Chart Ratings

[Guest]

Alright, BEA, it's time we sat down and had a serious talk. The rather insidious evil that is chart ratings has grated on my conscience for basically as long as I've been here, and an_outlaw's most recent hijinx, along with the discussion that has followed, has alerted me to the fact that I'm not alone.

"Evil?" you say. "But the charts are what this site is built on! Certainly it must be one's primary objective as a member of this site to have a high-rated chart that is loved by the masses."

No. No it mustn't. Let's start with the basics:

A chart is, essentially, a tool one can use to conveniently list his/her favorite albums for the viewing pleasure of other users. It's true that charts can be good for recommendations, as well, but this does not strike me as their primary purpose (I realize some disagree; I will address this shortly).

So what is it exactly that we're rating? If someone's chart is an honest reflection of his/her favorite albums, then certainly it must be ludicrous to deem his/her chart less valuable simply because one does not agree with his/her taste. What is even more bizarre to me is the way people will give a lower rating because a chart is not "varied" enough. Why do I say this is bizarre? Statistics seem useful in this situation. Consider the following example:

Dennis is an avid music listener. He has listened to precisely 1000 albums and has ranked every one of them. Among these 1000 albums are 10 Beatles albums. Now let's assume, for the sake of argument, that this ranking is completely random (considering the subjectivity of musical taste, this should be a fair assumption). In other words, any particular album is equally likely to be in any given rank. Statistically speaking, a top 10 consisting solely of Beatles albums is just as likely as any other particular top 10 (disregarding order). And yet, if Dennis, after serious contemplation, determines these 10 Beatles albums to be the 10 greatest albums he has heard, he will inevitably be punished by low ratings should he choose to stay true to his tastes in making his chart.

Is this horribly improbable? Well yes, but as I pointed out, so is any other particular top 10. Now, the obvious rebuttal to this is something like: "Sure, it is technically possible that an avid music listener will have 10 Beatles albums in his/her top 10, but it is incredibly unlikely. More likely than not, any user with a top 10 consisting of a single artist is rather close-minded regarding music." Okay, granted, but is an average chart rating of 30/100 really a suitable punishment for being a statistical anomaly? Hardly. Even if a user reveals him/herself to be fairly ignorant in the forums, to give a low rating because of this would be more a rating of the user than of his/her chart.

Now, how about comments? Okay, yes, comments are great, and rating a chart based on the quality of comments seems pretty fair (though it could be argued that this is a judgment on the chart creator's oratorical prowess, not on the chart itself). Obviously, though, this is not what people are basing their ratings on. Just look at MrFrogger's chart. Despite being ranked #3 on the site, it is almost entirely devoid of comments. I don't mean this as a criticism of his chart (if I did that would rather undermine this entire post), but if comments were the driving force behind most ratings, it should be fairly obvious that Frogger's chart would not be rated so high.

So what about recommendations? As previously mentioned, the purpose of the charts is to share your favorite albums with the world. So what if almost all of your favorite albums happen to be pretty well-known? Now, some users use one-album-per-artist or other such rules to make their charts more diverse, thus increasing the quality of recommendations, but it seems perfectly logical that a user would gear his/her chart towards simply representing their 100 favorite albums as honestly as possible. Now, I'm not about to go full Naples on you guys and insist that my method of chart-making is the only proper one, but it is a reasonable one, and so users should not be punished for employing it.

The average chart rating that I give out is currently at 92/100. I don't see any reason to give a chart a rating below 90. Why should I? To make the creator feel like his/her taste is less valuable? To make sure that chart gets less exposure? To scare the user into altering his/her chart to be more diverse, thus increasing its rating but simultaneously making it less genuine? Do you see how ridiculous this is? Charts should not be rated because they represent a user's taste, and taste is not something to be graded.

Now, I'm not going to start a rebel faction and go on forum strike until chart ratings are removed, but I thought I'd post this as some food for thought. To those of you who disagree: What is your criteria for rating a chart? In what way do you feel I've misrepresented the intention of chart ratings?

Apologies if this post was rambly. It's past my bedtime.

[Guest]

[Jackwc]

I'm sorry, I had to stop reading halfway because it was stupid.

If Dennis thinks those ten Beatles albums are the ten best albums he's heard, guess what? I disagree. So I'm rating his chart poorly. That's how this works, get it? I don't agree with his taste so I rate his chart poorly. That's kinda how this works. I know, right? Mind blown.
_________________
A dick that's bigger than the sun.

Music sucks. Check out my favourite movies, fam:
http://letterboxd.com/jackiegigantic/

[SingingPeasant96]

I never thaught about rating chart seriously, If the chart gave me impressive recommandations or something, I rate it high, I guess many of us do same. Why so serious?

Did you get bad rating from someone and get mad?
_________________
If you're feeling sinister, Go off and Listen to Indie Pop

[SquishypuffDave]

Chart ratings show how much a chart is valued by the community. For whatever reason. That's all there really is to it. Ratings don't have anything to do with what the creator thinks of the chart's content. Usually that's self evident: it's their favourite music. Ratings are for everyone else. If you're making the chart primarily for yourself, then it shouldn't matter. If you're making the chart for everyone else's enjoyment and you get a low rating, then either accept it or make the chart better.

[Polythene Pam]

Rate peoples chart if they care about what you think. They don't.

[Applerill]

I think it's a lot like Rotten Tomatoes Aggregation. These rating may seem "objective" at first, but you can see that it's rigged towards the unthinking majority.

Mulholland Drive has an 82% on RT, while the Avengers has a 92%. Yet if you tell any of the Sight & Sound editors that the latter is better than the former, they'll probably laugh at you.

This isn't to criticize The Avengers. But in that same way, chart ratings (and member rank) mean nothing.

I can feel pretty crummy about my 82/100 overall chart, but then I remember that some of my favorite people love it. That's more than enough for me.


PS: If you have Atoms for Peace on your overall chart, you are officially screaming out "I'm an idiot! Give me a 5/100!!" Just sayin' (Once again, not because it's a bad album at all, but because it uncannily shows how much cultural capital you have)

[Mother Nature's Son]

[Polythene Pam]

[ButterThumbz]

Surely, this is a "problem" which is easily solved. If you don't agree with rating charts, ignore the rating system and leave it for those who use it to worry about.