Open data / public intuitionA game in lines

You know the trend.
Do you know the number?

You see the first half of a real chart. Draw where you think it goes next.

Scroll to the chart

The real line is waiting below.

Start here

Draw the continuation.

Draw the continuation. Reveal the truth. Then try the next chart.

Chart 1 of 2UN Inter-agency Group for Child Mortality Estimation (via World Bank WDI)

Child deaths before age five in the world

Deaths before age five, per 1,000 live births. Draw what happened after 2010.

Child deaths before age five in the worldDeaths before age five, per 1,000 live births. Draw what happened after 2010.

Draw from 2010 to the right edge.

Why this exists

The gap is measurable.

Most public misperception research asks people to pick a number from a list. That tells you they were wrong. It does not tell you what shape they thought the world had.

Drawing does. Every line here is stored as a curve - not a single guess, but a whole belief about direction, speed, and where things settled. Aggregated across enough people, that becomes something you cannot get from a multiple-choice poll: a picture of the trend the public believes in, laid directly over the one that happened.

The findings are the point. The game is how they get collected.

The draw-then-reveal format was pioneered by The New York Times in You Draw It (Gregor Aisch, Amanda Cox and Kevin Quealy, 2015). Linework is an independent project, not affiliated with or endorsed by them.

How it works

Three steps, about a minute.

  • 01

    See the trend

    A real series from a government or research source, cut off at the point people usually start getting it wrong. No title giving away the answer.

  • 02

    Draw the line

    Drag straight through to the right edge. Wherever you think it went - steady, falling off a cliff, climbing. Your line is your answer.

  • 03

    Watch it reveal

    The real line prints over yours. Then you find out whether you were wrong in the same direction as everyone else.

From the data

What people get wrong, in aggregate.

  • +18

    points

    Average overestimate of how much violent crime has risen since the early 1990s, when it has in fact fallen.

  • 3 in 4

    players

    Drew the child vaccination rate below the real figure - the single most underestimated chart in the set.

  • −12

    points

    Average underestimate of how far the teen birth rate fell after 2007.

Illustrative figures. Real per-chart findings publish here once enough people have played each one.

Find out how wrong you are.

An account keeps your streak, remembers which charts you have already answered, and adds your lines to the aggregate.