Scientist's statistical model predicts Sweden to win Eurovision again

As the Swedish capital Stockholm prepares to host the final of the 61st Eurovision Song Contest tomorrow night (Saturday, May 14), Swansea University glaciologist Dr Martin O’Leary has again published his annual predictions of the winners and losers, based on a statistical model.

Dr Martin O'Leary 2Last year, Dr O’Leary, who is originally from County Kildare in the Irish Republic and now lives in Mumbles, Swansea, correctly predicted a win by Sweden’s Mans Zelmerlow with the song Heroes.

This year, Dr O’Leary, a research officer with Swansea Glaciology Group in the University’s College of Science, predicts Sweden will win again with entrant Frans and his song If I Were Sorry

Dr O’Leary’s model, which uses Bayesian algorithms, predicts Austria, Ukraine, Australia, and Russia will be hot on the heels of Sweden as the countries most likely to win, with the UK’s entry, Joe and Jake with their song You’re Not Alone, as 14th most likely to win.

The Eurovision fan has closely analysed the results in the contest for the past four years using technology and the skills developed through his day-to-day work as a climate change researcher, looking at interactions between glaciers and the oceans, particularly on iceberg calving events (when chunks of ice break off at the end of a glacier).

In 2012 his model correctly predicted Sweden’s Loreen would win with the song Euphoria, and the model was just one off in 2013, predicting a win by Azerbaijan’s Farid Mammadov with the song Hold Me, who came second.  

In 2014, the model favoured Sweden’s Sanna Nielsen with the song Undo, which placed third in the final, with Austria’s Conchita Wurst taking first place with Rise Like a Phoenix.

In 2015, Dr O’Leary introduced changes to how his model works and there was one notable change to the contest, which again affects this year’s Eurovision.

In a change from his previous predictive models, Dr O’Leary didn’t use betting odds to determine song quality.

“Although this was a really appealing idea, it didn’t really work very well in 2014,” he said.  “The resulting predictions were both wrong – not necessarily a problem – and far too certain – definitely a problem.  I tried some tweaks to reduce this problem, but the best solution seemed to be to remove it completely.”

The key change to the contest last year for Dr O’Leary’s model was the inclusion of Australia as an entrant for the first time, to celebrate the competition’s 60th Anniversary and it’s ‘Building Bridges’ theme.  Australia has been invited to participate for a consecutive year, with Dami Im representing the country with her song Sound of Silence.

“From a modelling perspective this is a bit annoying, because we have very little historical data on Australia compared to the other European countries,” he said.

“According to this year’s model, they’ve still got a very respectable shot at the title. They've got a strong female solo singer, which is usually popular with the juries, but it's also a solid pop song which should do well on the televotes.”

Dr O’Leary, who was last year featured by Wired after he recreated the frozen continent of Antarctica in Minecraft, has previously worked at the Department of Atmospheric, Oceanic and Space Sciences at the University of Michigan and at the Scott Polar Research Institute at the University of Cambridge, where he achieved his PhD.

He said: “Predicting Eurovision means looking through the mess of individual votes and trying to understand the patterns that emerge. It's not always certain what the reasons are behind any one vote.

“Bayesian algorithms provide a way to work with large quantities of uncertain information, and see the rules which underlie it.

“The same tools which reveal the secrets of Eurovision can also help us to look beneath the ice sheets of Greenland, and help us to understand the complex processes which are driving climate change.”

You can follow Dr Martin O’Leary on Twitter @mewo2, or visit his dedicated Eurovision web site here.