The original article already mentions 'leotard bias' as the problem that athletes from smaller countries face some prejudice in judging compared to the ones representing a superpower, and letting viewers judge would amplify that problem much more - it wouldn't even be about what show/PR the athlete puts on to the audience but highly influenced by global politics and what their country's leadership does.

That strategy is not without flaws either: once you let the public decide you create a system where the incentive is not "who skates better?" but rather "who can play the audience better?". The EuroVision Song Contest, for instance, has a well-known issue where the public often votes for/against a singer not because of their song but rather because of geopolitics [1]. And that's before we discuss organized voting rings.

A related but less clear point is whether it's fair for the skater to be judged by people who have no idea about skating. If my code were to be negatively judged because (say) the person in charge judges the amount of comments rather than their content, I would probably be unhappy about that.

[1] https://en.wikipedia.org/wiki/Eurovision_Song_Contest#Politi...

And then 3000 Japanese fans outvote everyone else :)

I’m looking into Eurovision judging as a curious example (half votes come from public and half from judges) :)

Viewers are not impartial. Viewer voting would mean the winning team just becomes whoever has the most fans watching.

People will buy sim cards or accounts and automate spam voting.

Voting? That will never do!