[Cheddar]

I'm sure that over the 20 years of this site's existence that this conversation has been had before, but I wanted to throw this out there because I've been thinking about as I've been putting together my 2025 album chart. As I've been listening to more albums and the chart has been changing, it's been sort of sad to see albums I greatly enjoy lose their points as they got pushed too low or fell out of the chart all together. There's probably 30-40 albums that are now off of my chart that I think could be considered top 100 quality, but will never get rank score points from me.

My thought is to have some way to incorporate album ratings into the point system for albums. Now, this is not a ratings-based site. It's a rankings-based site. The charts are the heart and soul of how albums perform, and I think it should remain that way, but I think having a way for rewarding albums with good ratings could do a couple of things. First, it could increase engagement with the rating systems on the site. If there is a direct reward to an album for being rated well, then more people would be incentivized to use ratings. Secondly, it could give more visibility to albums that aren't as popular and therefore don't make it onto as many charts. There's a lot of great music out there, and this would be another way for albums that aren't as widely popular to gain visibility in the aggregated BEA charts.

As I've thought about this, I've thought about how this sort of idea could be implemented in a way that wouldn't be able to be gamed or abused as well as something that wouldn't suddenly inflate scores or cause wild upheaval in the charts. You can't just give points based on a flat ratings threshold, which could be quickly abused, and I don't think you could make it a scaled point system like each level of chart has without giving ratings too much undue influence on the system that's already in place. What I've landed on is what I believe to be a way to reward ratings that is simultaneously subtle, not chart breaking, and can't be easily abused.

How it would work is as follows: For every rating you give an album, BEA looks at that rating against your own average across all of your album ratings. If the album rating is equal to or above your personal album rating average, BEA awards one rank point to the album. Otherwise, no point is awarded. For an example, I currently have an average album rating of 82. This means that if I give an album a BEA rating of 85 or above, the album would get a rank point because I rated it above my normal album score. If I rate an album 80 or below, it wouldn't receive a point. You could even give it a unique metric display, displaying how many people rated it "above average" or something like that. Basing it on average rating means that it would be harder to game (unless you suddenly try to rate a bunch of albums poorly en mass to lower your average, which I believe would be fairly easy to spot) and only awarding a single point means you still need a plurality of people to think that this album is above the quality of their average rating to really affect the rank score. You'd also probably want to have a rating threshold so that only accounts with a certain number of ratings (maybe 1000) could have this feature apply to their ratings. That way it's harder to influence your rating average or influence things immediately upon joining and starting to rate albums. It also means that many albums might not even get that point if an album gets included in a chart, but gets a rating below someone's average. I think it would add a bit of nuance to what ranking an album in a chart vs. rating it means on BEA.


There's a chance this wouldn't be very useful or would just inflate all ratings equally and be moot. Y'all can shoot this down and put me in my place in this case if need be. I am a newbie around here (I made my account last year but didn't become active or my make my first chart until a few months ago) and don't mean to jump in and immediately tip over the apple cart. It was just a thought I had wishing it would be nice for albums I've rated but not yet added to charts or have been pushed out of charts to still be rewarded since I still like them a lot.
_________________
My overall top 100 is far from fleshed out at this point, so here are some charts I am actually proud of:
2025

[albummaster]

Hi Cheddar, thanks for putting so much thought into this and taking time to post. Yes, the general idea has been discussed before, but not for a while. In the background I've spent a lot of time over the years pondering how we could combine ratings and rankings into a unified system. But every time I've gone down that rabbit hole, it always gets too convoluted (whether it's derived from user averages, rank score multiplied by overall average rating etc, there's so many ways to do it).

The core issue with using user averages as a baseline is that it would introduce a "moving target" problem. If you rate five new albums highly tomorrow, your average changes. Suddenly, an album you rated weeks ago might drop below your new average and lose its bonus point through no fault of its own.

There’s also a concern that once you introduce a rule like “above my average = contributes”, people start adjusting how they rate things to influence that threshold. So instead of just reflecting how much they like an album, ratings can start to become about how to influence the scoring system.

On top of that, there's a problem for new users who don't have a big rating history yet, not to mention the fact that everyone scales their ratings differently (a 70 from one person might mean the same as an 85 from somebody else). Some people only rate the albums they like and so average ratings are generally inflated with a low standard deviation (& other people will then rate an album lower than it should be to game the averages). These issues could be solved, but add extra complexity to the overall system and things are relatively simple and easy to understand right now.

I totally agree with the underlying issue being not able to add more than 100 albums into a chart which might cause albums not making the cut to be more buried than they should be, but the more straightforward direction that has been discussed quite a bit before is to allow bigger charts (which preserves the existing logic). Bigger charts are planned to be launched as part of the premium membership scheme and are ready to go in the background once the decks are clear.

(EDIT: Sorry, overlooked your workaround for new users where this only applies to users with X ratings, but other parts still valid about complexity and influencing how users rate items. Moving target issue also remains to a degree depending on qualifying threshold).
_________________
Bluesky · Facebook · Spotify · X