#starting position

[pratchettquotes]

Of the three things that most people know about the horse, the third is that, over a short distance, it can't run as fast as a man. As Rincewind had learned to his advantage, it has more legs to sort out.

There are additional advantages if a) the people on horseback aren't expecting you to run and b) you happen to be, very conveniently, in an athletic starting position.

Rincewind rose like a boomerang curry from a sensitive stomach.

Terry Pratchett, Interesting Times

[elettradancer]

[eurosong]

Hi there, folks! We finally have the running order for this year's Eurovision - I've been waiting so long to get my hands on one of these! I of course had to take a look at starting positions to try to gain some insight into how countries may have done in the semi-finals, using both what the juxtapositions made suggest to me, and how previously songs did in those positions.

The evening will kick off with Cyprus and close with San Marino. The opening and curtain-closing positions have a lot of commonalities in this producer-created running order era - they tend to pick up-tempo songs that did relatively well in the semis (or, amongst big 6 in the closing position, are relatively well liked) but not considered contenders for the win, and indeed, both countries' odds of winning dropped since the running order was released.

The famed 'death slot' of second has constantly been used for 9th place qualifiers instead of 10th, oddly enough - with the sole exception being Spain's doomed Tu canción in 2018. It seems like Albania, who were also placed here in 2019, may have repeated history. Following them, Belgium and particularly Israel strike me as having probably been marginal qualifiers.

We then get the first of many unusual sandwiches of fan favourites - two uptempo, lyrically feminist songs from Russia and Malta. Despite the latter being a bigger bookies' fave, they may have underwhelmed, particularly in the televote, to get put after the powerhouse of Manizha and in a position with a worse track record. Portugal are next, contrasting well with the two songs proceding it and in a slot that's had a few big successes. They almost certainly did well with the juries are probably outperformed expectations with the televote.

Serbia follow in a slot that has contained nothing but top 5 finishers in their semifinal, and I firmly believe Hurricane did the same in SF2 this year. UK is up next in an historically decent slot, albeit one in which they themselves have underperformed before. The banger vibes may be undermined by having been proceded by something with even more energy. Greece gets one of the worst slots in recent years and the worst of the latter first half. Coming after two uptempos and before a bookies' fav like Switzerland makes me think they seriously underperformed.

Speaking of Switzerland, they're in a slot that has hosted successes like J'ai cherché, YLMWA and Amar pelos dois - and the juxtaposition with what surrounds them makes me think they won their semi or came close. Iceland follows in a decent slot; they may contrast positively with Switzerland and definitely will with Spain's ballad. Moldova has a pretty strong position before Germany, who are slotted into the worst position statistically of the entire second half. Finland come after them, a jarring juxtaposition, and in a starting position that has seen mid-table semi-finalist qualifiers but also two no-hope Big 5ers. My guess would be they managed 6-7th in the semi.

Bulgaria will contrast well with them, and whilst their starting position is statistically not one of the best, the cordon sanitaire it enjoys of being the only down-tempo song with two up-tempo numbers side by side make me think Victoria managed top 3 in her semi-final. Lithuania are in an a priori great position, but between the emotional moment from Bulgaria and the chaotic, infectious energy of Ukraine, I fear they may have underperformed in their semi.

France enjoys a similar cordon sanitaire to Bulgaria, reflecting their favourite status. Azerbaijan has the job of following Barbara and makes me think Efendi could have only scraped into the final. Norway is next in one of the historically best spots of the final and may have outdone expectations. The hosts, Netherlands, have been given a rough deal by having to precede big fave Italy. Sweden, whilst in one of the most advantageous spots historically, may well be very squeezed between Måneskin and Senhit w. Flo Rida.

[eurosong]

Good evening, folks, and welcome to today's statistical map and breakdown. I have always been a fierce critic of producer-picked starting order, because there is no doubt some positions in the show are more favourable than others. Indeed, just a few hours have passed since the order was revealed and it's already had an impact on the odds, which we'll be reexamining in the next few days. Before that, let's take you through the final to see which countries' qualification chances are looking better or worse after the starting order was released, taking into account historical qualification rates in that position over the past five years.

We start with the two countries who will be raising the curtains - Cyprus and Armenia. The honour of starting the show is usually considered a plum position, but openers have only qualified 60% of the time, and usually only one of the two does - the last time both openers went through was 2014. With Cyprus'  momentum and Armenia's generally very strong qualification rate, I have the feeling 2019 might be an exception like '14 was.

2nd place is dreaded in the grand final but is not a great position in the semis either, with only 30% qualifying from it. Montenegro's chances were doubted strongly from the time Heaven was picked at Montevizija, but the 2nd place could be even more damning for a marginal Irish song that could potentially have passed from a more favourable spot. Third is also not a great spot - last year, we had two qualifications there from Albania and Serbia, but neither stormed through, and before that, only 3 songs out of 10 from 2013-17 had qualified from 3rd. Not looking good, as a result, for Finland and Moldova. Neither song went through last year from 4th place, and it's generally a difficult position from which to qualify with only 40% rates. I predict Switzerland will get past this easily, but Poland might struggle.

5th and 6th, on the other hand, are a small purple patch with 70% - good news for Slovenia, Czechia, Latvia and Romania. I could see 3 of the 4 at least getting through. 7th and 8th are less fortuitous but a few big favourites have come in 7th in recent years like Netta - I think only Hungary will make it in SF1 and will not be coming in the top spots, whilst I wouldn't be surprised to see SF2's Dano-Swedish sandwich's constistuent pieces both go through. 9th is one of the best positions of all, with 80% of all songs starting there going through, and whilst neither Serbia or Austria have a big following this year, perhaps the dissonance with the upbeat songs preceding them will allow them to create a memorable moment. 10th has a decent qualification track record of 70%, but in the last two years, the song in SF1 in that position has qualified, whilst its counterpart in SF2 has not. I would not be surprised to see the same happen, with Belgium going through and Croatia falling. 11th and 13th have generally bad track records with 40% qualification, sandwiching a stronger 12th. Last year was a departure from the norm, with the song at 13th being much stronger and passing where the other two didn't - easily in the case of Cesar, barely in the case of Hungary. This year, that new trend seems set to continue - in SF1, Georgia is very doubtful for qualification and Australia is hit and miss, whilst Iceland is extremely hyped; in SF2, equally hyped Russia gets the thirteenth spot after an underwhelming Lithuania and a strong Malta whose live performance may nonetheless disappoint.

14th tends to be a non-qualifier dip before either a hyped or high profile entry (like Monsters or Dance you off last year or Beautiful mess in 2017). The formula tends to be repeating itself with Portugal, one of the most-talked about entries, and Norway, one with a fair degree of fan hype, coming after Estonia and Albania, who haven't got as many tongues wagging. Both countries have been really done dirty, especially Albania, a power ballad directly following a power ballad and preceding two hot favourites. They're following that up with two very strong songs in their own right from Greece and the Netherlands - 16th is not usually a great place but I feel its average qualification rate will jump after this year, even though, in Greece's case, it has the weight of being the penultimate song - last year, both penultimate songs qualified, but in 2016 and 2017, none did. I fear the position may seal Macedonia's fate.

Amongst the biggest winners in terms of qualification chances, we always have to include the song that comes last. Only one show-closing song in 10 in the last five years has not qualified - Latvia's Line back in '17 - and being the last song before the lines open certainly seems to make an impression on voters. We will almost certainly be seeing the return of Azerbaijan to Saturday after last year's NQ, and Björkman has given an almighty boost to San Marino's qualification chances by letting Serhat perform last, who gets a “very favourable” rating despite performing 17th because he’s the curtain closer. It’s a poisoned challice, though, as going by previous years' convention, one of the two, if they qualify, will be placed first in the final and underwhelm in their Saturday results. A full data table is available below.

[eurosong]

Good afternoon, folks, and welcome to one of a bumper set of analyses we’ll be sharing over the next couple of days. This one deals with the producer-led running order, which, since its institution in 2013, has often unfortunately given us a taste of how well songs did in the semi-finals. A song receiving the same starting position as songs that did well in the semis of previous years often indicates that it also did well in the qualifiers, whilst the opposite is true when a song is placed in a death spot. When it comes to the position of the automatic qualifiers, it often gives us an indication of how well Björkman thinks it will do.

We have taken data from ESCs since 2013, but now have enough data to exclude 2015 - with 27 songs, it was a major statistical outlier. We note what rank they got in their semi-final and find an average SF result. As we can see, there are very few “neutral” positions - and there is a world of difference between being put in 11th place, whence two songs have won, and 15th, where most songs placed there only scraped by. 

There are four extremely advantageous starting positions in this era of ESC. The first is 11th, where both Conchita and Salvador performed, as well as a warmly received performance from Amir of France in 2016, which was seen to have gotten the Gauls out of their doldrums. The only swing and a miss in this position was Cascada - who, nonetheless, were expected to do well. This is a spot for semi-final winners or big 6 predicted to do well; and it seems that Björkman believes that Michael Schulte can give Germany a reasonable hope of a good finish this year. 

18th place is also very strong. It’s been home to two winners, EDF in 2013 and Sergej in 2016, as well as Softengine in 2014, who were rewarded for their country’s best semi-final placing (3rd) in years with a top spot. Even the weakest song that performed in this spot, Never give up on you last year, got one of the UK’s best positions in several years. So 18th is also a place for semi-final winners or a reward for positive momentum at the contest. Equinox, who take the spot this year, probably didn’t win the semi but I feel definitely came top 3.

Traditionally, in the home stretch of the show, 21st and 24th have been the best spots to receive. In 21st place, we’ve had a string of podium finishes - Alcohol is free (2nd), Running (3rd), 1944 (2nd) - with the only dud being Perfect life, who can probably thank that providential position to a surplus of unfancied big 6 songs being put into the second half last year. In 24th, we’ve also had mostly big semi successes: I feed you my love (3rd), Calm after the storm (1st), Loin d'ici (7th), I can't go on (3rd). Hungary and Ireland, receiving these prime slots, may have outperformed expectations and grabbed a podium place in their semis, though there is the risk that Ireland scraped by - there are so many up-tempo songs in the second half and Björkman could well be using Ryan as a sacrificial lamb, one of the few slow songs in that half to create a bigger impact for Cyprus, put in 25th place. This spot used to be one of the worst in the second half, but it’s becoming a new pimp slot, given how Kristian Kostov was also awarded it last year.

What about the show openers and closers? Over the last 5 years, the opener usually outperformed the closer, but neither were deemed contenders for the win. Mélovin is following in the footsteps of Imri, Laura Tesoro and Mariya, artists that did relatively well in their semi but slumped in comparison in their finals. Metamoro are given the task of ending the show, often one that the Big 5 are afforded, and must be considered as top half contenders but, in Björkman’s eyes, not winners.

There are some positions that can definitely be considered death slots. The most infamous in ESC history is 2nd, where qualifiers who were only 9th in the semifinals tend to be put, but 4th is also a seriously bad position where marginal qualifiers tend to be put - like Nathan Trent, Samra, Krista Siegfrieds and Pollapönk. None of those names made it into the top half in the finals. Getting put in 4th made Lithuania’s odds with the bookies, which had been soaring since their sweet SF appearance, go down like a lead balloon. Meanwhile, Björkman must not be expecting strong scoring from Spain or otherwise might want to avoid a back-to-back Iberian set of wins, controversially putting an automatic qualifier in the death spot of 2nd.

There are also death spots in the second half - if you came last in your semi and you drew second half, you may often find yourself in 15th or 17th position. In 15th, we’ve had two last qualifiers - Sanja Vučić and Demy - as well as a Big 5 (Bonnie Tyler, who came 19th) predicted to not do so well and the Tołmaczowy sisters who got only 6th in their semi, a weak position for Russia. So, it looks quite probably that Denmark only barely scraped through this year. In 17th, we had two tenth place qualifies too, Tinkara and Nina Kraljić, but there is a glimmer of hope for Finland, starting in this position tomorrow - ByeAlex (8th in semi) and Jowst (5th) both overcame the starting placement to grab a top 10 finish.

[eurosong]

Good afternoon, folks, and welcome to today’s statistical map, which is all about the running order for this year’s semi finals, which has just been released in the last hour. Those who’ve been following the blog for some time will know that I’ve been a big critic of producer-chosen running order - you can greatly help a song’s chances by artificially fabricating a positive contrast for it, and you can also do the opposite, so it’s incredibly unfair for a country’s chances to be helped or hindered by someone unaccountable. 

The position in which songs perform has also proven to have quite an impact on how well songs have fared - when 90% of songs in starting position 12 qualify for the final, but only 30% in position 3 do, there is clearly a statistically clear dis/advantage at play relating to where in the evening you perform that should be determined at random, not by an individual with vested interests. If you’ve got a subtle, low-tempo ballad that ends up getting sandwiched between two other similar acts, if given that position by lot, you’re unlucky; if given it by design, it feels almost an act of sabotage. 

So, who are the winners and losers of the starting position decision? As has been the case for the past few years, Denmark, Sweden and Finland have each received pretty good slots, with the latter two performing in positions with an 80% qualification rate. Russia gets a warm welcome back with the best slot, statistically, in the first half, a position shared with Lithuania - although, in the latter’s case, the highly positive starting position may be cancelled out by the fact that it’s appearing as a cushion between Czechia and Israel, two of the most hyped entries of the year. Ukraine get the generally favourable show-closing position, and whilst Ireland share that typically favourable 18th position, the benefit of closing the show may be felt instead by Cyprus in 19th. 

There are six countries in truly make-or-break positions with 50% qualification rates: opening their respective semis are Norway and Azerbaijan, a position that can be a real mixed blessing. In the past 3 years, only one show-starter of the two has managed to qualify, and I have the feeling that this may happen again. Over the past 5 years, only one of the two songs in #7 has qualified - and 80% of the times, it has been the one in semi-final 1, which bodes well for Israel and not so well for Moldova. Similarly, I expect the country to qualify out of the couple on #4 will probably be Belgium over San Marino...

On the other side of the coin are the countries put in positions with less than a 50% qualification rate. For the second time in a row, Austria has been given a bad starting position, and this time around, most of the disadvantageous starting positions have been given to countries in a “very curious” almost geographically contiguous area in central Europe and the northern Balkans, with Switzerland, Hungary, Slovenia, Serbia, Romania and Albania all having disadvantageous starting positions. I will never understand how a producer determining who gets the best and worst slots will ever be fair play.

A full table of the qualification rates of the positions from 2013-2017 can be seen below.

[eurosong]

Thoughts on the starting order

Good (very early) morning, folks! It's 5 am and I'm still up because my adrenaline levels haven't dropped since the start of the semi-final. The end of Eurovision season is in sight and I'm as excited as I am apprehensive.

Those of you who've followed this blog for a while will know that I'm an ardent opponent of the producer-created running order - here's an archive of posts just from ESC '16 and '17 where I explain why it is inherently unfair (the long and short of it: it gives advantageous and disadvantageous positions to songs, which goes against the spirit of the competition) and how the producers choosing the running order often fall on the same tropes - using semi-final results to determine where to place the qualified songs.

As outlined in this post, there are positions in the running order that the producers consider much stronger than others, into which they tend to stuff the high-flying semi-finalists and Big 6s that they think will do well - like the 16th act before the end of the show, which is usually 11th place (but was 12th in 2015 on account of there being 27 finalists).

This prime position, almost as close to the end of 1st half as you can be, has only been occupied by a semi-finalist once in the previous four ESCs since random draw was abolished: 11th has been home to Cascada, who were tipped for a good position pre-ESC; Conchita, who won her semi and the whole shebang; Guy Sebastian, making Australia's much hyped début, and Amir, a huge pre-contest favourite. This is where Salvador Sobral has been placed - which makes me feel that he must have done as well as it seemed he did at winning over the fans.

The other strongest positions in the first half are 9th and 13th (10th and 14th in 2015). 18thj from last is where Gianluca (4th in his semi, 8th overall), Maans (winner of both semi and final) and local boy and 5th place in the final Frans were placed - this is where Francesco Gabbani has been slotted. A strong position, but not as good as Salvador's. 14 songs before the end (usually 13th place, 14th place in 2015), the last stop before the second half, is where Dami Im, Sanna NIelsen and Anouk have all performed - a semi final winner, a runner up and 6th place in the semi and 9th in the final respectively. This prime spot, surprisingly, is where Jacques Houdek has been placed. Will he emulate this high-flying female trio or the only blokes to have been assigned this position, the Makemakes with their unfortunate nul points?

Whilst the producers can help you, they can also hinder you. 2nd from the beginning is the dreaded death spot from which no one has ever won - but in recent years, 2nd from last is more of a death sentence. A slot for 3 last place marginal qualifiers from the semis (Georgia '13, San Marino '14, Albania '15) and an unfancied Big 6 (UK last year) seems a very odd place indeed in which to put such a bookies' favourite as Bulgaria. Its juxtaposition is unfavourable too, sandwiched between other favourites in Sweden and France - did it really underwhelm so much to warrant this, or are they actively trying to kill the kid's chances? After the "eventfulness" of this year, could they be afraid of a début Eastern host country taking the reins?

In other news, JOWST seems a surefire marginal qualifier, slotting into 10th position from end (17th, 18th in 2015) which has been home to two countries who qualified in 8th place (Poland 2015 and Hungary 2013) and two in 10th place (Croatia last year and Slovenia in 2014). Out of the quartet, only ByeAlex managed to surpass the mediocre semi final placement with a good finish in the final. It also seems that the city lights are dimming for Blanche, placed fourth from last (usually 23rd, 24th in 2015). It's not been a bad slot for Big 6 allotted it - Basim and Marco Megnoni - but when a semi qualifier has been given fourth from close, they placed 10th and 9th in their semi respectively. Both were Georgians, curiously - Nina and Nika & the Young Georgian Lolitaz.

On the "winners" side, we should also give a mention to Sweden (who perform 3rd from last, like Margaret Berger, the Common Linnets, and Polina) and the UK, who get given 9th from end (usually 18th, 19th in 2015) starting position of a third place semifinalist (Softengine), a semi final runner up (Aminata) and two semi-final winners (Emmelie deForest and Sergey Lazarev). On the "losers" side, their starting positions suggest that Austria, Greece, Poland and Australia scraped through.

[eurosong]

Good morning folks! As many of you will know, I have been a passionate opponent of producer-led starting position since it was introduced, and have been arguing for the return of the neutral and much fairer random draw - when there are places in the semi-finals in which songs almost always do well, and others from which songs never or very seldom qualify, these statistical advantages and disadvantages are great and should not be determined by producers, who are often far from neutral.

I've talked about that a lot, but there is another factor that makes producer-led draw both unfair and laughable, and that is that, by using the positions to which songs have been allocated in the final, one can often predict how well a song did in the semis. I'm not just talking about juxtaposition here - though that is often used to killer effect to: if the producers put you between two hyped songs, these days that means that either you did badly in the semis or you’re expected to do badly in the finals, like Jamie-Lee last year between Amir and Frans. I'm also talking about starting positions in the final that are almost always reserved for songs that did well in the semis, and other spots for those that did badly. Eurovision's producers can - and do - "reward" songs that battled to scrape through to the finals in 9th or 10th place by giving them a death slot in the final to kill off their chances of winning or even doing well.

I took data from the four Eurovision finals since the ill-fated producer-led draw was introduced. I noticed spots that were strikingly reserved for big songs, and others for narrow qualifiers. The one exception was often provided by 2015, because of its 27 songs - by counting backwards, though, (e.g. grouping songs that performed second last together - 25th in normal years, 26th in 2015) one sees some startling trends.

The most famous inauspicious spot of all is thought to be 2nd place, the cursed position from which no song has ever won Eurovision. Isn't it a dick move to give that death spot to a country qualifying for the finals for the first time? That's what SVT did for Czechia last year - the underlying philosophy being that if a song scraped through, it doesn't deserve a chance to shine and do better at the final. The only song performing "twenty-fifth from last" that did well in a semi-final was Golden Boy - which was put in 3rd starting position in 2015 behind the ill-fated Lisa Angell. The other songs in that category that started second - Cheesecake and Something - did mediocre at best, scraped through at worse - and both these and I Stand underperformed considerably compared to in their SFs.

Whilst 2nd starting position is famed for its unluckiness, what is becoming much more of a deathslot thanks to the producers is second last to perform - 25th usually, 26th in 2015. If your favourite song is put on just before the song that closes the show, kiss goodbye to their chances of winning or even doing well. In the past 4 years, the song that’s been put in second-last position either was a Big 6 (Joe & Jake) or the very lowest qualifier in their semi: Georgia in 2013, San Marino in 2014, Albania in 2015. 

Not only have they been given an inauspicious spot, but they have also had to perform after a major semi-final favourite, because third last (24th usually, 25th in ‘15) has become a confirmed “pimp slot,” where big songs have been placed: like those of Margaret Berger, third place in her semi and fourth overall, as well as semi winners Polina and the Common Linnets.

Another death spot is tenth last (17th usually - 18th in 2015.) In that spot, we’ve had two eighth places (In the Name of Love and Kedvesem) and two last places (Round and Round and Lighthouse). Only ByeAlex did well: the others got two 23rd places and a 25th place. This is in part, again, because they’re performing before a designated pimpslot: in ninth last position, we’ve had Emmelie deForest (winner of her semi), Softengine (third), Aminata (second in her semi) and Sergey (winner of his semi). If your fave is on 9th last, it’s certain they did well in the SFs.

If you’re in the first half, sixteenth from last (usually 11th) is the best place, being reserved for direct qualifiers expected to do well (Amir, Guy Sebastian, Cascada before their flop) or for winners (Conchita), whilst 21st from last (4th; 5th in 2015) is the worst, usually reserved for unfancied Big 6 (Electro Velvet) or marginal qualifiers like Pollapönk and Krista Siegfrieds.

Frankly, it is shocking and unacceptable that from starting position alone, we can infer so much information about both how a song did in the semi-finals, and how well or badly it is believed by the producers to be able to do in the finals. It borders on farce. A song’s performance in the semi shouldn’t seal its fate in the finals. Performers should not be able to kill off a song’s chances by putting it in between two favourites. It’s time to bring back the random draw already.