Published on 2026-07-25
Why do two unrelated lines suddenly move together on a chart?
Two lines can look perfectly in sync on a chart for one simple reason: each has its own hidden scale, stretched until they match. Here's how that trick works.
A statistician once built a website purely to prove a point most of us feel but rarely see demonstrated: he plotted the number of Nicolas Cage films released each year against the number of people who drowned in swimming pools, and the two lines climbed and dipped together almost perfectly. Tyler Vigen, the mind behind the 'Spurious Correlations' project, says it plainly on his own site: what he shows is mathematically true, and it is intentionally misleading. The confession is the whole lesson.
Two lines, two rulers, one illusion
Here's the mechanic hiding under nearly every eerily matching chart. When you put two different measurements on the same graph, films released and drownings, temperature and ice cream sales, spending and anything else, each measurement usually gets its own vertical axis, because the numbers live on completely different scales. And that's where the trick lives: whoever draws the chart chooses where each axis starts, where it ends, and how much space each unit takes up. Stretch one axis a little, compress the other a little, and two lines that have nothing to do with each other can be made to rise and fall in what looks like perfect step. Vigen's own site admits to truncating the axes on its graphs for exactly this reason: it's the fastest way to make two unrelated trends look like partners.
When the trick isn't a joke
Vigen's project is a wink, built to teach the lesson safely. The same device shows up in earnest, and a widely shared 2023 chart from the Federal Reserve Bank of St. Louis is a clean example of how. It compared the military budgets of six countries on one chart, with China and four other nations measured against a left axis running up to 300 billion dollars, while the United States alone was measured against a separate right axis running all the way up to a trillion. Because both lines were squeezed to fill the same visual space, China's line appeared to climb toward the top of the chart right alongside the US line. Newsweek's fact check laid out the real numbers behind that picture: US military spending stood at about 767.8 billion dollars in 2021, China's at about 270 billion, a gap of roughly half a trillion dollars that the shared visual space had quietly erased.
Two lines moving together on a chart tell you that someone chose two rulers. They don't yet tell you that the two things being measured have anything to do with each other.
That chart travelled fast once people started sharing it, precisely because it did what a good illustration is supposed to do: it made a comparison feel instantly graspable. The trouble is that the feeling it produced, two rising powers neck and neck, wasn't the one the underlying numbers supported. Nobody needed to alter a single figure in the dataset to produce that impression. They only needed to decide, quietly, how much visual room each figure was allowed to take up.
What actually deserves your trust
None of this means every chart with two axes is dishonest. Plenty of dual-axis charts are built carefully, with sensible ranges chosen to inform rather than to flatter a story: a company genuinely might want to show units sold alongside customer satisfaction scores, two measurements that live on entirely different scales for entirely honest reasons. The problem isn't the second axis itself, it's that a second axis hands the chart's author a second dial to turn, quietly, without you ever seeing the setting. A single-axis chart, where every line answers to the same ruler, can't play this game: if two lines cross there, they cross because the values actually met.
So when a chart shows two lines moving in eerie lockstep, the two seconds worth checking are simple. Read both axes, not just the shape. What number does each line start and end on? If one axis spans a few hundred and the other spans a trillion, the lines were never measuring anything comparable to begin with, no matter how neatly they overlap on the page. And even when the two axes are perfectly reasonable, remember what Vigen's whole project exists to prove: two lines can march together for years by pure coincidence, with no hidden mechanism connecting them at all. Overlap is a starting point for a question, never an answer by itself.
Sources
- Spurious Correlations (Tyler Vigen) (tylervigen.com)
- Fact Check: Does China Spend More Than U.S. on Defense? (Newsweek) (newsweek.com)