Swinging by for a bit of Actual Maths

So on my trawls around the Interwebs, there seems to be a general sense that the voting’s a bit different this year “But I don’t really understand how…”. For one afternoon only, me and my good friend Mr E.Xcel are going to be the randomised people of Austria and talk you through it!

First some basics – at heart, what we’re going to get here is still a 50-50 split between what the professional juries think and what the general public think, and at the end of the night someone (probably called Dodo, they usually are) is going to be saying “And finally… Tanganyika 12 points!!!!” and the crowd will go wild. Except it probably won’t be Tanganyika.

But let’s start at the beginning – a tolerable place to start – in Semifinal 1. ORF have gathered together a motley crew of 5 Genuine Musicians in a darkened room somewhere in Vienna, and are currently plying them with coffee and Austrian cake to get them in a good mood. They’ll need to be, because later tonight they’re going to enjoy Natalia Kelly’s dress rehearsal and then each of them will set about putting all the other 15 songs in order of preference. Yes, each of them will get to say “That genuinely is the worst piece of shit in the whole semifinal.”

According to Mr E.Xcel and Mr R.Andomizer, this is how each of our Austrian musicians is going to rate the competition.

Juror 1: 1 NED 2 IRE 3 MON 4 LIT 5 CYP6 UKR 7 BLR 8 SLN 9 EST 10 CRO 11 DEN 12 RUS 13 SER 14 BEL 15 MOL
Juror 2: 1 MOL 2 LIT 3 RUS 4 CRO 5 MON 6 CYP 7 SER 8 IRE 9 EST 10 SLN 11 BLR 12 UKR 13 DEN 14 BEL 15 NED
Juror 3: 1 RUS 2 UKR 3 NED 4 EST 5 MON 6 MOL 7 DEN 8 SER 9 LIT 10 BEL 11 CRO 12 BLR 13 IRE 14 SLN 15 CYP
Juror 4: 1 MON 2 LIT 3 SLN 4 DEN 5 EST 6 UKR 7 BEL 8 CYP 9 MOL 10 CRO 11 NED 12 SER 13 IRE 14 RUS 15 BLR
Juror 5: 1 SLN 2 BLR 3 CRO 4 RUS 5 EST 6 NED 7 UKR 8 DEN 9 MOL 10 IRE 11 BEL 12 LIT 13 CYP 14 MON 15 SER.

Clearly they haven’t compared notes on these entries nor, apparently, actually listened to the same songs. This can happen.

So what next? Oh, now it gets all mathsy-wathsy as ORF’s chief mathser has to work out how the jury overall has ranked the 15 songs. So what he’ll do first of all is to add up the positions that the five jurors gave to each song: Estonia, on that basis, ends up with a total of 9 + 9 + 4 + 5 + 5 = 32. Mr E. Xcel tells me that the totals for the 15 songs work out as follows (and they’re ordered lowest score first – in this calculation, lowest wins):

Montenegro 28, Lithuania 29, Estonia 32, Ukraine 33, Russia 34, Slovenia 36, Netherlands 36, Croatia 38, Moldova 40, Denmark 43, Ireland 46, Cyprus 47, Belarus 47, Serbia 55, Belgium 56.

You’ve possibly spotted that there are a couple of tied scores in there, between Slovenia/Netherlands and Cyprus/Belarus. In both these cases, the jurors had a show of hands to decide which they preferred – three of the five preferred Slovenia and Cyprus, so they get the better ranking.

Mr Mathsy-Orf has one more job to do with the jury scores, then he and the jurors can go to bed tonight basking in the knowledge of a job well done – and that’s to turn the list into a simple ranking to lodge with the EBU to use tomorrow night.

JURY RANKING: 1 MON 2 LIT 3 EST 4 UKR 5 RUS 6 SLN 7 NET 8 CRO 9 MOL 10 DEN 11 IRE 12 CYP 13 BLR 14 SER 15 BEL

Goodnight all, sleep tight… bloody hell it’s morning already. That was quick!

Let’s let those darned Televoters loose on it and see what they think of it all. Probably something entirely different. At least the sums are easier with them, just let ’em all wote wote wote for their favourites and count the piles. And the televote ranking according to Mr E. Xcel is:

TELEVOTE RANKING: 1 EST 2 MON 3 LIT 4 IRE 5 DEN 6 CYP 7 BLR 8 SER 9 NED 10 BEL 11 MOL 12 SLN 13 RUS 14 CRO 15 UKR

My gosh those Austrians do like their Estonian ballads and dubstep spacemen, don’t they? Let’s merge these two rankings one more time. Montenegro scores 1+2=3, and by the power of Maths the full combined ranking is:

Montenegro 3, Estonia 4, Lithuania 5, Ireland 15, Denmark 15, Netherlands 16, Slovenia 18, Russia 18, Cyprus 19, Belarus 19, Ukraine 19, Moldova 20, Serbia 22, Croatia 22, Belgium 25.

Once again, we’ve had a few ties to sort out in that lot. Televoting take precedence, so Ireland beats Denmark, Slovenia beats Russia and so on. We’ve got ourselves a combined ranking, boys – we’ll be allocating scoreboard points soon, and it’s still Tuesday. Just about.

COMBINED RANKING: 1 MON 2 EST 3 LIT 4 IRE 5 DEN 6 NED 7 SLN 8 RUS 9 CYP 10 BLR 11 UKR 12 MOL 13 SER 14 CRO 15 BEL

And so – eventually – here are the results of Austria’s semi-final deliberations:

Belarus 1 point
Cyprus 2 points
Russia 3 points
Slovenia 4 points
Netherlands 5 points
Denmark 6 points
Ireland 7 points
Lithuania 8 points
Estonia 10 points
And finally, Austria’s 12 points goes to… those whacky dubstep spacemen from… MONTENEGRO!

Whizz ’em over to the EBU, count ’em all up, and before you know where you are fake envelopes are being opened and everybody’s happy. Except for six of them.

Then everyone comes back on Friday and Saturday and does exactly the same stuff again, except they’ve got 26 songs they need to deal with (or 25 if they’re one of them).

It’s as clear as mud really, isn’t it? I’d just shrug your shoulders and have another drink if I were you, to be honest!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Anonymous
Anonymous
10 years ago

I’m shrugging and I’m drinking. Here’s an idea, Cyprus have had a beastly year, can’t we just give all the points to them?

JimW
JimW
10 years ago

Actually that’s one of the better explanations that I’ve seen Nick. It will be interesting to see how much of a difference this will make in the overall voting.

It most likely won’t dilute the neighbourly voting patterns because let’s face it a jury (even of music professionals) will probably have the same influences as the general public in their country. So for example a Greek professional jury will probably still be quite generous towards the Cypriot song, a Ukrainian professional jury will still most likely vote high for the Russian song and so on.

What it will probably have an effect on is the diaspora voting patterns because even if Country X gets top marks from it’s disapora in Country Y the jury for Country Y may place the same song very far down their vote. That means that when the jury and televote are combined County X is likely to end up further down the list compared to before when only the Top 10 in each list was considered.

This is also going to be magnified more in the final where there are 26 songs as opposed to the semi-finals where there are only 16/17 songs.

Darren
Darren
10 years ago

Good stuff Nick!

What you didn’t say here, is what the result would have been under last year’s system. In that scenario, rankings 11th downwards were irrelevant to the combined score. Therefore, the new system really hurts those countries that finish at the bottom of either ranking. In your particular example, Ukraine.

The results under last year’s system would have been:

Slovenia 1 point
Cyprus 2 points
Russia 3 points
Netherlands 4 points
Ukraine 5 points
Denmark 6 points
Ireland 7 points
Lithuania 8 points
Estonia 10 points
Montenegro 12 points

So, where this is going to have a telling effect, is when there is a large difference of opinion between judges and televoters. The obvious scenario is a song, disliked by most, that has a large expatriate vote. Armenia, anyone?

nick_at_esc
10 years ago

Great shout, Darren. Not sure that Armenia’s going to be the one that gets burnt by that effect though, as it’s got some possibilities for jury love – perhaps Serbia might be at risk?

We might find it working the other way though if a country that’s very effective at gaining jury support does absolutely zip all in the televote… and I think we might have a prime candidate for that in semi 2, in the shape of plucky little San Marino.

Darren
Darren
10 years ago

It’s all very interesting, that’s for sure.

Diaspora vote is the obvious disconnect between the televote and jury, assuming that the majority of the jury members aren’t of the diaspora and voting with national prejudice in mind.

Neighbourly voting, e.g. Cyprus/Greece, and the states of Yugoslavia, may be less different between jury and televote, because there is a common language (usually), a common culture, and familiar names that may mean even the juries’ votes are skewed to their neighbours.

Another thought I had is that each juror fully ranking all the songs may reduce or even eliminate the recency bias that we are used to; i.e. the benefits of being drawn nearer the end of the competition. By ranking all the songs, you are not picking out the 10 you like best, but actually considering what you liked the least as well as what you liked the most. I don’t see that as penalising the early acts. However, the televote will still definitely be skewed in that direction, and analysing televote vs jury vote should be enough to prove to the EBU that the running order does have an effect on televotes.

The other effect of this, is that the “Marmite” songs will do less well. If a song gains 10% of the televotes from 25 songs, it will win, even if the other 90% of the population actively disliked it. This could mean that very polarised songs, like “Hard Rock Hallelujah” may not have had such an easy ride under this system. I can imagine that Romania’s entry this year is going to do far better in the televote than the jury vote as a consequence.