There are good books, very good books and then there is The Book of Why.
This is definitely a book like no other (at least when comparing with the ones I’ve already read). It did just not increased my knowledge as it made me saw the world in a different way.
The main theme of the book is the importance of the causality and effect in almost everything. Until a few years ago, these two concepts were like tabu to statisticians and to scientists in general. But then, due to enoumerous factors, in which new researches and the publication of scientific papers are included, the game changed and we now see, gradually, causality and effect taking higher importance day after day.
The author structures the book in a pragmatic way. In the beggining, presents to us the ‘Stairs of Causality’, which contains 3 steps. Each one is introduced and discussed in detail. The list of examples and real-life cases is huge and I must confess this was my favourite aspect of the book.
And talking about favourite, my favourite chapter was definitely the one where the Simpson’s and Monty Hall paradoxes were explained. The first is a little more complex to explain than the second, as it envolves not only theory but maths. It has, though, a lot of real-life examples, some of them responsible to ocupy my mind during some time. The second, my favourite paradox, which I already knew, was explained like I’ve never seen. I’ll explain it now so that it can become to this book as a trailer is to a movie.
It all began in a TV contest, where participants had to choose one of three doors, without knowing what was behind each one. The only thing they knew was that behind one of those doors was a car and behind the other two there were two goats. After the first selection, the presenter opened a door with a goat behind (it didn’t matter which door was selected as there were two goats; if the participant selected the door with the car, there were two to the presenter choose from; if the participant selected the door with one goat, there was still another one behind one of the remaining two doors) and asked “Do you want to change the selected door?” The question that became the base of the paradox is “What is the probability of winning if the participant answers yes?”. Some of you might be thinking “Well, it’s 50%, without discussion, because there are two doors to reveal, one that contains the car and the other one that contains the other goat”. But now comes the interesting part: the answer is not 50%, but 66%. Why? That’s simple. First, imagine the number of possible combinations. There are 3: the car being behind door number 1, 2 or 3, considering the goats are indistinguishable. Now, for demonstration purposes, suppose the participant always selects door number 1, no matter where the car is (in fact, the participant doesn’t know…). If the car is in door number 1 and the participant answers ‘yes’, then looses the game, because the change made him/her select the remaining goat. But in the other two cases (car behind 2nd and 3rd doors), if the participant answers ‘yes’, he/she will win, because the selected door was hidding a goat and the presenter opened the door that was hidding the other, so the other door, the one the participant then choosed, is hidding the car. That said, the answer ‘yes’ represents winning every 2 out of 3 times (2/3 = 66,(6)%). Surprised?
That said, I couldn’t finish this book review without saying that the last chapter was special, as it represents the vision of the authors about Big Data, AI and the future of these fields.
To conclude, I would like to congratulate the authors for their courage and bravery of pursuing the investigation in this area even when the times were not favorable. Their behaviour resulted in this magnificent book, which I just do not rate 5 out of 5 because sometimes it requires the reader to re-read explanations and math formulas to understand the point of view of the authors, which can make it a little difficult to a person without a background in maths and science to understand everything. Still, I truly recommend this reading to everyone, no matter the field of expertise.
Rating 4.7/5
Read in 2021