How do you solve a problem like Imperija?

The Macedonian Eurovision song selection is normally a much more sedate affair. Either they have a long drawn out national final and everyone cries fix but doesn't really care too much, or else they pick Kaliopi and everyone worships at her feet.

This year, Vlatko Lozanoski and Esma Redžepova were a popular choice, but no sooner had 'Imperija' been presented in a live show on Macedonian TV than complaints began to fly in the national media, and a few days later we learned that the song might be changed. Earlier this week, it was confirmed by MKRTV that this would be the case, and on Friday we will hear the new song for Macedonia.

So what happened? It's not at all clear, but let's try and explain.

Firstly it was said that the song would undergo 'changes in the arrangement and lyrics'. The reasons for this weren't particularly clear - the lyrics themselves aren't particularly political or controversial and the arrangement is fitting for the song, however the decision was made after "clearly stated views and contructive suggestions of experts and Macedonian audience".

MKRTV were apparently unhappy with the standards set in the video in particular; they'd let the artists and video producers have a completely free rein this year, and the resulting product was not 'dignified' enough to be presented to a European audience. It appears they were generally unhappy with the whole thing from the moment it was completed, but with a show to broadcast, had no choice but to run with the original song and video for the presentation.

The major bone of contention was the prominent appearance of buildings and monuments from the 'Skopje 2014' project, a government building program. Skopje 2014's aim is the construction of museums and other buildings, as well as monuments, in a neo-classic style, which has been strongly criticised for "constructing nationalistic historicist kitsch" as well as just generally being very expensive and a waste of money in what's not exactly a cash-rich economy. In effect, the whole video plays as a propaganda piece for the achievements of the current government, made up of a nationalist, right-wing party (VMRO-DPMNE).

It's perhaps somewhat unusual and unexpected that MKRTV would speak out so publicly against the government in this way, which leads me to suspect there's somewhat more to this than is being said. Certainly, even if the video were the root cause, there'd be no need to change the song itself.

However, that is what's happening. One of the reports wrote that Lozano had threatened to leave the project if the song wasn't changed, and indeed it seems he's got his way. The new song will be written by Darko Dimitrov and Lazar Cvetkovski (the latter is Lozano's own producer) with Simeon Atanasov, the composer of 'Imperija'. Esma herself will reportedly sing in the Romani language this time, rather than in Macedonian, and the song will have much stronger influences from the world of gypsy music after criticism of the 'ethno-motifs' in 'Imperija'.

So, who knows what really happened? One of the articles mentioned 'Eurovision representatives' being involved in the discussions, so perhaps there was an element of the EBU that disapproved of the blatant political posturing - though for once it seems it wasn't due to annoying the Greeks! There's certainly more to it than meets the eye, but to summarise what we know for sure, the main objection within Macedonia seems to be that there was an element of 'ethno-kitsch' to both the song and the video, which it seems will be eliminated in favour of something more genuine.

And what potential does the new song hold? I have to admit that, while I liked 'Imperija', I was a little disappointed in the overall product, but this new song doesn't sound like it will have the same accessibility as the previous. With strong gypsy influences and opportunities for both singers to show off their vocals, it could potentially wind up having little chance in a European context.

You can hear 'Imperija' below (the video has disappeared from the internet), before the new song is released on Friday.

Does the draw make a difference? Why we'll never know

The EBU's decision to allow SVT to decide the running order of the upcoming Eurovision Song Contest in Malmö has caused the shit to hit the fan and rain down over the Eurovision message boards in a fashion barely seen since the days of Marlain.

But is it going to be a game-killer for the contest? Does a song's starting position make a big difference for its chances? Maybe. In the past few weeks and months, fans have come up with plenty of theories to prove that the draw does make a difference, and that SVT are out to screw the chances of the songs they don't like.

I intend to show you that those theories are generally a whole load of hookum.

"But in 2009, 14 of the 20 qualifiers came from the second half of the semis."

Two points. Firstly, 14/20 is hardly an overwhelming correlation. Secondly, using statistics like that out of context ignores one of the greatest variables in the contest - the quality of the songs. One of the main reasons it's impossible to compare statistics like this on a year-by-year basis is that the songs and performances are, by their very nature, completely different each year.

Without even looking at what songs were drawn where in 2009, that argument contains obvious fallacies, as it assumes that all songs are born equal. They aren't, and by just looking at the bald statistics you're completely failing to take this into account.

Taking 2009 as an example, the first halves of those semis include 'Aven Romale', 'Copycat', 'La teva decisió (Get a Life)', 'Let' tmou' and 'Firefly', songs that most fans would agree never had any hope of qualifying. This just shows that judging the numbers and the statistics without taking into account the songs themselves is pretty much a futile exercise

"But Haba Haba is a great song, and was drawn second and failed to qualify."

Sure, you think it's a great song. Maybe everybody at your party loved it too. The problem is, of course, that there is no hard and fast way to judge a song's quality, and on the night, not enough people liked it and voted for it. And in music, unlike science, there is no right/wrong, no measure of good or bad.

We simply cannot argue that if a song had been drawn in another position, it would have got another result. Ultimately, we cannot know. Perhaps Haba Haba did only fail to qualify because it was drawn second. Or perhaps it did badly because most people didn't think it was very good or very well performed.

Ultimately, it's impossible to directly compare songs - all songs are different and we only ever see them in one starting position - it's not like we ever get the opportunity to place the same song in different positions with all the other influencing criteria intact, just to see what would happen. In summary, it's the songs themselves that are the biggest variable, and have the biggest influence over their own final result.

(Incidentally, as a side note: in researching this article and looking for an example for this section, it struck me just how many songs drawn second have actually qualified since 2008 - either six or seven out of ten, depending how you view Sweden 2008.)

"But we do see them in two different positions - in the semi and the final!"

This is a fair point, and there are examples of songs which qualify and then get a very different draw in the final, and get a very different result. Two examples would be Latvia in 2005, which scrapes through the semi then performs last in the final where it finishes 5th, and Finland in 2011, which qualifies comfortably, then has to go on first in the final where it bombs.

But are they a direct comparison? I would argue that the semi-final and the final are completely different beasts, with greatly differing audiences. Firstly and most obviously, since 2008 the final has all competing countries voting, while the semis only have a selected half. This will naturally affect voting patterns, especially if you got stuck with (or separated from) your neighbours in the semi-final.

The final itself also tends to have vastly higher viewing figures than the two semis. We therefore have a different demographic voting in the final - the semi viewers are perhaps more dedicated fans of the contest or members of a diaspora, whereas the Saturday night show attracts a wider range of people. And on a Saturday, viewers are perhaps looking for something different. The whole family may be watching together, while others may be at a party with significant amounts of alcohol consumed. All of this leads to different voting patterns in the final than in the semi.

And aside from the voting patterns of people watching at home, the final itself is a totally different contest to the semis. The hopeless songs that made you look competent are gone, and perhaps have been replaced by two acts from the other semi-final, doing the same thing you're doing, only better. With completely different competition, and different people voting, the results between a semi and a final are always likely to vary.

All in all, we can never prove that the draw is the decisive factor in a song's result. There is evidence that points that way, and indeed the Latvia and Finland examples mentioned above present a pretty strong argument, but ultimately, with the strongest variable being the song, and a lack of any opportunity to do any kind of representative testing, we can never prove anything for sure.

So, in summary, most fans making up theories and 'proofs' on the internet are talking complete rubbish.