Are rating systems worth having?

On top of the subjectivity of rating systems, you have the massive problem of selection bias.

I noted this in another thread recently, but the average score, over more than 40,000 ratings in our database, is just shy of . So either that should be treated as a baseline or (more likely), people tend to watch/rate things they're guessing (more correctly than not) they'll like.

The problem actually reminds me of film criticism writ large: a lot of people feel like they could be movie critics, I think, in part because they imagine the passion and enthusiasm they have for writing about films they love (or sometimes, films they hate). What they don't think about is having to review some incredibly boring, fails-in-ways-that-aren't-even-interesting genre exercise, and having to find the enthusiasm to say something insightful about that. Same kind of thing here, where ratings in general are hurt by the fact that people are not obligated to watch everything, or even a little bit of every type of thing, with any regularity.

First off: don't hesitate to just straight-up ask these things. Lots of people happy to help.

In this case, in the Movie Forums Site Stuff forum, it's the very first pinned thread.

You can also see how someone did something by quoting their post and looking at the code that produced it. And in this case, thankfully, it's also totally intuitive (the tag is just RATING).

Well, if you've been in 3 or 4 out of 10 pain for 2 or 3 weeks, we'll be hearing from you and if not, we should because something is wrong with you. And thanks for you $40, maybe I'll go to the movies.

One of the unanticipated pleasures of being a critic is going in to see a film blind and having the happy expectancy violation of seeing something good, something which you would've never known existed and which you would have never self-selected. It can really knock you on your can. As much as you want to, you can't research everything. And very quickly, you learn that there is much to be recommended for going in blind.

• If it is bad and you never would've watched it, that's not a surprise.
• If it is good and you would've watched it, that's not really a surprise either (you would've self-selected it in the expectation of pleasure).
• If it is bad and you would've watched it, that's an unwelcome surprise.
• If it is good, but you would not have watched it, that's a welcome surprise.

There are compensations to the grind.

I don't think the selection bias is a total inherent negative. The people who rate it are commonly within the target audience, although admittedly not always, and then the target audience is probably making sure it's worth seeing, so they'd be clued into whether that's the case or not, and it would continue that way. A Marvel film could get an IMDb score of like 8.2 and I won't see it because I don't like comic book movies. However a horror film with a 7.1 would appeal to me because it's a genre I like and would be more interested in watching.

There will always be reasons why a rating system is flawed and should be taken with context and qualifiers. Does that mean they're useless? Nah I wouldn't say so. They're useful for a general appraisal. Naturally if you want more depth to a critique, then you go looking for those reviews of trusted reviewers. People wouldn't say "What'd you give it out of ten?" if it didn't hold some weight. You can always go more in-depth and discuss the reasoning behind the number you give, like we do here a lot. Talking to people isn't a bad thing.

__________________
"Some day this war has to end."
"Wash your mouth out with soap!"

For me they are. If 10,000 people rated a movie 4/10, chances are I am not watching it.

I didn't call it negative at all, let alone totally or inherently. I just called it a "problem." Because it is: it clouds our ability to decide what "average" means and what people are trying to convey with their ratings.

As you alluded to, people adjust over time and, consciously or otherwise, come to grade on a curve, assuming they pay attention to it or not. This works well enough, except that people a little more academic and serious might make a special point to avoid these habits/pitfalls, in which case we find that one man's average is a full point or two higher than another's.

As others have mentioned, this is significantly alleviated when taking lots of ratings together, though. In my experience it's very rare for a terrible film to have incredible overall ratings, or vice-versa. But that's a pretty modest kind of utility.

I usually check the ratings before I watch a movie, although I often end up watching it anyways, even if it has low ratings. I like having rating systems, but often they don't end up being a determining factor in my watching or not watching a movie.

my personal rating system is dumb so i assume most people's are also kind of nonsense so idk what it all means when they're all added together. that being said a movie i'm interested in is way more likely to jump to the top of my watchlist if it has an average rating of like 5/10 than if its an 8/10 and idk what this means either.

Fair enough. It can be a problem, like you say. I think there are a lot of variables when it comes to any rating, average or individual, so it's hard to determine what the reader can take to heart.

For myself, I use average ratings of a bunch of people usually when I am on the fence about whether I'll see a film or not. If a film sounds really good and I want to see it and its IMDb score, since that's the site I use most often, is like 5.8, I'll still watch it. If the film has a 7.1 versus a 4.7, then I'm more likely to watch it. It doesn't guarantee I'll like the film, but it is promising. On average, a 7.8 is going to be better than a 5.2. I don't think anyone except me enjoys The Good, the Bad, the Weird more than Godfather. I'm sure there are exceptions like that though that prove the "average rating being a gauge" rule though. The numbers here are totally random and have no bearing on anything.

Ratings are not an exact science. Everyone has his or her own criteria for why a movie gets a 7/10 or a 3/5 or C- or whatever. The thing with having an 8/10 or a 3.5/5 is they're quick. Would you rather see 500,000 ratings as one average or read ten four-paragraph reviews if you're in a hurry? I think ratings and reviews have their advantages and disadvantages and one must determine what he needs at the time when it is prudent to do so.

OK thanks,

Let me try:

Red River (1948, Howard Hawks)

Edit: Success!!

Also I was thinking too about the comment made how people generally tend to gravitate toward things they like or have an idea they'll like as far as rating skewing a bit higher. I was looking at my LetterBox'd account and that is so true. It's the same name as on here iluv2viddyfilms and my rating average is pretty high because of rewatch from films I know I love... for instance I've been rewatching a lot of musicals I love lately because of the countdown so stuff like that.

__________________
"A candy colored clown!"
Member since Fall 2002
Top 100 Films, clicky below

http://www.movieforums.com/community...ad.php?t=26201

See your post makes the point why review ratings are at their most useful when a written review is also included.

Red River (1948) Why not 5/5 because of this:

The wagon train scene when the girl is shot through the shoulder with an arrow, she doesn't even flinch...she actually smiles. Say what?


As much as I like Red River it's not a

as it is not perfect.

It's not even in her shoulder; it's in her shirt. Maybe she's ironically smiling at the Indian's poor aim? I'm going to hold fast to the notion that a score is helpful and a detailed review is even more helpful. It depends on what you need at the time.

Right it's not in her shoulder, but the film clearly shows that she is suppose to be shot through the shoulder and they even pull the bloody arrow out. It's totally cool if you or anyone rates Red River a 5/5. I was just demonstrating how my 4/5 rating had more meaning with more information. BTW that image and passage I posted was from my MoFo review.
My Red River (1948) full review here.

That's just one brief scene though. I don't think it holds much bearing on the plot and is not worth factoring into the final assessment. That's just my two cents though.

__________________
IMDb
Letterboxd

Actually for me the ending is what kept Red River from being a 5/5. Remember I did score it a 4/4 and the vast majority of the film is perfect. But I view the ending is flawed...

Well, I do agree with you there.

I saw Red River a few years ago and wasn't a fan. I think Wayne and Clift got on my nerves, and some of the scenes didn't resonate with me. I'm a mite picky with my Westerns. It wasn't even a vendetta against Wayne nor Clift nor Ms Dru; the performers are fine. I just didn't like their characters.

Sorry; I'm straying off topic. My review wouldn't be as good as some, which harkens to the score rating number thingy being imprecise and subjective. Corax made a good point that the outliers will average themselves out with a large enough average. That's why it's an average.

I think it could've been a great film. They just dropped the ball at the last second and erased some of its potential.