zlacker

IMO audience score is more of a proxy for movie quality minus movie expectations. A lot of niche genre films do okay because 1: the only people who watch them are enthusiasts and 2: nobody goes in with high expectations to begin with. If everyone watched them and reviewed them they'd do much worse.

For a mainstream example, take Fast X. It's an objectively stupid movie with a great audience score - because it's exactly what it says on the tin! Nobody is confused about what they're watching. Nobody thinks they're going to get terrific drama or romance or suspense. They're going to get the 9th sequel to a comedy action movie about dudes driving cars.

>>Abraha+(OP)
Yeah, the audience score is in no way an accurate measure of quality because it is provided by a self-selected group of people who both paid for the movie and went online to rate it. The end result is that a lot of movies viewed as failures will have high audience ratings as long as they can reach some small passionate audience. Just looking at some current movies:

Blue Beetle: 78% critics, 92% audience

Gran Turismo: 63%, 98%

Elemental: 74%, 93%

Meg 2: 29% 73%

Haunted Mansion 38%, 84%

Indiana Jones: 69%, 88%

Little Mermaid 67%, 94%

Those audience scores are not "more accurate" in any way. People who are forced to see these movies like them less than people who chose to see it.

There also really isn't anything currently that fits into "the political correctness / marketing budget of the movie" claim of OP. It seems like they are just buying into cultural war nonsense. The closest I can find is Barbie and its critic score is 5% higher than the audience score, so not much of a gap.

>>slg+w7
The one useful aspect is that if you only care about the scores of movies you were already interested in, the audience score actually is quite accurate. If you already know "I am the sort of person who might pay to watch the little mermaid" then you can have good confidence you'll like it based on the audience score.

>>Abraha+(OP)
Agreed, but you can infer a lot with it.

My favourite way of judging movie quality is checking what kind of movie goers hate it and why.

>>mpsprd+Gc
This does not always work. For example: Knock Down The House started off with an excellent audience score and stayed that way for about a year...until Tucker Carlson mentioned it on his show and then it plummeted to its current score. So how can you infer whether the quality is good?

>>slg+w7
The ratings for those movies are so far away from how I felt about them (they all landed < 30% for me) it actually gives me an idea for a site.

You rate movies. The consensus ratings you see are based on people who have rated movies similarly to you overall.

>>Abraha+(OP)
Personally the way I see it:

High Audience Score: HA High Critics Score: HC

HA + HC: Great movie to watch

HA + no HC: An entertainment movie but don’t expect a masterpiece

No HA + HC: Avoid unless you have obscure tastes or if you are pretentious

No HA + No HC: Probably garbage

>>Abraha+(OP)
I feel like these are added dimensions beyond existing rating systems but I haven't found a good way to capture the data or communicate it.

At a high level what I really want is two ratings: Global rating and does this deliver what it promised. A greasy spoon is objectively not a good restaurant but it scratches an itch and you have certain relatively low expectations of it so in the context of greasy spoons generally I might rate the restaurant 5 stars even if globally I'd give it two stars.

As you say with Fast X: objectively it is not a good movie but it absolutely delivers what it promised. People who like that movie series will be pleased with it so in that context it deserves a positive rating.

As a follow-on I want to tell the system about the things I like + the things I hate. Then I want the system to give more weight to ratings from others who both like and hate the same things. I honestly don't care if critics or audiences liked the movie... I want to know if people who in some way think like me enjoyed it.

>>nebula+8g
you sample bad reviews from users and read them. You can most of the time filter it out the classic noise:

-Pretentious watchers hating on summer blockbuster movies

-Political things and review bombs

About your specific example Steam comes to mind, they have a great review timeline feature that allows to filter out review bombs.

>>mpsprd+Sp
I suspect the negative reviewers got what they wanted: they feel better because they 'helped' in damaging someone they hate in some small way and anyone who just glances at it would pass on the film.

One feature that would be nice would be a filter to filter out reviewers based on certain criteria

-only one review

-only has reviews on certain films

-account life is less than specific threshold

That DB query is probably too expensive to run on a free site though. An app that scrapes the RT reviews and filters out based on this criteria has been on my list of things to build.