This year, we had a 50% jury system to weed out the bias that you get when actual people choose which songs are the best. Did it work, though?
Here's a map:
What it shows is who voted for whom. I've only marked the top 3 votes (12, 10, 8) from each country, and have not included any votes for the countries in the top 5 (Norway, Iceland, Azerbaijan, Turkey, UK) on the basis that these probably were the best songs. Countries in green had an entry; countries in red had a vote but didn't make the final; countries in yellow didn't enter this time round; countries in grey have never entered.
What's immediately obvious is that there are still regional biases. What seems to happen is that the countries in a bloc vote for the best song in their bloc. For example, the countries that used to be Yugoslavia went for Bosnia Herzegovina, which was indeed the best offering (I myself liked it); the other Balkan countries went for Greece, though. Estonia benefitted from being part of the ex-Soviet bloc and proximity to the Nordic bloc, but Iceland and Norway took away most of the top marks or they would have done much better; their real battle, though, was against Lithuania, because the better of the two would be the one to pick up the bulk of the shoo-in votes to which the region is entitled.
There were some howlers, of course. Spain picked up 20 points, of which 12 were from Andorra, and no-one was shocked by Cyprus's customary 12 for Greece. Even the top 5 countries had different voting patterns: the UK made its total from mainly 3s and 4s (presumably from the songwriting juries, who know Andrew Lloyd-Webber's pedigree), whereas Turkey was a love-it-or-hate-it prospect with 6 12s, 3 10s and 15 0s (one of those 12s came from the UK, gawd knows why; one of the 10s came from Germany, which has a large Turkish immigrant community and always returns a high mark for them).
So, the jury system probably did smooth out some of the worst excesses, although they still remain.
Norway would have won anyway.