The idea is to put the reader through the kind of brain-taxing maths-based problems that are given to physics candidates applying to Oxford University. And some of these are genuinely entertaining. In particular I found the sections on logic problems, perpetual motion machines and estimating highly enjoyable - the estimating section consists of what are often known as Fermi problems, though Thomas Povey seems not to have heard of that name. (There is a whole book of these called How Many Licks.)
What I found myself doing was reading the problem, having a think about what the shape of the answer might be and then flicking though the answer without reading it in detail. If I'm honest - and this is probably why I never made it as a real scientist - I didn't really care what the actual answer was. That just seemed like grunt work. But thinking around the problem was genuinely stimulating.
However, I did find a number of the topics - geometry and various areas of mechanics for instance - sufficiently dull that even getting a vague idea of the direction that should be taken was rather meh. It's a shame that there weren't more genuinely interesting topics. Now, admittedly by limiting topics to those that high school students should know there is a natural tendency to the duller subjects, but the perpetual motion section showed you could make basic mechanics and energy considerations approachable - it's just a shame there weren't more exotic interpretations like that.
Overall, then, I surprised myself by getting more out of the book than I thought I would, and despite expectations, I don't think I will have nightmares as a result of reading it either. I even had the delight of having recently researched one of the estimates that Povey uses in his Fermi problems, and could feel a little smug as he was almost an order of magnitude out (as long as you consider Americans rather than Brits). In the spirit of the book, I'm not going to tell you which estimate it was, or why there are special circumstances that make the answer in the book closer to correct than it should be.
However, this book certainly isn't for everyone who would read a conventional popular science book. I'd go as far as to say that it's not for most popular science readers. But if you fancy doing physics, maths or engineering at university - or wish you once had - it is an absolute must-have buy.
Review by Brian Clegg