What do Spotify’s audio features tell us about this year’s Eurovision Song Contest? 🤔

Eurovision, not particularly my favorite music event of the year. However, all the related predictions in advance always seem to strike my interest, as well as the (may I say so?) rather distinct music that passes by now and then.

Spotify does their yearly predictions as well — though they have significant more data it seemed fun to figure out a way to compare the previous Eurovision winners with this year’s contestants.

Spotify Audio Analysis

I’m not going to get too much into this because there’s tons of documentation to be found here, but to summarize; Spotify’s API gives you the ability to extract several audio features of a song. The available features that also have been used for this analysis are:

• Danceability: Danceability describes how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable.
• Acousticness: A measure from 0.0 to 1.0 of whether the track is acoustic.
• Energy: Energy is a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy.
• Instrumentalness: Predicts whether a track contains no vocals. The closer the instrumentalness value is to 1.0, the greater likelihood the track contains no vocal content.
• Liveness: Detects the presence of an audience in the recording. Higher liveness values represent an increased probability that the track was performed live.
• Loudness: The overall loudness of a track in decibels (dB). Loudness values are averaged across the entire track. Values typical range between -60 and 0 db.
• Speechiness: Speechiness detects the presence of spoken words in a track. The more exclusively speech-like the recording (e.g. talk show, audio book, poetry), the closer to 1.0 the attribute value.
• Tempo: The overall estimated tempo of a track in beats per minute (BPM). In musical terminology, tempo is the speed or pace of a given piece and derives directly from the average beat duration.
• Valence: A measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry).

What makes a winner?

It would be interesting to see how these variables would demonstrate if we were to look at previous Eurovision winners. I did some investigating and extracted all the audio features of Eurovision tracks that have won over time, of this dataset I only kept the last five years. As you can imagine, music evolves throughout the years and therefore it seems more valid to take this as benchmark compared to all 60+ years.

The table below shows you the average of these values when aggregating the winning contestants.

Knowing this, we can look up the audio features for all 2018’s contestants and see on what audio features they differ from the Eurovision winners average.

After collecting all the individual values for these 43 songs we can calculate the index based on the Eurovision winners average. Tracks that deviate the least (with an average index) from 1 resemble the previous Eurovison winners better.

The tracks that rank best based on deviation are:

1. Alexander Rybak — That’s How You Write A Song (Norway)
2. César Sampson — Nobody But You (Austria)
3. Netta — Toy (Israel)
4. Vanja Radovanović — Inje (Montenegro)
5. Jessica Mauboy — We Got Love (Australia)

How does this differ compared to the official predictions?

I would be surprised when these variables are taken into account when composing the official Eurovision odds. That’s why I was curious if there are some differences in ranking. Turns out, there were quite some actually.

As mentioned earlier, I’ve compiled a top 5 ranking of Norway, Austria, Israel, Montenegro and Australia. Three of these tracks are included into the top 10 ranking of the official odds. Where I’ve listed Vanja Radovanović — Inje as #4 least deviant track, odd-makers feel less confident about this artist, labeling a #41 position to Montenegro. I also seem to be (or are they? ;-)) quite mistaken about San Marino (-37) and Iceland (-33).

Of course, these odds change continously and this is just a snapshot 11 days before the event starts.

The below graph shows you how much this approach deviates from the main Eurovision odds.

Eurovision’s mood

By looking at the valence and tempo variables we can determine what either the most uplifting or saddest songs are.

The saddest songs are coming from Lithuania, Portugal and Latvia whereas the most uplifting songs are represented by Norway, Moldova and Israel.

Russia has a valence (mood) score which is least deviant to the benchmark. When looking at the tempo, the least divergent are Poland, Georgia and Malta.

What happens after Eurovision?

Winning Eurovision is great, but after standing in the spotlight for millions of people you want your music to live on after the event and potentionally fetch an international hit record. Though I feel this has changed over time and there are exceptions to every rule, Eurovision music is more distinct from the mainstream international hits. Something that gave me the idea to compare the audio features of all 43 competing Eurovision tracks with the current top 200 hits on Spotify. Because if you don’t win Eurovision but have a good match with the today’s popular music you might win after all, right?

If we look at the audio features of today’s top 200 hits and compare these with the contestants we get the following top 5 ranking:

1. Mikolas Josef — Lie To Me (Czech Republic)
2. DoReDos — My Lucky Day (Moldova)
3. Madame Monsieur — Mercy (France)
4. Vanja Radovanović — Inje (Montenegro)
5. Elina Nechayeva — La Forza (Estonia)

When comparing these outcomes with the official Eurovision odds you can see that I’m still quite off but less compared to the equation of taking the Eurovision’s winners average.

Have a look for yourself

All the findings are retrievable via an interactive Tableau dashboard, which gives you the ability to search for Eurovision tracks.

Eurovision Dashboard

Disclaimer

1. This is by no means a serious odds calculation but a fun approach to look at Eurovision from a different angle. The only thing taking into consideration are the acoustic variables of a song.
2. The official odds were extracted 11 days before the event — these rankings may change the following days.

You can listen to Spotify’s 2018 Eurovision playlist over here and judge for yourself.

Thanks for reading! If you enjoyed reading this article some 👏🏻 are appreciated.