I finally see Eurovision through the eyes of my accountant.

Admittedly the first eleven minutes were a slog. But if you persist, you’ll be rewarded with diagrams.

There is one major problem with the spreadsheet, it doesn't really take into account the previous year's winner autoqualifying for the finals, they are getting treated as if they got to the finals through semi finals and placed whatever they did instead of rating them the same way as the big 5.

metrics and numbers were not confusing at all, rubic cube kid was

I think the reason why the UK and Ireland were doing so great in the past is because of the language rules. The competition was also less diverse musically back then so the prettiest ballad usually won. (Not to say there aren't exceptions to this rule - my favourite winner of all time is from 1995 - but that was often the case)

it's me, im the person who loves eurovision and venn-diagrams, and statistics and rubiks cubes. Also "sorry it's hard to do, cause like it's hard to do, so like whatever" are words I intend to live by

My brain kinda hurts, but I gotta appreciate the work put on this one. Thank you!

The language rule seems like it really helped France, Britian and Ireland specifically. I say this based on the abyss they seemed to find themselves in from 1999 onward when EBU skipped that rule.

Finally, the nerd shit eurofans have been waiting for :D

You're certainly going beyond what I ever imagined doing with eurovision statistics. It's awesome.

I claim the right to complain! Thats an important Accomplishment for us germans, especially against the british, our dear brothers and sisters in the art of complaining...

This is an epic showdown between Romantic intuition, as represented by youth, and Enlightenment rationalism, as associated with experience. The director uses ironic juxtapositions: the sterile plasticity of a Rubik's Cube with the solver's fluid approach, the light-heartedness of pop music with the relentlessness of numeric classification. The climactic cry of "Dad, why'd you put me in the most boring video?" symbolizes the eternal rebellion of one generation against another. It is positively Turgenevian, or would be if I had ever read any Turgenev. In conclusion, douze points to Alex.

Something I've learned from this is that Norway proved you can change for the better, and tbh, I'm so proud of my country for that😆

I believe that an even clearer indication of vote tampering during Azerbaijans hot streak shows if you look at how many points certain individual countries gave them during that period. In several instances you could see seemingly random countries without any cultural ties to Azerbaijan consistently giving them high scores. If I remember correctly, Malta was the most blatant example, having a several year long streak giving them top marks. The year after the voting allegations emerged, those voting patterns disappeared almost instantly.

Also! I just noticed that Georgia is missing from the spreadsheet. Georgia actually has the 4th lowest 5-year average with an average scaled score of 0.104 from 2018-2023. Georgia also had their worst 10-year average in 2023. Hope this helps!!

Great maths! Well done

You guys have regained my support after the clarifications regarding Azerbaijan.

So much work into your videos and you bring them so often!!! Brilliant, big admirer!!!

Hey! I'm really in the center of the Venn Diagram of nerds into eurovision, so loved this! Just an idea, maybe naive. It seems a bit abrupt to consider a 10 year rolling average. It makes sense that a result from 10 years ago stops being relevant, but 2012 disappearing all of a sudden in 2023 is still abrupt. How about a "weighted" rolling average? For example: now, 2023 counts 40%, 2022 counts 25%, 2021 counts 20%..., 2013 counts 5%. It could make things smoother I guess

The amount of work that had to go into this is crazy. Great job!

Omg i love you ive been trying to figure this out for months

This is brilliant, a very well put together and in depth way to properly rank countries at Eurovision. Well done lads! Looking forward to your field day with this year's results

Wow guys You rock! Thanks

Really nice video, you guys! I have spent way too much time considering formulas to be able to compare Eurovision results between varying circumstances, and have come up with a couple of suggestions that make your’s seem slightly too simplistic. (Yes, I know, I’m likely insane and even have the papers to prove it from my psychiatrist.) Nobody has been willing to discuss this with me, however, so if producer-Matt ever feels like refining his models further I’m definitely up for that conversation!

Great video! Immediately subscribed!

Wow! This is next level ‘Overthinking Eurovision’. Fantastic video team as always. 😊🤯❤

Hey Overthinking it. Love your videos, I always find them intriguing for how deep you go into topics, great for adding to my trivia knowledge. One aspect of Eurovision I've always been fascinated by is the backing singers. They've historically played an important part of performances, adding depth and harmony to the lead vocals. But overtime, theyve been gradually sidelined in favour of more focus on the lead and backing dancers. Starting in 2015, when thet weren't required to be on stage, and then in 2021 when they weren't required to be live anymore. By not requiring live backing singers at all its less likely you get cases in the past where artists started as a backing singer before becoming a lead singer (Hera Bjork, 🇮🇸). Could you do a breakdown of backing singers history please as well as there statistical importance in doing well in recent contests. Is it better to have them live on-stage, off-stage or pre-recorded. Thanks, keep up the good work

Love this ❤ fantastic video 🎉 How did you accomodate for the host country’s into these results especially if they are not part of the big 5

I just love-love-love numbers, thanks for this great vid. But then what exactly am I doing here on a music -elated channel???

This is amazing

Hey folks, great video! Thanks for such a thorough analysis :) Just had a look at the data and noticed that cell BX68 (Italy's 2021 scaled score) is missing a formula. Just wanted to point that out so it can be updated ^^

Very very very interesting!

ngl this level of analysis with eurovision stats is what I did for my coursework in the IB, so this made a lot of sense! This is what I love about eurovision!

I love to see another channel that finally looks at eurovision placings fairly!! The system explained is almost perfect (I think my own was practically the same in 2021) but since then I've found some other ways to improve it: - for big 5 countries (or any other automatic qualifiers) you only set the amount of participants as the amount of finalists if the only countries they placed above were also AQs. It feels strange for example to consider France 2021 as 2/26 instead of 2/39 when Europe considered them better on the night than countries that got through the semis. This way every country is out of the full contest apart from the few AQs typically at the bottom of the final. - considering average points received per country definitely makes more sense than just points overall, but it still creates some injustices- if you were casting these points among 16 countries then it would be much more likely to score points, and less likely to be in someone's top 10 in a final of 26. I go into explaining my system that fixes this problem in this video, although for semi final nqs it only really results in a few minor switch arounds: kzfaq.info/get/bejne/atZnmaSalczLZ6s.htmlsi=p2SBltU97dYuhfVj - also a pretty small one but for any tie (e.g. joint 2nd) that country's placing is considered as the average of the places they occupy (e.g. 2.5) before applying this to the scaled formula. As the 4 1st places in 1969 are all official winners, this doesn't apply to them.

Happy 20.1K plus subs!

Loved the interposed kid sections. Lightened the heavy stats right up and was also very cute :) That said, it'll be interesting to see what changes in the stats this year's competition will bear out...

I sadly have to go to work so i can't finish the video yet but this is gold. Matt's son made me laugh. This is exactly the kind of statistical model i would have tried to create if i was given the same task. 😂

Honestly, I don't follow Eurovision. I just come here to see your fantastic audio work. Doing great!

I work in data processing and visualisation, so I loved hearing about your methodology 😂 I've thought about how to equalize the data over the years before, and I think your system is very well thought out. Though I do miss differentiating between a narrow win and a landslide. I think if you only take the finals data, you could do a % of total points won, to see over the years who won the biggest, what years were very close (1969 comes to mind ;) ). A different set of data, but also quite interesting I'd say

We need something similar to analyze countries' jury votes points vs. telecast votes points each year.

I came to this video expecting to see The Netherlands considering their 8-year streak of not making it to the final, but I had not considered the placements in the actual semi-finals. Great video, thanks!

I like to think that a big 5 country/auto qualifier would have got through the semis if they beat a non auto qualifying country in the final. If you have a cluster of autos from 23-26 you know they probably would have nqd if they had to compete in semis, if you have one in 15th above a bunch of balkans they deserved to be in the final

I saw another channel fix the big 5 problem really nicely. If a big 5 country beat any non big 5 country their perecentage would be out of everyone, but if they were last (or only had other big 5 nations below them) it would be out of how many countries there were in that semi EDIT: Could also fix the problems with winners auto qualifying and australia 2015 btw

Don’t know what others are talking about. I followed every word of the data crunching. Makes sense!

Great video, as a data nerd, this is right up my alley! Though for future reference, the Norwegian broadcaster is NRK (Norsk Rikskringkasting = "Norwegian State Broadcasting"). If you put in a P instead of an R you get Norsk Pikskringkasting - don't google translate that. Or it could be Nynorsk Pressekontor which would be a whole other thing entirely, and if they were in charge, we would probably see more contributions like this year's Gåte every year.

Had to focus, not used to that, but it was worth it!

Math! yay One small thing I thought about while watching though: comparing points received per participating country is skewed too, just in the other direction, eg. getting one point from everyone on average is bad with 11 countries participating but impressive with 100. Admittedly, the difference is probably smaller than with just comparing points received. I just found it funny because you were specifically using it to compare semifinals with different participant numbers.

i did watch all this. it took 8 hours cos i needed to take pauses, but i did it! I also need to say that I did not like how you use map from 1917 (or earlier) as backdrop from you scoreboard.

Congratulations overthinkers for this impressive work 👏🏼 The way you explained your method is absolutely fantastic, it should be taught in schools! I like the simple way you chose to deal with the “big 5” problem. Alex is a real star ⭐️ But I need to point out that I disagree with the way the points are calculated for the non qualified countries. For the countries with high points, more countries lead to more points due to more total points awarded (as you correctly say). But for the countries with low points, which are the ones we care about since they didn’t make it to the final, more countries lead to higher possibilities of low or no points due to increased competition. Thus your method is not fair for the countries in semifinals with more countries.

I like data!

Your math teacher would be so proud... :D Your scaled score is based off what is called the "Cumulative Frequency"

As a stats lover, this is a gift

Now I can back up my jokes about Germany with maths! Thank you!

As a nerd who loves graphs (and Eurovision since I became a fan last year), this video was incredible to watch ! I'm so glad I discovered your channel this year. Thank you for always having interesting things to say about stuff for which I never think there's anything interesting to say (I have no idea whether this sentence makes sense in english, but you get the idea, I guess ?)

Statistics? Yes ! I enjoyed this video way more than I might admit ;-) It might not tell the whole thing though in the bad streak compartment, because in some period countries could not participate the next year if their result was too bad, but that wouldn't be taken into account in 10 year streaks because a countries couldn't have made 8 participations (except the big five). I know this especially because my country Belgium literally played pingpong around the 2000's because we did so badly, but we don't show up in the 10 year bad streaks, but do in the 4 year one which was 2009 where semi-finals existed. This is no criticism though, just a fact that's not able to show up in statistics that I happen to know about.

I found a mistake. You didn't take into account that the hosting country (e.g. Denmark 2014, Austria 2015) is like a big 5 in that year. Austria 2015 came last with 0 points, but you give it a score of .359

Some other KZfaqr uses a system which compares the Big 5/host countries to just the finalists if they finish last or only above other autoqualifiers. But if Big 5/host beats at least one semifinalist in the final, they get compared with all the countries including the ones that didn't get through semifinals. For example, France 2022 ranked 24/25 as they only beat Germany in the final. But UK 2018 was 24/42 as they beat Finland (qualifier) in addition to Portugal (host).

I just know you're leaving Nebulossa for the end cause you have SO much to say about it

4th Edition D&D was probably not the right edition of D&D to use an example, but I appreciate the joke and the data analysis. Great work guys!

Is that Alasdair Beckett-King on Alex'es t-shirt?

I fully expected Alex to be a grown man. Sorry Matt… 😆

18:03 Wouldn't the same also be true for Australia since they debuted in 2015, making 2023 their 8th participation (not including 2020)?

That kid is hillarious,yall should have him on more often

4:16 But but but-- what IS The Algorithm, Alex???? 😲

14:04 The Fragmentation of the German broadcasting system might be part of the problem. But it is important to know two things: These "regional" Station aren't t even exceptionally small on their own, WDR alone has an designated reach of over 17 million people (North-rhine-westphalia) that's more than most participating eurovision nations. Second: These Stations are part of the ARD group, it is possible to cooperate, with each other under the umbrella of ARD, If, and only If the Stations decide to... I am under the impression, that there isn't much interest for the ESC, except for the grand finale.

Instructions unclear, there’s chocolate in my diagrams…

I don't think anything says "I'm in it to win it" quite like Azerbaijan's first 5 years at Eurovision... They placed 1st, 2nd, 3rd, 4th, and 5th - not in that order of course but like, DAMN! :D

As a Czech, I wheezed. We always have something special 😅

But if the UK and Denmark tied in 10th place in the grand final with the same score, they should have the same scaled score. The Big 5 (and host) handicap should only be applied where they don't beat any qualifiers in the final.

I'd say if uk and germany were competing in semis each year, their stats would have gone up significantly. It's just the fact that they are automatically competing with stronger songs that have already proven to be better than 5-6 songs from the semi they have qualified from Also prior to this year big5 only had to perform once while others performed twice over the course of the week, not to mention that someone on Reddit has discovered, that in the finals national juries usually give points to songs that we have competed with them in the same semifinal!

Sorry to break it to you but Eurovisionworld already calculated the semi position thing for every non-qualifying country for every single year haha

Statistics and Eurovision! What a mix! I think I can convince at once my husband, statistics and maths freak to get interested in Eurovision . Thanks!!! 😅 ( great videos by the way, for sure a lot of work to do, and I hope fun ❤)

Question If I wanted to see how a country's result in Eurovision 2024 (say for Example Italy coming last in the Grand Final this year with 0 Points) Could I add a row and add in this example Italys Results and the spreadsheet will work the same or would I have to change either the code or Formula (or maybe even both) by the way I'm not saying I dont like Italy this year or I think it will do badly because in both cases the opposite it true I like Italys song and I think it will do well that was just a hypothetical change I could want to make if I wanted to see how it would impact Italys stats. Also speaking of Italy I have noticed 2 things that do well in Eurovision that Italy (and Sweden) do very well but Im not sure if its just im my head or if there is something to it that is when your National Final (I know calling Sanremo a National Final is controversial but I will ignore that because for the purposes of this calling it a National Final is a good enough term for it) is bigger than Eurovision itself (in the country the National Final is selecting for) it does well in Eurovision my theory for this is the song is selected because the country it represents loves the song. Which leads into my 2nd thing that does well at Eurovision. When a country loves the song that represents the Country it does better in Eurovision than a song that the country doesn't even like despite the fact that a country can not vote for itself.

Random unfounded 4E slam out of nowhere, 4E best E. :)

I always wonder what would happen if Germany sent someone who was actually famous before the contest like, say, Peter Fox. But anyone who's already popular probably didn't want to risk their reputation.

Damn to 11 mins 18 seconds I lost because I didn't see "Actual data starts at 11:18."

If you are not yet members of Mensa, the high-IQ society, you should definitely apply for an admission test!

1.) what about 2004 to 2007 with only one semi-final and 14 countries automatically qualified for the grand final? 2.) 10:38 can you really say "Denmark was competing against 38 countries" or (my theory) rather 32 countries, as it only competed against the countries in its own semi final, not against the 6 eliminated countries from the other semi?

And even then it’s still not perfect because every semifinal had a different qualitative field, a different set of voters and effectively had one country been in another semifinal - the result would have been entirely different. So there you go 😅

Wow!!

The problem I see with Albania and Cyprus' comparison is that Albania would get better results anyway because the points were divided by less countries, thus the result is higher for Albania. To make it fair, we should take into account how many points they would get if they had the same amount of voting countries each semi final. Either Albania 58/21-1=2.9 or Cyprus 63/19-1=3.5. Also, we could say that there are an average of 20-1 voting countries in each semi [(19-1)+(21-1)/2=20-1]. That makes Albania 58/20-1=3.05 and Cyprus 63/20-1=3.32. So, Cyprus would be 28th and Albania 29th.

Sam ryder redeem the uk

You maniacs!

Why isn't Georgia included in the spreadsheet?

can you do the math for results by running order? like analyse how bad 2nd slot actually is

Why is Australia is missing from the results sheet?

Wow Germany did worse than UK... also what was the point of the Rubix cube portion?

Not trying to burst your statistic, but how do you fit the rest of the world vote in count added this year?

Whilst I enjoy this video to a degree, some of the math isn’t matching the way it should. If looking at rankings there should be some kind of weighted multiplier for winning, coming top 5 etc. At a very basic level for example, you state that Ireland winning 4 times in 5 years is only the 8 best rolling average. I’m sure if you ask any of the remaining 7 if they wanted to trade results, they would all say yes. Finally - there is the probability of getting continuous results. Ie the odds of winning are low, odds of winning twice are impossible, winning 3 times is unbelievable etc

1979 norway had oliver, its not average everyday dress... ?

Hi Matt, just FYI: "England" and "United Kingdom" aren't the same. Sincerely, a Scot 🙏🏴󠁧󠁢󠁳󠁣󠁴󠁿 P.S. love your channel

Me, Norwegian gen z: we usually do very well My mom, Norwegian boomer: but we always get 0 points? Mind you, she's been watching eurovision every year I've watched it (at least from 2005 and onward) and she's still stuck on how poorly we did 50 years ago lmao

Is this how you take the fun out of Eurovision? 😆😆

Scoring countries in semi-finals based on average points per country isn't entirely fair and actually biases smaller semis. This is because each country can only give 10 other countries points. For example, if you are in an 11 country semi you only need to beat 1 other country in the eyes of each voting country to score points, while in a 19 country semi that number goes up to 9. Overall I think average points is a better system than total points, but it's still not entirely accurate. The only problem is I have no idea how you would create an entirely accurate system.

0:29 please dont call the UK "England", its like calling the Netherlands Holland or the USA California

honestly, the further we (Germans) get from the last place the more uncomfortable it makes me. I think getting last place suits us very well indeed 😊👍

#SaveGermanyFromTheNDR #RettetDeutschlandVormNDR

11:26 the video starts

Keeping up with my earlier comment about complaining: 🙂 Maths, not math. The full word is Mathematics. The UK & Brits, not England & the English. Wales is gagging to join ESC, they participated in Junior ESC already & i would give them better odds to win than auntie Beeb. In reply to another comment; indeed, it's the Netherlands, not Holland. The kingdom of the Netherlands is 4 constituent countries, of which the European part is by far the largest. Continental Netherlands has 12 provinces & 2 of them are called Noord-Holland & Zuid-Holland. The country had been a kingdom for a while now, but before it was the Republic of the seven united Netherlands, the Dutch republic for short. In between the Dutch republic & the current kingdom there was some hullabaloo with a French dude called Napoleon. During a small part of that period there was a sattelite state of the French Empire called the kingdom of Holland, but us Dutchies mostly don't talk about that. So please use the name the Netherlands to speak about our country, thank you in advance. Oh & this was the most wonderful content. I'm in love with MB... i think... It will blow over soon, i'm sure. Kid got an awesome shirt on.

so, marry me? Or how to say that it was awesome?