Thinking, Fast and Slow

• There is a fast thinking System 1 which is used most of the time and acts immediately and a slower working System 2 which does more analysing
• There is e.q. the need to always do pattern matching and find a reason behind something - like why a stock price changed
• The halo-effect is described. If a teacher reads two essays from the student and the first one is better, then the second one is getting a better grade even though it could be shit
• The law of small numbers is interesting. In smaller sample sizes, the outcomes are likely more extreme. E.g. smaller schools are more likely to be way worse or way better than bigger schools - not because the schools are actually better, but because there are too few people for a good statistic.
• The Anchor effect is also explained. When a question has an anchor like "is this house worth 100k?", then the people will start their estimate from the anchor (100k), so for a real estate agent a higher anchor is important
• Simple formulas are statistically better than complex ones or intuition
• As an entrepreneur or risk-taker in general, you have to be an optimist. Statistically you will most likely fail, that's why only optimists are founding companies - if they would be realists or pessimists, they wouldn't even start their endeavour.
• When a company gets acquired, the acquiring company normally trades lower on the stock markets. That's because usually the managers of the company which buys the other think they can manage the acquired company better than the previous management which is most of the time simply not true and the Wall Street knows this fact, that's why this discount happens.
• Management of a big company is normally overvalued. The CEO of a public traded company has most of the time a lesser impact on the company than other factors like competition and current market.

The Goodreads "star" method of reviewing lends itself to a particular fallacy. I'll call it the self-selected evaluator's bias. I've mentioned this before in other reviews but this seems to be the place it belongs and (statistically speaking) you probably haven't read any of my other reviews. (Come to think of it, statistically speaking you aren't reading this one, but I digress.)

Imagine someone writing a review of Star Wars. If you bothered to watch it in the first place and are now bothering to write a review of it, I can predict that you liked it a lot and that you generally enjoy Science Fiction. You are not a random reviewer, but a self-selected biased one. (You could argue with this--I found myself continually arguing with Kahneman's certain pronouncements throughout reading T, F & S as I will hereafter refer to the book--but you'd still have to admit, my example may not be the best but the bias I describe exists.) The end result is that books tend to be overrated unless they are so horrible (or attract the wrong audience by poor marketing) that a fan is offended in some way.

But no one believes the ratings of books are some kind of science, do they? What about people who are evaluating their life? Or their degree of pain? What of people who think the metaphor of a hedonometer to measure happiness makes any kind of sense? I think the measurement of the unmeasurable, the disguising of the ineffable as a number and passing it off as precise and doing statistics with it, are offensive. Yes, I am offended by this kind of thing and so, by the halo effect, I am underrating a book which contained a lot that I liked because of this feeling. My dislike extended to the book as a whole.

And, also, I am helping the star rating system by my negative "correction."

The author should (if he hasn't already) go out and read Weapons of Math Destruction (the "insert book/author" feature seems not to be working for me at the moment--convenient since I wish to complain about the belief in the book's advocating algorithms as preferable to human decision making and the computer code is failing to work) for an opposing point of view.

To some extent, this book has rediscovered Freud. Sigmund has been much maligned lately so maybe some of his key discoveries could use a repackaging with out his name on it. System 1 is similar (not identical, mind you) to the Id and System 2 to the Ego. Or maybe System 1 to what Freud called The Unconscious and System 2 to the Conscious. Freud's discovery that the conscious mind is not what rules our lives is akin to this book's "discovery" that fallible System 1 makes many more of our "decisions" than System 2 which we think of as "us."

I also wish to point out that the distinction T, F & S makes in the discussion of the "framing fallacy" in which it is asserted that framing makes us lose touch with "reality" that much, if not all, that passes for reality is just the current most popular frame. There are many such frames in the book masquerading as fact, the measuring of the unmeasurable being just one example. The book is so extremely culture-bound that it never occurs to Kahneman in his example of how much one enjoys one's car, that some readers (e.g. me) don't have a car. He probably never realized that, back in 2011 when this book was written, referring to someone as a "paraplegic" would be criticized in 2017 as politically incorrect.

Lastly, far from his audience being watercooler gossips, I imagine it will become a must-read for spin doctors and advertisers. I am reminded of the witticism of someone on reddit who said:
It's not enough to have a threat model anymore; if you're not
asking yourself at the earliest planning stages how something
can be repurposed to hurt people, and putting design and
engineering effort into preventing that from happening, you're
building a weapon.


All this said, I recommend this book to libertarians. Especially any who may be members of my family.