Who has the best song in 2021? That’s the question we are all extremely eager to find an answer to. And like everyone else in the fan community, we here at Good Evening Europe, will of course try to answer that question before EBU has a chance to.
Earlier in the series we looked at the two Semi-Finals, the participant’s voting histories and the relative success of each running order slot. For the First Semi-Final our statistics based prediction got 7 of the 10 finalists right. Not bad for a prediction that does not look at bookmaker odds or song quality. For the Second Semi-Final the statistics did somewhat worse, only getting 6 of the 10 finalists.
For reference, the most wrong (the wrongest?) anyone could have been is 4 out of 10 for the First Semi-Final, and 3 out of 10 for the Second Semi-Final. Our statistics did better than the worst, but 10 out of 10 is the holy grail of Eurovision predictions, obvisouly 🙂
You can read all the entries in the 2021 Statistics series here
Statistics
The two main parts of our statistics series, are running order rankings and individual voting histories of all the eligible voters for an event. It is common knowledge in the community, that the running order has an impact on an song’s final ranking. Previously we saw that the slot an entry is allocated, can have a large impact on the final ranking of the entry. Or it certainly looks that way. There is really no way to know for sure without turning back time, re-allocating every entry for an event and hoping nobody notices, and broadcasting it all again. The only sure thing is that there is a definite trend, and that some running order slots are worse than others.
The other part of our statistics deals with voting histories of participating countries. As we noted previously in the series, some countries tend to favor specific other countries. Neighbors and cultural simialities both seem to have an effect. So much so that EBU has taken steps to try to level the playing field a bit for the semi-finals at least. With only two semi-finals, EBU has limited options to deal with favoritism between countries. There is still a very real and visible trend toward voting for a country regardless of the quality of the song they enter the contest with. The good thing for our purposes is, that it gives os a mountain of historical voting data that we can use for a prediction of a winner on saturday.
Running Order
We start by looking at the running order. We give each slot a rank based on the average ranking the entry achieved each year since 2010. This gives us a good picture of any trends for each running order slot. The shorter the bar is, the better the slot has historically performed in the contest.
The chart is interesting in itself, because it shows a clear bias toward the ends of each half. More interesting is the fact that there isn’t a clear difference between the 5 year average and the 10 year average. There is some, but it is not wildly different. The reason this is interesting is that, in 2016, the local producers of the show were given the responsibility of allocating the slots. The draw want from a straight random draw of allocation, to the one in use today, where contestants simply draw for which half they will perform in. It is also interesting to not that many of the “good” running order slots can be found towards the end of the first half. The draw of death remains slot 2, there is no way around it.
Friends Voting For Friends
The other part of our statistics adventure deals with how countries vote for other countries. It is a well known fact that some countries award more points to certain countries, than the rest of the contestants. Similarly to how we did it for the First and Second Semi-Final, we have taken a look at the voting history of every participant in 2021, and matched it up to all the 26 finalists this year.
As we can see, Russia, Sweden, Italy and Bulgaria are the countries that most consistently score high overall.
Predicting a Top 10
With statistics for running order and historical voting patterns, we can begin to guess which countries are going to do well in the Grand Final 2021. To make the numbers comparable, we assign each a rank based on how “good” the number is, compared to the other numbers in the same series. A little bit like how the juries rank each entry in Grand Final from 1 to 26.
Each bar represents the rank assigned to the country. The lower the bar, the better the rank. With this complete we can now, finally, come to a conclusion about a statistical Top 10.
| # | Country | Combined Rank |
|---|---|---|
| 1 | Sweden | 2,25 |
| 2 | Italy | 3,25 |
| 3 | Portugal | 8,75 |
| 4 | Switzerland | 8,75 |
| 5 | Norway | 9,25 |
| 6 | Russia | 9,5 |
| 7 | The Netherlands | 10,25 |
| 8 | Malta | 12 |
| 9 | Ukraine | 12 |
| 10 | Israel | 12,25 |
With that our Top 10 is complete. Sweden takes the crown in front of Italy and Portugal.
