Corrections
We want to stick to our own Statistical Style Guide: “If you get it wrong, admit it.” So here we are, admitting the bits we got wrong.
Page numbers refer to the original hardback and may differ in other editions. Updated October 2026.
Page 25 · Chapter 3 · Sample sizes
Probability intervals
We said: “If you roll 100 times, your 95 per cent probability interval is between eleven and twenty-five: just 10 per cent of the possible outcomes.” It should be 15% of the possible outcomes.
We also said: “If you roll 1,000 times, then there’s a 95 per cent chance that you’ll roll a 7 between 145 and 190 times. Your range of outcomes has narrowed to just 4 per cent of the total.” It should be 4.6%.
Page 56 · Chapter 8 · Causality
Causes come before effects
We wrote: “causes usually have to come after effects”. We meant before effects. D’oh.
Page 71 · Chapter 10 · Bayes’ theorem
The rare-disease example
We got our sums wrong. If you have a million people, 100 of whom have some rare disease, and a test that is 99% specific and 99% sensitive, it would correctly tell 989,901 healthy people that they don’t have the disease; incorrectly tell 9,999 healthy people that they do; correctly tell 99 people that they do have the disease; and incorrectly tell one person that they don’t.
So the chance of having the disease after a positive result is just under 1%. We got several of those numbers wrong in the book and ended up with about 8%, which was incorrect.
Page 73 · Chapter 10 · Bayes’ theorem
Andrew Deen’s conviction
We say Andrew Deen was convicted in 1994 but that his conviction was overturned in 1993, which obviously can’t be right. He was convicted in 1990. We should also note that he was convicted in the retrial, which we hadn’t realised at the time of writing.
Page 74 · Chapter 10 · Bayes’ theorem
1%, not 10%
We mistranscribed a line from the study discussed here. It describes breast cancer in 10 women in every 1,000 — or 1%. We wrote 10%.
The example should use 1% for the proportion of women with the disease. That makes the subsequent answer — roughly a 10% chance of having the disease after a positive test — consistent. Using 10% would instead give about a 50% chance.
Readers of the Polish edition also reported this error in October 2026.
Page 87 · Chapter 12 · Has what we’re measuring changed?
The year was 2020
We said: “Then, on 26 June 2019, several states agreed to include ‘probable’ deaths”. The year should have been 2020.
Page 128 · Assumptions in models
The Treasury, not the Home Office
We said the Home Office modelled the impact of Brexit. It should have been the Treasury.