Eurovision votes, country by country: where many emigrants live, the public favours their homeland and the jury does not

Where many emigrants live, the public pulls away from the jury

The same comparison for every pair of countries with no shared border and no close language, in the 2016–2026 finals: how much more the public gave than the jury, after taking away the gap seen in the rest of Europe. Pairs are grouped by how many people born in the performing country live in the voting one.

Below a thousand emigrants, public and jury give almost the same (−0.43 points apart, 503 pairs). Above 100,000, the public gives on average 1.5 points more than the jury (54 pairs).

Of the public 12-pointers in finals with juries, 34% went to a neighbour or a country with a close language, and 9% to a country with no shared border and no close language with over 100,000 people born there living in the voting country. For juries, the same two groups received 21% and 4% of the 12-pointers.

Publics favour eastern countries with many citizens living among them; western ones, hardly at all

Where more than 100,000 citizens of a Central or Eastern European country live (Turkey and the Caucasus are counted here too), the host public gave that country in finals an average of +3.0 points more than the song got from the rest of Europe. For western countries with over 100,000 citizens in the host country (Germany, France, Italy, Portugal, Spain, the United Kingdom and the others) the public gave +0.22. The same countries’ juries: −0.23 and +0.12. The calculation leaves out neighbours and countries with a close language and uses the 30 host countries for which Eurostat publishes citizenship.

The chart shows, for each country with enough votes, the average points it received from the public and from the jury of the countries where more than 100,000 of its citizens live, in finals and semi-finals with juries, leaving out neighbours and countries with a close language. Poland gets 7.4 from the public and 1.1 from the jury, Romania 6.1 and 1.6, Germany 0.5 and 0.9.

The gap appears separately in each half of the period: +3.2 against +0.29 in 2016–2021, +2.8 against +0.15 in 2022–2026. Without Ukraine, whose 2022 win drew votes from everywhere, the gap stays (+3.3). The east–west difference in public-minus-jury points has a 95% confidence interval of 1.8 to 4.3 points, computed over 47 country pairs.

Germany is the case that does not fit: 771,000 Romanian citizens live there, and the German public gave Romania on average no more than the German jury did. Some of the 969,000 people born in Romania who live in Germany are ethnic Germans (Saxons and Swabians) who left mostly between the 1970s and the early 1990s.

Both ways: the extra points go in one direction

Being neighbours and sharing a language work both ways; emigration usually does not. Across 78 country pairs, at least 20,000 people born in one country live in the other, and at least ten times fewer the other way round. The host public gives the home country, above the song’s level, on average 1.4 points more than it gets back (68% of pairs). The same countries’ juries: −0.19 (51%).

An example: Lithuania's public gave the United Kingdom 0 points in nine of ten finals; the UK public gave Lithuania 12 points in eight of fourteen shows. About 207,000 people born in Lithuania live in the UK.

Moldova and Romania: 12 points that emigrant numbers cannot explain

Moldova's public gave Romania 12 points in four of five shows since 2017, and 10 in the 2022 final. The number of emigrants cannot explain these points: only 3,545 people born in Romania live in Moldova. Romanian is Moldova's official language, the two countries share much of their history, and Moldova's jury also gave Romania 12 points three times.

The other way round there is a diaspora: 81,000 people born in Moldova live in Romania. Eurostat counts 9,740 citizens of Moldova there; people who also hold Romanian citizenship are counted as Romanian. Romania's public gave Moldova 12 points in all 8 shows in which it could.

What these data do not show

• Emigrant numbers do not explain most votes. The model that knows them predicts the public vote only slightly better than the one without them; being neighbours and sharing a close language improve it more.
• An emigrant count is a count of people, not of votes. Nobody knows how many of them watch or vote, or what share of the public vote comes from them. Where the UN publishes no number for a pair of countries, the card says so and the pair is left out of the calculations by number.
• Juries are not neutral: they also often give 12 points to neighbours (21% of their 12-pointers, against 34% for the public).
• Nothing here suggests fraud: voting by phone, text or app is open to anyone in the voting country, and people also vote for the song of the country they came from.
• Russia after 2021 and Belarus after 2019 are missing; they no longer take part. In 2022 the EBU replaced the juries of Azerbaijan, Georgia, Montenegro, Poland, Romania and San Marino with a calculation from other juries; those votes are left out. The 2023, 2024 and 2025 semi-finals had no juries. The online vote from the rest of the world is left out.
• Citizenship data cover 30 countries of the European Union, the European Economic Area, Switzerland and Turkey; the United Kingdom only up to 2019.

Method and sources

Points, for each show, country and jury, come from the detailed voting tables of the Wikipedia article on each contest (CC BY-SA), which cite the results the EBU publishes on eurovision.tv. That is 16,132 votes from 30 shows (finals and semi-finals, 2016–2026; the 2020 contest was cancelled), between 1,935 pairs of countries. The totals checked match the official ones once the online vote from the rest of the world is added back: for example, Romania’s 232 public and 64 jury points in the 2026 final.

Emigrants: the number of people born in one country living in another, from the UN International Migrant Stock 2024 (CC BY 3.0 IGO), and the number of foreign citizens on 1 January, from Eurostat’s population statistics by citizenship (2014–2025). Counting by birthplace also includes children born abroad to people who later returned home, and people such as ethnic Germans born in Romania; that is why the east–west comparison uses citizenship.

“Above the song’s level” means: the points a country gave, minus the average the same song got in the same final or semi-final from every other country, separately for public and juries. “East” means the former communist states of Europe, plus Turkey and the Caucasus. The tests were written down before counting; a model trained on the 1998–2008 finals predicts the public vote in the 2016–2026 finals slightly better when it knows how many emigrants live in each country (the mean squared error falls from 6.33 to 6.19, in squared points). Emigrant numbers alone help more than distance, but less than being neighbours and sharing a language.

• Wikipedia, “Eurovision Song Contest 2016” … “2026”, detailed voting tables (CC BY-SA 4.0)
• EBU, eurovision.tv, detailed results
• UN DESA, International Migrant Stock 2024 (CC BY 3.0 IGO)
• Eurostat, population on 1 January by citizenship
• Natural Earth, borders (public domain)

The points on this page come from Wikipedia and may be reused on the same terms (CC BY-SA 4.0). CC BY-SA 4.0