You probably know about Nate Silver from his work on FiveThirtyEight, the popular polling analysis website.
In The Signal and the Noise, Silver talks about the history of polling and prediction, from basketball to weather, earthquakes, politics, economics, chess and more. Silver explains the common failings that apply to predictions, and how humans in general are just not very good at understanding uncertainty.
One of the pervasive risks that we face in the information age, as I wrote in the introduction, is that even if the amount of knowledge in the world is increasing, the gap between what we know and what we think we know may be widening.
Bayesian statistics takes centre stage throughout, with frequent examples showing the application to real world scenarios and showing how it's usually better than other statistical methods.
The Signal and the Noise provides a great overview of how statistics is used in the real world, and where breakthroughs are being made. It also highlights common mistakes practicioners make and how susceptible we are to making mistakes in how we approach uncertainty in the real world.
I'd absolutely recommend this to anyone interested in learning more the real world applications of statistics.