Monday, September 23, 2013


Nassim Taleb’s latest book Antifragile is hard for me to describe succinctly. It contains a number of interesting ideas that made me think, which is a very good thing. On the other hand Taleb sets a new standard for incomprehensible meanderings. Taleb sees himself as part philosopher, but I can usually understand the writings of philosophers.

Two of Taleb’s previous books had messages important in shaping a sound investing strategy. Fooled By Randomness teaches that we tend to mistake skill for luck and see patterns when there is just randomness. Most sensible investors should conclude that they do not have the skill to trade against investing professionals even if they feel like geniuses in a rising market.

Taleb’s The Black Swan teaches that extreme events are much more likely to occur than standard theory based on the Gaussian bell curve predicts. However, I preferred the treatment of this subject in Benoit Mandelbrot’s earlier book The (Mis)Behavior of Markets. Investors should conclude that they need to consider how their finances would be affected by extreme events such as equities dropping by 50% at the same time as losing one’s job. Considering such a possibility seriously dampens enthusiasm for leverage.

I can’t say that Antifragile adds much to these two investor lessons. Perhaps it reinforces the need for investors to protect themselves against extreme events.

Taleb labels anything that tends to be harmed by random events as “fragile”. For the opposite, he does not use “robust” because he is not looking to describe things that are not harmed by randomness; he wants a word for things that are actually helped by randomness. For this he coins the term “antigfragile”.

Readers can be forgiven for thinking that all things fragile are bad and antifragile good. This is not the case. For example, lottery tickets are antifragile because the downside is very limited and extreme randomness gives a huge upside. However, lottery tickets are still bad because the expected return is so low.

Giving examples of the incomprehensible sections of this book would require unreasonably long quotes, so I’ll just give a few amusing short quotes:

Risk “is both predictive and sissy.” Antifragility is “nonsissy.”

“We have managed to transfer religious belief into gullibility for whatever can masquerade as science.”

“Science is an anti-sucker problem.”

Taleb is no fan of newspapers “with their constant supply of causes for things.” I read newspapers but understand what he means whenever I read an inane headline like “markets drop amid Syria concerns.” We seem to yearn for explanations, even when there is no reason to believe they are correct.

“Risk is in the future, not in the past.” This is a great criticism of data mining. If you build an investing strategy that can handle all the shocks seen in some stretch of the past, you may be leaving yourself wide open to new types of shocks or larger shocks.

Taleb is no fan of people who make economic predictions but aren’t held to account. “Overconfidence leads to reliance on forecasts, which causes borrowing, then to the fragility of leverage.”

The people that Taleb most dislikes are labeled “fragilistas”. This seems to mean someone who advocates economic policies that make our economy more fragile. If a given shock causes some amount of damage, a more fragile economy would suffer more damage from the same shock.

In a diatribe against the metric system Taleb says “A meter does not match anything; a foot does.” He goes on to go through several British units claiming they are all natural in some way, but that metric units are not. This is all nonsense. The only virtue of British units is that they are familiar to some.

Apparently, applying ice to injuries is a “naive version of an interventionism,” and “there is no empirical evidence in favor of reduction of swelling.” I’ve played a lot of sports in my life and I estimate that I’ve used ice to control swelling perhaps 1000 times. This may seem like a lot, but it’s only 3 times per month for about 30 years. Of the perhaps 100 times I was too lazy to apply ice I regretted it almost every time. There may be good examples where medical intervention has questionable benefit, but using ice to control swelling is a terrible example. I think this is a case where Taleb is able to come up with a very intelligent-sounding reason to avoid doing something he doesn’t want to do.

“Corporate managers have incentives without disincentives.” This is very true. Stock options give corporate managers huge paydays if their risky leveraged strategies pay off. If the strategies flame out, investors eat the losses (or taxpayers for too-big-to-fail businesses).

One of the most coherent parts of the book is Appendix II on economic models and how they blow up. Unfortunately, I didn’t understand any of the remedies discussed in Table 12.

Overall, I’m glad I read this book because it made me think about a number of subjects in a new way. However, I have a hard time recommending it to others.


  1. The book is about derivatives and the mindset required to manage derivative trades and positions. Not a single mention in your "review". It sounds like you wasted your time.

    If you prefer, Taleb's paper is enlightening:

    His Authors @Google talk is probably the best public version of his talks:

    1. My review was about "derivatives and the mindset required to manage derivative trades and positions." Perhaps you missed it because I used the same approach as Taleb used in the book to barely mention anything about this main theme.

      Seriously, though, I don't think I wasted my time because reading this book hardened my views against using leverage in my personal finances, which I think is a good thing. I probably could have achieved the same result less painfully by simply rereading The Black Swan.

  2. Good review, Michael. Another that isn't very complimentary to Taleb - I'd love to see a debate between these two - it would be an intellectual cage fight of the highest order.

    1. @CanadianInvestor: That's an interesting review. However, a debate between these two woud have far too much ego-bumping to be watched without a couple of beers first.