Numbers don’t always mean what they seem to mean | David Spiegelhalter and Anthony Masters

The Evening Standard reported last Wednesday that the number of Covid deaths rose to 207, making it the deadliest day in almost six months. The number of deaths within 24 hours of a positive test was actually slightly lower than the week before. What went wrong?
It seems that it was due to a common misconception that, even after 18 months, the daily death number represents deaths occurring within the last 24 hours. However, it is only deaths that were reported during that time period, and there are fewer reports on weekends and holidays. Only 50 deaths had been reported on the previous bank holiday weekend. The spike of 207 added to the backlog.

Statistics may not always mean what they seem to. Some statistics don't match our common understanding. For example, politicians might count a new wing for a hospital or major renovation as a hospital. Some are subtler, such as the BBC headline "Covid: Disabled people account for six out of 10 deaths in England last Year", which is accompanied with a photo of someone in a wheelchair. This context defines disability as a long-term limitation of daily activities that is not related to old age. This definition was also fulfilled by 54 percent of non-Covid-19 deaths, according to the Office for National Statistics (ONS). This is due to the observation that Covid victims are the same as those who die from other causes. Covid targets the weak and vulnerable just like normal people. Some people are at greater risk of getting it.

ONS found that the most significant risk to people with severe learning disabilities was due to living in a residential care home or similar facility. Tim Harford, an economist, talks about the dangers associated with premature enumeration. He focuses on the numbers rather than explaining what they mean. We are not alone. We are not immune.

It is best to take a moment to look at the numbers and verify that they are what you believe.