Lies, Damn Lies and Statistics

Mark Twain (popularising Benjamin Disraeli) famously noted there are three kinds of lies: "Lies, damn lies, and statistics." It’s a phrase that has never felt more relevant than in today’s 24-hour news cycle, where the same data set can be used to announce a "roaring recovery" or a "national catastrophe," depending on which headline you click.

The catalyst for this thought was a recent BBC News report regarding the UK’s latest GDP figures. The statistics showed 0.6% growth in the first quarter of 2026.

On the surface, this is an objective piece of data. However, the way it was framed across the media landscape provides a masterclass in how statistics can be used to push polarising narratives.

The Great News Spin: Focusing on the Delta

For those looking to broadcast a message of success, the KPI of choice was the percentage growth relative to peers. We saw in the past headlines declaring the UK amongst the fastest-growing economies in the G7. By focusing on this specific delta (the rate of change), the narrative becomes one of momentum, turning corners, and outperforming expectations. It is a valid use of statistics as growth is inherently better than contraction, but it is a narrow lens that prioritises the "speed" of the car over where the car actually is on the road.

The Terrible News Spin: The Per Capita Reality

Conversely, critics and more sceptical analysts looked at a different KPI: GDP per head. While the total economy grew, the amount of economic output per person actually fell over the previous year. From this perspective, the recovery felt hollow to the average person whose standard of living hadn't improved, and indeed increased cost of living is front of mind. When compared to pre-pandemic levels (2019), the UK’s growth remains sluggish compared to historical trends as the population grows and inflation takes a huge chunk out of any nominal growth. This is also a valid use of statistics, but it is a lens that prioritises the "destination" and long-term context over the immediate progress.

Good vs. Bad Use of KPIs

The difference between a "good" use of statistics and a "bad" one often comes down to context compared to cherry-picking.

Bad Use (Cherry-picking)

This involves selecting a specific starting point to make a trend look better than it is. For example, reporting that "crime has fallen by 10% this year" while ignoring that it is still 50% higher than it was five years ago. Or, when it happens, celebrating that inflation has hit the 2% target without mentioning that this doesn't mean prices are falling; it just means they are rising more slowly on top of the 29% increase we've already endured since 2020 (source: Bank of England).

Good Use (Contextual Transparency)

A healthy use of KPIs involves showing the "Why" and the "How." It means acknowledging the margin of error, showing long-term trends alongside short-term wins, and using multiple metrics to provide a 360-degree view. For instance, reporting on high employment rates is good, but it becomes great reporting when it also accounts for record levels of economic inactivity due to long-term sickness. And perhaps it is represented in a visually appealing graph over an appropriate time period.

The Healthy Sceptic’s Conclusion

Data is not the one ultimate truth; it is a tool used to approximate the truth. In an era of ‘big data’, it is easier than ever to find a number that supports your pre-existing bias.

Does this mean we should stop believing in numbers? Not at all. And we should remain optimistic. After all, 0.6% growth is objectively better than -0.3%. Progress, however incremental, is still progress.

But we must also remain healthily sceptical. When you see a statistic used to justify a "great" or "terrible" headline, ask yourself: Who is presenting this and what are they looking to achieve? What is the timeframe and trend? What had been the expectation? And what metrics are they choosing NOT to show me?

Used judicially, Data forms Information, which becomes Knowledge, and ultimately drives Wisdom. In the end, the most important statistic is the one we use to measure our own critical thinking. Don't let the headlines do the maths for you.