In fact, it has been shown that not smoking would be the biggest improvement overall for the healthcare system (less is more).

Overall, strikes in hospitals did not let more people die, as only the very important surgeries were performed. It has been shown that people would live longer if the medical budget was cut, and where medicines would be administered only in the most extreme cases -> most medicines are administered uselessly.

The Lindy effect also works with food. If you want to make sure food is healthy, only eat what we have been eating for 1000 years.

Furthermore, the key to good nutrition may also be in eating at random times and not regularly as we do, and eating random stuff and not per diet like we do.

Herbivores eat regularly. Carnivores do not. Predators go for long periods without eating, then feast. And it’s never “balanced”.

As a result, you may be eating a lot of meat at some point, then a lot of vegetables at another, then fast.

Fasting has been shown to be excellent for health.

Finally, breakfast is not a natural thing – it may even be harmful. You’re not supposed to find food upon waking up, but should hunt to get it first.

-> go for a run in the morning.

Book VII: The Ethics of Fragility and Antifragility

Under opacity, some can hide risks and get away with them.

The only way to mitigate risks is through skin in the game.

Chapter 23: Skin in the Game: Antifragility and Optionality at the Expense of Others

In older societies, there still were heroes, people that decided to assume the downside for everybody else.

Today, it’s the opposite. Politicians get elected not to assume the downside, but to benefit from the upside.

This leads to another triad.

Those with no skin in the game that benefits from the upside; those with skin in the game that neither benefit nor suffer from anyone; and those that assume for other people – heroes and adventurers.

In the past, courage was dying in battle. Today, it is to stand for an idea, or for certain values.

Those that don’t want to take risks for their opinions or take on opinions that carry no risk, are not as honorable.


A lot of professions are becoming antifragile at the cost of our fragility: government employees, academic researchers, journalists.

Hammurabi’s code, now 3 800 years old, made sure to get everyone’s skin into the game.

Eg: if a house collapses and kills the owner, the builder will be put to death. If it kills the son of the owner, the son of the builder will be put to death.

-> The Romans asked the engineers to stand under the bridge they built. Ethically, someone voting for war should be at the front in the battle.

Same goes for predictions. Predictions originating from someone with no skin in the game aren’t worth much.

The problem is that in our modern world, intellectuals with a lot of power make recommendations without skin in the game. They have privileges, and no obligations.

Karl Popper thought ideas competed and that the least wrong of them would survive. But it’s wrong. It’s the people/societies with the right set of ideas that survive. The others die.

Chapter 24: Fitting Ethics to a Profession

Someone who lives with 2k per month without being dependent on anyone is freer than someone who spends 1 million a year being dependent on his job.

A lot of people, as they earn more, spend more. As a result, they are never free.

The Tyranny of the Collective

Academics with unusual ideas won’t get jobs because they don’t have the ideas that everybody has.

So everyone is teaching and learning stuff that everyone knows is wrong. No one has enough freedom to break the cycle.

Chapter 25: Conclusion

Everything gains or loses from volatility. Fragility is what loses from volatility and uncertainty.

For more summaries, head to

Did you enjoy the summary? Get the book here!

Want more?

Subscribe to my monthly newsletter and I'll send you a list of the articles I wrote during the previous month + insights from the books I am reading + a short bullet list of savvy facts that will expand your mind. I keep the whole thing under three minutes. 

How does that sound? 

Leave a Reply