Eurovision Voting: The Statistical Anomalies

When you look at statistics there’s “unlikely”. And then there’s “it just won’t happen”. Fair flipping of a coin will very occasionally have the coin land heads 10 times in a row. It will never land heads 1,000 times in a row. Looking at the voting patterns we have a coin landing heads 10,000 times in a row.

I totaled up how many times each country gave 10 or 12 points to other countries for the last 2 finals (2010 – 2013). I used both 10 and 12 because bought votes will not always be able to overcome a Loreen getting 12 points, but it should certainly be able to pull in at least 10 points. So what do the numbers show? There are 44 countries that participated at least some years. And this is what we have.

Happily Single

Austria, Belgium, Germany, Italy, Montenegro, Poland, & San Marino did not give any country top votes (10 or 12) more than once over this period.

Croatia (Bosnia & Herzegovina), France (Sweden), Hungary (Sweden), Israel (Sweden), Latvia (Russia), Serbia (Bosnia & Herzegovina), Slovakia (Sweden), Sweden (Denmark), Switzerland (Serbia), & the United Kingdom (Ireland) only gave top votes (10 or 12) twice to one other country (listed in parenthesis).

So 17 countries or 39% of the countries participating do not favor any country. This is a clear baseline of what a fair vote should show.

The Egregious Cases

So now let’s look at the worst cases:

  • Albania -> Greece 4/4
  • Armenia -> Georgia 3/3
  • Azerbaijan -> Ukraine 3/4
  • Belarus -> Ukraine 4/4
  • Bosnia & Herzegovina -> Serbia 3/3
  • Bulgaria -> Azerbaijan 3/4
  • Cyprus -> Greece 4/4
  • Iceland -> Denmark 3/4
  • Ireland -> Denmark 3/4
  • Malta -> Azerbaijan 4/4
  • Moldova -> Romania 4/4
  • Norway -> Sweden 3/4
  • Romania -> Moldova 4/4;
  • Russia -> Azerbaijan 3/4
  • Turkey -> Azerbaijan 3/3
  • Ukraine -> Azerbaijan 4/4

Some of these are clear cases of neighbors voting due to cultural affinity. It also shows that Greece is an unappreciating spouse given love by Armenia and Cyprus, without returning it. And something’s going on with Ireland loving Denmark but Denmark isn’t getting this kind of votes anywhere else so this is unlikely to be fraud. (But really Denmark, you need to give Ireland a call.)

And the rest are Azerbaijan. Statistically this is incredibly unlikely. This is the coin flip landing heads 10,000 times in a row.

So how did the 39%, the happily single, vote for Azerbaijan? They’re our guide for a fair vote. 5 of the 17 gave Azerbaijan 1 vote, the rest 0. Over 4 years, across 17 countries that span Europe, 5 of the countries gave Azerbaijan top votes once, 12 never.

The above numbers are enough to show a large enough anomaly that the EBU needs to investigate specifically what occurred with the vote from these 5 countries. And the EBU needs to release the raw vote totals so we can see where these results came from. If the televote in this countries were similar to most other countries and the jury vote has been delivering these results, then there is a major problem with the jury selection.

Two closing thoughts on this. First it is possible that one of these five is a legit case of they love Azerbaijan. The equivalent of the Ireland -> Denmark love affair. What’s statistically impossible is five. Second, there have been allegations that other countries have tried to buy votes. If any have done this, they sure suck at it.

 

To read more of our coverage of the 2013 Eurovision voting scandal, click here.

David T is a Colorado-based correspondent for wiwibloggs.com. You can keep up-to-date on the latest Eurovision news and gossip by following the team on Twitter @wiwibloggs and by liking our Facebook page.

Photo: Eurovision.tv (EBU)