An Ordinary Chart That Does Everything Right
July 1, 2026 • 6 min read
Came across this chart from The Washington Post's Department of Data, plotting how Americans rate songs against how old they were when each song came out. The finding is simple enough to say in a sentence: we love the music of our teens, and our ratings slide from there for the rest of our lives.
What makes it worth slowing down on is how ordinary it is. It's a scatter plot with a trendline, close to what Excel or Tableau would hand you by default, and nothing is plotted in a clever new way. The clarity comes from a handful of small choices, and those choices are what turn a routine chart into something that reads like a story.
![[object Object]](/_next/image?url=%2Fblogs%2Fmusic-nostalgia-chart%2Fchart.jpeg&w=3840&q=75&dpl=dpl_H6Bngu9wq9UiLZVSu7MrBbDSjVra)
How it's built
Each faint dot is one song age, the gap between when a song came out and how old the listener was at the time. The horizontal axis runs from the years before you were born, through birth at zero, and out into the years you were alive. The vertical axis is a rating centered on zero, rising for songs people liked more and dropping for songs they liked less.
The dark line is the one real piece of processing on the chart, an 11-year moving average that smooths the scatter into a readable trend. Take it away and you still have an ordinary scatter plot that anyone would recognize. The chart type was never doing the heavy lifting. Almost everything that makes this land comes from how it is labeled.
The takeaway title, and the subtitle under it
The strongest thing on the page sits right at the top, before you reach a single dot:
All good music came after I was born, and before I hit 35
How Americans rate songs, based on their age when the song came out
The bold line is a headline that names the takeaway. Rather than describe the data, it hands you the conclusion, so you start reading already knowing the point the chart is about to make. The lighter line underneath is the descriptive subtitle, and it fills in what the headline glides over: what is being measured, and against what. Together they read like the opening of a short article.
It's a formula, and a dependable one. Newsrooms reach for it constantly:
- Takeaway title: the one sentence you'd say out loud if you had to sum up the chart.
- Descriptive subtitle: what is being measured, over what, so the claim can be checked.
Most charts built inside companies do the opposite. The title names the data, something like "Song ratings by age," and the reader is left to work out why it matters. Naming the takeaway up front is the cheapest storytelling upgrade available to you, and it costs a single sentence. Say the point, then let the chart back it up.
Who is it drawn for?
Every chart is pulled by the same few forces: the audience it is for, the amount of detail it can carry, and the tone it needs to strike. Audience is the one people skip past, and it quietly sets the other two.
Each step out is a wider, less familiar audience.
The innermost circle is the people who know the data itself, close enough to read the pattern straight off a bare chart. A little wider are the people who know the context but not the numbers. Wider still, and you reach a general audience with no background and a few seconds of attention. Every ring out is a step away from shared context, and a step you have to make up for somewhere on the chart.
A newsroom is aiming at that outer ring, which is the hardest case to design for. The wider the audience, the more of the interpreting you have to do on their behalf, because you can no longer count on anyone pulling the pattern out on their own. The same shift happens the moment a chart leaves the people who made it. Something that reads perfectly to the analysts closest to the data can lose a room one ring out that needed far more spelled out.
Annotations that read the chart for you
This is where the audience choice starts paying off. Instead of leaving the shape of the line for the reader to decode, the chart sets its conclusions right beside the data they describe. Where the trend climbs, the note reads "We like the music of childhood and love the songs of our late teens." Where it falls away, "But our ratings turn negative in our mid 30s and keep falling."
People don't always know what they are meant to notice, and a callout placed right where the eye already is gives the pattern its meaning on the spot. Someone who would never pull a trend out of a scatter plot still comes away with the story. That is the heart of data storytelling: put the sentence next to the shape it explains.
Two quieter touches pull in the same direction. The raw dots are dropped to a low opacity, so the underlying data stays visible without fighting the trend line for attention. And the line is labeled where it sits, "Dark line shows the 11-year average," so nobody has to look away to a legend and decode a color before they can read the chart.
The axes do you a favor
Two axis decisions are easy to miss and worth stealing. The first is the vertical scale, which reads outward from a labeled zero: songs people rated higher go up, songs they rated lower go down. That kind of directional axis is uncommon, and it spares the reader from working out what a bigger number is supposed to signify.
The second is the horizontal axis, a small choice that heads off a real stumble. The obvious way to plot this is one axis running from negative years up through positive ones. But a negative age asks the reader to do a little translation every time they glance at it, and with an unfamiliar measure that friction adds up fast. The chart instead counts down to birth and then back up through age, split cleanly into "Years before birth" and "Years old," with zero marking the moment you were born.
No one has to read a minus sign as "before I was born." The label carries that for them. It is the kind of detail that only ever helps, and it costs nothing to get right.
What I'd push on
There is little to push on here, and my one note is minor: the color. Plain blue is the color every charting tool reaches for first, so a chart this well made ends up looking a touch more generic than its ideas deserve. If the goal were to stop a scroll, a more distinctive color on the trend line would earn more attention without changing anything else. It's a small design take, and not a necessary one.
None of that undercuts the lesson. A default scatter plot, a title that names the takeaway, a subtitle that keeps it honest, notes sitting on the data, and two axes that save the reader a translation. No new chart type required.
