Bias in album ratings

I had a thought about album ratings this morning and just thought I'd share it here even though there's no real practical application for the idea.

I believe that for album ratings to be anywhere close to objective, newer albums should receive bonus points simply for being newer. Of course one might wonder, why even strive for something like that since ratings are inherently subjective? Well, imagine that you never heard of Opeth. You're just checking them out and you go to their page on this website to look at the ratings for a good place to start. My theory is that if this was done as an actual experiment, and we had these first time listeners listen to just one album and rate it, with a huge sample size of people listening to different albums, there is virtually no chance that Still Life or even Blackwater Park would come out higher than Ghost Reveries OR Watershed, the way they are in the ratings right now.

This whole thought came from listening to the new Opeth album. There's a good chance it'll grow on me but my first impression of it is... kind of forgettable, honestly. However, I realize if this was my first time hearing their style, this album would have me obsessed. So, the mismatch between its objective quality and my subjective experience of it at this point is significant. In this case it's bias against the familiar but bias towards the familiar can cause the same thing.

There's also the matter of the general downward trend in the ratings of any band's albums over time. Sure, there are exceptions but they are just that, exceptions. Is it logically likely that all bands tend to get worse the longer they are active? Not for any reason I can imagine. Maybe I'm way off here, but it's something I can't unsee, to the point that I look at the rating of a new album and think "okay, it's this band's 12th album (or something) so it's probably a lot better than that", and quite often, it is.

The subjectivity never really goes away no matter what when it comes to ratings. Not only that, people have different standards that they judge by that, while less subjective, nonetheless lead to different rating standards. No matter what, good albums will always rise up through the ranks. But great albums usually stand out from the start. The difficulty with metal is that we often judge this or that album firstly again the previous works of the band, then against other similar bands. And lets face it, a lot of albums take time to grow on people. That has always been the case, and it is especially true in metal. The other thing to consider is that if there there is a 10 there should be a 1 somewhere, and if a ten constitutes your favourite albums of all time, then a one should be the most detested. But What constitutes a 1? Are we only talking about metal? If so, then why is the lowest ratings usually no less than 3 or 4 with only a few exceptions. This tells us that people are factoring in other kinds of music into their ratings (which I typically do). If a Madonna album is a 1 then the worst metal album is probably around 5, but if a 1 in someone else's eyes is Metallica's Lulu, then they would have a very different rating standard and 10 would likely be vary rare indeed for that person. In any case it is a subjective rating and only time and patience will sort it out and I don't think a "newness bonus" would fix the problem, because what about a "Lack of production technology" bonus for albums released prior to 1990? It's a slippery slope offering objective bonuses to an inherently subjective activity.

Written by DarkVoid on 02.04.2025 at 09:14

if there there is a 10 there should be a 1 somewhere, and if a ten constitutes your favourite albums of all time, then a one should be the most detested.

I personally agree on this take, but not necessarily everybody would or should I'd say. It depends a lot on whether you interpret the rating scale in relative terms (which I do), saying 10s are for my most beloved albums and 1s for those I like the least, or you take the numbers by an absolute interpretation (such as the MS rating scale suggests) where a 10 is 'perfect' and a 2 is 'pure shit'. People going for the latter could just argue "Metal will always be perfect or near perfect for me" or "There's no shit in metal", and that addresses the other point you make.

Written by DarkVoid on 02.04.2025 at 09:14

if there there is a 10 there should be a 1 somewhere, and if a ten constitutes your favorite albums of all time, then a one should be the most detested.

I can kind of agree, but I would say the majority of users listen to albums from more popular bands and these are always going to range from about a 6-10. Albums that deserve a 1 fly under the radar for the most part. The real explanation for low ratings comparatively is that most albums pre-2005 were here before the site had a decent user base so users retroactively vote on their favorite classic albums and those albums didn’t get the criticism that new releases get from the roughly 20 users now who listen to releases each week and rate them accordingly. Most people aren't going to remember a bad album they listened to from the 90s and go rate it, but they’ll rate a bad album that came out last week. The other thing to consider for a 1 rating is that most artists put effort into their music and unless something is absolutely grating on your ears it’s not likely to get a 1 unless someone is manipulating top 20 votes. If you went to buffet and tried everything and liked all of it a decent amount and someone said give every dish a rating 1-10, you’re not likely to give a 1 because even the bad parts have something you like. I think most albums if they were rated totally objectively would land somewhere between 6 and 8 with a few outliers on either side.

I knew it subconsciously this will lead to create a forum very soon. Wish I'd have posted this before.

Good flowing.