1 Comment »Posted on Saturday 20 June 2009 at 7:45 pm by Jacob Aron
In Getting It Wrong, Mathematics

Cliff Arnall is the king of the “formula for” story. Earlier this year I wrote about his equation for calculating the date of Blue Monday, his self-styled worst day of the year.

At the time I failed to mention that Arnall actually trots out this rubbish not just once, but twice annually. When summer rolls round, it’s time for the happiest day of the year, which according to Arnall’s formula was yesterday.

The “story” was picked up by the Telegraph, Daily Mail, and Sun. Fact-checking obviously doesn’t occur on the happiest day of the year, because it seems that Arnall is still dining out on Cardiff University’s reputation, despite the institution making it very clear he only worked there as a part-time tutor.

I suppose its time to take a look at the formula now, but by this point do you really need me to tell you it’s nonsense? Here, in all its glory, is the “complicated equation” needed to calculate a day’s happiness rating, along with the variable definitions:

O + (N x S) + Cpm/T + He

  • O: Outdoors
  • N: Nature
  • S: Social interaction
  • Cpm: Childhood memories of summers
  • T: Temperature
  • He: Holidays

Not sure about the difference between outdoors and nature, and surely the value will be the same for each day; O = N = 1, unless there is a second outdoors that I don’t know about. Social interaction could actually be quantifiable, perhaps the number of conversations in a day, but it’s pretty unclear.

Cpm and He are both very bad notation. What is wrong with just C and H? The extra letters don’t add anything, they aren’t even an abbreviation, but they could easily be confused for additional variables. I guess this way looks more “scientific”.

In fact, the only scientifically measurable variable, temperature, is what makes this “formula” fall apart. Assuming you have at least some memory of your childhood, Cpm/T will rapidly grow to infinity as the temperature drops to 0 °C and completely dominate anything else in the equation.

I don’t know about you, but I thought it was pretty warm out yesterday. It seems that Arnall’s Blue Monday, January 19th, would be a much better candidate for happiest day of the year according to this formula. Maybe he accidentally got his bullshit mixed up with his bollocks, and gave us all the wrong date. Now that’s a thought that makes me smile.

  1. One Comment

  2. Ha!

    I hadn’t actually seen his equations – I’d assumed “blue monday” etc. were a statistical “result” from adding a load of time-series predictions together badly, but obviously not.

    It’s moments like that that make me glad I work in an Advertising/PR company (as mathematician/programmer) – at least I can make sure one company never releases anything like that.

    By Tim Wintle on Saturday 20 June, 2009 at 11:29 pm

Sorry, comments for this entry are closed at this time.