One can only read so many Game of Thrones novels. Not because they are too violent (Oberyn Martell’s death, AMIRITE?). But literally, because George RR Martin won’t just give the people what they want- which is ALL the books, ALL the time. With only the gods, old and new, aware of when his next book will be published, it is necessary to seek other reading opportunities. Being learned people, I offer to you, the faithful readers of this blog, the first of (hopefully) many posts about engaging, interesting books about statistics.
I know what you are thinking: wait- what? Books about statistics? What is even happening right now? They are not “statsy” like you get reading a text book. The books collected in this series of posts represent an informative, but engaging lot. Some might even make the argument that many of the titles that will eventually be included in this list would be appropriate for undergraduate student consumption.
You may recognize Nate Silver as the guy that who accurately predicted the outcomes of 49 of 50 states in the 2008 presidential elections in 2008 and went 50 for 50 in 2012. Prior to that, he attracted quite a lot of attention for developing PECOTA, a system for forecasting and predicting the career development of Major League Baseball players.
The University of Chicago grad was laid over during travel when he got the idea for the blog FiveThirtyEight (which gets its name from the number of electors in the US electoral college). Since, FiveThirtyEight has grown into a smart source of info a broad range of topics- from the 2016 elections, to predicting Academy Award winners, and sports. In the world of news and internet punditry, he comes across as a sensible, calm voice of reason, amidst a lot of people losing their minds over the most recent political and election development.
He makes the argument in The Signal (and in other places) accurately predicting outcomes in elections means you have to have some way of accommodating for and correcting bias in polls. This means that in order to adequately understand risk and probability, you need to accurately assess the extent to which estimates of sample outcomes accurately represent the reality from which they were taken.
Silver discusses several examples that help challenge traditional thought when it comes to understanding probability, risk, estimates, and assessments. He talks about the series of bad estimates and predictions that caused the 2007 housing and mortgage crisis, how so many television pundits got the 2008 presidential election of Barack Obama so wrong, the promise and pitfalls of “big data”, and the epic battle between chess grand master Garry Kasparov and IBM’s Deep Blue computer program. He has a very lovely way of breaking down very complicated concepts into relate-able and understandable knowledge.