The Expected Goals Philosophy
Authors: James Tippett
Published: 2019, UK
About the Author
James Tippett Currently works as a commercial executive for a gambling company, oddschecker.
He previously worked for a company called smartodds which this book revolves around. He didn’t mention any experience inside a sports team although the book extensively covers Brentford, Matt Benham. Benham owns both Smartodds and Brentford FC.
If you wanted a better understanding of expected goals/points, I would veer towards Sprawlball by Kirk Goldsberry (Review | Buy) which is basketball based. It is a far more visual and rich book, although similarly repetitive.
Who is it for?
This book seems targeted at people who want to learn more about expected goals although it doesn’t really fit that in the end.
I would say this book would be of interest to football (soccer)/Brentford FC fans. There is not really enough depth in the insight to be useful for gamblers, coaches or data analysts in particular.
What is it about?
This book is supposed to be about an insight into expected goals. It doesn’t really give you an insight into what it actually is. It glosses over the detail of how it’s calculated and the strengths and weaknesses of different company’s models.
What the book does well is bring to life different uses for Expected Goals. How gamblers use it, how coaches and scouts can use it.
The book shows the example of how Brentford use expected goals find undervalued leagues to find cheaper players that they then intend to sell on for much higher prices. It makes the point that even Ronaldo doesn’t score hugely above his xP every season and in fact, the difference between Ronaldo and his peers is that he creates more of those chance, hence increasing his expected goals.
What did I take from it?
I took some uses of expected goals from the book. BF Sports Analysis have built an Expected Points model for Rugby for our clients. Our model from a team point of view is to see how efficient teams are from certain scenarios and see if there is an area for improvement. What we hadn’t looked at yet is a team’s total xP as an indicator of performance by itself. So now we researching if a team has a higher xP, does that indicate strong performance, likelihood of winning or correlate with strong phase play in rugby?
Should I buy it?
Probably not. If you are a Brentford fan or interested in sports gambling you might be but I would probably stray away from this book otherwise. I didn’t enjoy the tone of the book which was almost discussing the stat as a cult like figure.
If you’re looking to learn and understand the “Expected goals/points/shots gained” style models, I think you would be better off reading Sprawlball or reading an online blog of it if you wanted something related specifically to football/soccer.