THE TIGER THAT ISN’T Seeing Through a World of Numbers by Michael Blastland and Andrew Dilnot

SIMPLEXITY: The Simple Rules of a Complex World by Jeffrey Kluger

THERE ARE THREE KINDS of lies – lies, damned lies, and statistics. That famous statement – attributed by Mark Twain to Disraeli – on the potential of number-crunching to mislead nestles at the heart of Michael Blastland and Andrew Dilnot’s clear-eyed and concise offering.

They are producer and presenter of Radio 4’s More or Less, a programme that has been exposing the misuse and misrepresentation of statistics and figures for years now, so they know a thing or two.

These days, politicians, journalists and scientists continually spout endless streams of numbers and figures to support policies, enhance news reports or explain research, without really understanding what they are talking about. Yet these statistics mostly go unchallenged, as if the mere fact that numbers are involved implies the absolute certainty of what they are saying.

It doesn’t, of course. When the Government announced that it would spend £300 million on extra childcare, did anyone calculate that this equated to only £1.15 per child per week? What seemed a big number turned out to be paltry when broken down to a human scale.

Similarly, when it was hysterically announced on the BBC that “for every alcoholic drink that a woman consumes, her risk of breast cancer rises by 6 per cent”, did anyone stop to think, hang on, that’s garbage? It implies that a woman going out on a weekend bender is guaranteed to get breast cancer. In that case, the increased risk was 6 per cent if the woman had a drink every day of her life. With the base-line risk of breast cancer at around 9 per cent, that means that the actual increased risk is 6 per cent of 9 per cent, or just over half a per cent.

The Tiger That Isn’t is full of such examples of garbled logic, ranging from glaring to more subtle misrepresentations of information on the part of politicians, the media and scientists. Looking at subjects such as averages, targets, sample sizes and faulty comparisons, Blastland and Dilnot expose flawed reasoning and explain how it came about.

Sometimes they do let the culprits off the hook, tending towards the view that those who misuse statistics do so out of ignorance. This may be true in some cases, but undoubtedly there are often much more sinister motives at work.

Governments clearly deliberately manipulate numbers to suit themselves, while journalists jump on the most shocking results they can find, rather than the most likely ones, in order to sell newspapers. And scientists are not innocent either – the more headlines they can grab with a piece of research, never mind the veracity of the results, the more likely they are to get a grant for further research.

The only down side of this book is its rather smug tone. Blastland and Dilnot are, perhaps, not natural authors, and may be more suited to radio than the written word. There is a hint of sanctimonious righteousness that does the book no favours – really, the authors should have let the numbers speak for themselves.

The central premise of Jeffrey Kluger’s entertaining Simplexity would doubtless be closely scrutinised by Blastland and Dilnot. Essentially, he suggests that simple things can be more complicated than they seem at first, and complex things more simple.

It’s a neat idea, but not one that finds much supporting evidence within the book. Simplexity is a trawl through the science of complexity and simplicity, much of it based on work carried out at the Santa Fe Institute in New Mexico – a hotbed of blue-sky thinking where scientists from across the disciplines swap ideas about physics, biology, computing, social sciences and much more.

Across them all, an array of similar systems can be found, ranging from absolute chaos on the one hand to absolute robustness on the other. Complexity occurs on the curve in between, and the SFI spends a lot of time trying to model such systems mathematically.

Kluger leads us through such diverse topics as the stock market, human evolution, traffic congestion, warfare strategy and how babies learn to speak, with an easygoing and readable style that reveals plenty of juicy nuggets of information, but doesn’t add up to a defence of his opening statement.

The problem is that, although the basic idea behind a mathematical model of a complex system might be simple, the more you try to tweak it to represent real life more accurately, the more complex it becomes. Trust me, I know, because I used to do mathematical modelling for a living, and it was a pretty complex task at times.

My quibble with Simplexity is not with the book itself, rather with its unsubstantiated central claim. While some chapters don’t work as well as others (notably those on art and sport), in general this is a finely written and well researched look at how scientists are tackling the mathematical representation of complex systems in new ways. Not simple, then, but a lot of fun, nonetheless.

Read exclusive extracts from The Tiger That Isn’t in Times2 next week

THE TIGER THAT ISN’T Seeing Through a World of Numbers by Michael Blastland and Andrew Dilnot

Profile, £12.99; 173pp

Buy the book here at the offer price of £11.69 (free p&p)

SIMPLEXITY: The Simple Rules of a Complex World by Jeffrey Kluger

John Murray, £16.99; 244pp

Buy the book here at the offer price of £15.29 (free p&p)

Michael Blastland and Andrew Dilnot are at the Edinburgh International Book Festival on Monday August 20 at 7pm. Call 0845 373 5888 www.edbookfest.co.uk

Times Recommends

• The 100 Best Books of the Decade
• The Times picks the 50 best paperbacks of 2009
• The conversation: James Ellroy