cached data more cached data even more cached data

The home of amusing statistical juxtapositions since 1662. Still in beta. (more info)

Is this a coincidence?

graph: Annual Incarceration Rate (USA)

Annual Incarceration Rate (USA)

Author: Rob
Source: CEPR analysis of data from BJS, Census and Calahan (1986)

graph: Unemployment Rate in U.S.

Unemployment rate (USA)

Author: Rob
Source: US Bureau of Labor Statistics via Google

Permalink to this pair | Comment


About

The belief that correlation implies causation is a compelling one. It can drive you to believe that ‘they’ are watching your every move, that nightlights cause myopia, or that the thunderbolts are hurled by a bearded senior citizen atop Mount Olympus.

And sometimes it’s just funny.

Each of the two graphs you see is randomly selected from a selection of about twenty, and occasionally they look eerily similar. When this happens, it obviously means that there is some sort of connection between them.

It would of course work better if the system used some sort of algorithm to return pairs that fit better or are more likely to be funny. This isn’t there now because Rob is not a clever man (rather like this dude).

The idea for this website was stolen from the Redditors Unidan and hypno_beam, as a result of this comment. This story is also available via the medium of badly-drawn vector art.

Submit your own graph

Can you think of a graph that we’re missing? If so, send it in! You can use the convenient form below or e-mail it to coincidenceithinknot [at] gmail.com.








Graphs can be in one of the following formats. In order of preference:

  1. A 350×250 image (preferably PNG)
  2. A graph in a spreadsheet (pretty much any format i.e. .ods, .xls, .xlsx etc)
  3. A series of numbers or data (again, pretty much any format – spreadsheets or CSV are preferred)

Submitted material will appear after moderation. Graphs in formats other than the ones above are fine, but will take slightly longer to appear in the list of graphs. Please only submit graphs which you have created, graphs which you have permission to upload, or graphs which are in the public domain.
If possible, please mention the source of the data used in your graph in the description box.