Ratings of Bruce Springsteen & Radiohead albums

@Greg:

Edited by jamesbaldwin - June 05 2023 at 15:32

jamesbaldwin wrote: And this is the reason why, as I have already written, to obtain a coherent ranking of quality (beauty) of the music, it is necessary to make sure that all the voters have listened to all the nominated albums, as happens in film or song festivals.

@Stressed Cheese

jamesbaldin wrote: @Greg: "229: Chico Buarque - Construção 4.03 / 6,634 ratings/ 76 reviews 230: Led Zeppelin - s/t: 3.92, 33,054 ratings, 331: Neil Young - Harvest: 3.98 / 18,328 ratings / 603 reviews Led Zep IV is considerably higher rated, with 4.09 / 5.0 from 41,923 ratings and higher ranked at 64, but I am just using the first as an example when you see three in a row with such different numbers of ratings. That's a huge discrepancy in the numbers or ratings. Beauty is in the eye of the beholder, the numbers are something else. Which of these three is actually the most appreciated? I would say Led Zeppelin because it has the most appreciators even if it has a .11 lower rating than Chico's album." Greg, I know very well Led Zeppelin, I never listened to Chico Buarque. Perhaps, if I listened carefully to it, I might like it better than Led Zeppelin. Or maybe less. Therefore, I would not really say that Led Zeppelin has more appreciators than Buarque, I would only say that he is better known in the north-western world, where most of RYM's users reside, and therefore, being more listened to, it has more ratings.

I just meant something that doesn't need saying, that Led Zep has more appreciators at RYM, not talking quality (that has not been my focus, my focus has been in trying to interpret the numbers and what one can infer from them). I just find it interesting that sequentially ranked albums have such variance in numbers of ratings while having similar average ratings, plus or minus .11, or .O6 out of 5. Okay, that can be considered a lot (liken it to the miniscule difference in time sometimes between three people in an race). We had been talking about appreciation, Led Zep has come into the conversation before, you felt a lack of appreciation or recongition as I recall for Neal Young (sorry if I now misinterpret you), but it is such a numbers game. The charts are really complex things, and when dealing with the kids of numbers they do, try to balance known-ness (number of ratings) and the perceived quality or appreciation of an album (average ratings). The numbers determine the ranking, but one can weight things differently to achieve at different rankings. M@x once changed the algorithm at PA and it resulted in very different rankings. There's more to the story when judging the cumulative appreciation index than just viewing the rankings.

The number of ratings, in other words, has very little to do with an album's score.

And the number of ratings can have a huge affect on the ranking depending where. At PA, it makes a very big difference, and at a certain point having more ratings can hurt your ranking. It can be counterintuitive but mathematically sound.

The Quality Score can go up or down by making the number of listeners the same for each album.

Yeah, if you use some systems, but some systems don't work well for huge numbers of users raters. I'm not trying to make points about preference or what system is better, I'm submitting some of my observations of one specific system based on a very limited set. What is the better system depends on many factors, like obviously what you are trying to accomplish, what the data is meant to represent, how the data is taken etc.

And this is the reason why, as I have already written, to obtain a coherent ranking of quality (beauty) of the music, it is necessary to make sure that all the voters have listened to all the nominated albums, as happens in film or song festivals.

Beauty is in the the beholder and therefore it depends on the individual is how I tend to think of beauty in art.. Some systems are better to find consensus. That can be a better approach when trying to come up with a best of list, and I shared my thoughts on that earlier in the thread. Now I still have major issues with groups who listened to shortlists based on larger list which is itself limited yearly list because then I would think everyone should have listened to all albums, otherwise it's just a best of whatever albums were considered and listened to. I'm not a great admirer of best lists period, especially when it to arts that is so subjective. I don't take best list or ranking that seriously even though they can be a useful too, for me discovering music if I narrow my parameters, and still I might have to check out many myself before I find one that resonates with me. What matters to me most is if I like it, but if many others like the same, then it does become easier to find usually.

So if by coherent ranking you simply a consensus based on an agreement of the qualities of a set that all are familiar with, then okay. I don't treat RYM's list as a list of quality despite the fact that I find plenty of good qualities in the music. I don't claim, and I hope I don't act like. my tastes are better than others.

With what you quoted maybe this better expresses my interest, and offers insight on what I was trying to bring up. It's about getting that balance where they want it.

To sum all of this up, I would rather focus on a quantitative analysis of RYM's charts than a qualitative analysis. I think when talking about RM's chart specifically that makes more logical sense if one is interested in logic.

Stressed Cheese wrote: That's the tricky thing. Ideally, if you're a site like RYM, you'd want to take quality into account above all else. But you'd also not want really obscure stuff to infest (for lack of a better word) the tops of the charts too much. I know that they won't consider the average rating of a release with a single rating a true average, but they could do something like Letterboxd, and just not show an average at all until there are an X amount of ratings. Then again, their ranking system already filters releases like that out, so I guess it doesn't really matter at that point.

Edited by Logan - June 05 2023 at 19:21

Stressed Cheese wrote: Please note that I was typing this post and then 3 more posts appeared after I finished it, so this is mostly just in response to Lewian... Lewian wrote: Err, what!? If RYM users do the ratings, the ratings tell us about RYM users and not about anybody else. I don't see why, given an artist that has enough ratings, you can't say that an album's RYM rating isn't a pretty good indication of what the average person thinks of it. Yes, not everybody uses RYM, certain demographics won't be as represented (it's a western website, and people over a certain age don't use the internet as much). You can see this reflected in the amount of ratings (e.g. Fishmans' most rated album actually has more than U2's most rated album, even though U2 on the average still has more, so Fishmans was kind of a bad example on my part). But I think that this has more of an effect on the amount of ratings, than on the actual rating itself. Because there isn't really such a thing as "RYM users" as a general group. I mean, there might be, to some degree, but it doesn't translate over to the album ratings as long as each album/artist still attracts only the appropriate sub-group of users. E.g., there might not be that many people on RYM who grew up in the early 70's compared to people in their 30's (just guessing, but I think it's safe to assume), but as long as prog albums get rated by people who appreciate prog, metal albums get rated by people who appreciate metal, etc. I don't see how this would skew things too much. Now maybe I'm wrong. Maybe The Who's Tommy would have a much higher rating than 3.73 if there were more retired people on RYM, I don't know. But the fact still is that you wouldn't rate that album if you didn't at least have an interest in classic rock. Plus, I don't think RYM is by any means perfect, and they're transparent about the fact that the ratings reflect the here and now. If an album used to be really beloved or hated, and people now view it as overhyped or a hidden gem, that'll be reflected. But then again, you can only rate stuff based on your own perspective. And by the way, I haven't checked a ranking but I'd be pretty sure their ranking algorithm would prefer a U2 album with loads of ratings and a somewhat lower but still good average to Fishmans with a higher average and a handful of ratings only. But I'm not sure because of lack of transparency. (I can see that albums with fewer than 100 ratings and an average larger than 4 do not feature high in rankings.) I do think the ranking is somewhat affected by popularity, yes. The RYM faqs only mention user activity/reviews, but it wouldn't surprise me. At least for the ranking, rating I have my doubts. I would be interested to find out as well. For example, Tommy should be a couple points higher if you count every rating equally (yes I just calculated that...), but can the fact that inactive users and users who only rate at the extremes get discounted account for this entire gap? Probably an easy way to find out is to calcuate the average by hand for a couple of albums and see if they all end up being higher than what RYM gives as the rating. I assume that would be the case but I don't feel like doing that now, tbh. What is the "scientific sense" of the word that could apply here? I'm a scientist and "validity" statements are part of my bread and butter, and I can't make sense of this statement. It is what it is, it isn't what it isn't, and Lorenzo is just as well entitled to his preference of curated lists as you are preferring RYM. Ok maybe I was talking out of my ass a little, and the validities I am experienced with from my studies don't really apply here, truth be told. I'm kind of struggling to express this properly in english for some reason, but what I was trying to get at is that if you look at a RYM chart, that'll more likely reflect people's overall opinions than a more curated list from music critics. I am certain of that. Like, if someone gets more out of a list by Rolling Stone or whatever, that's fine. I think it's not very valuable (though possibly interesting to read through), but it's ultimately just for fun, and RYM ultimately is just for fun as well. Nobody is going to enjoy their favorite artists more or less because of what RYM thinks of them. But asking X amount of people from the music industry, or who write music review for a living what they think are the best albums, while interesting, is not going to translate over as well to what people think in the real world. But anyway, that's enough sounding like a RYM shill for now.

Edited by jamesbaldwin - June 06 2023 at 06:43

Edited by Lewian - June 06 2023 at 07:19

^^ (to Lewian) I just tried to explain my thoughs. "It is what it is and I like it for that" is not how I would phraze it, because it's not what I believe to be true. I don't think like you do. I don't doubt for a second that the collected knowledge of a music community like RYM is of a greater value than 650 000 random people.

With all this talk, I registered at RYM again. I had an account years ago. My user name is LoganPA1 (same as I had for a youtube channel which I cannot access anymore). The Logan is for my PA username which is a reference to Logan's Run due to my propensity for run-on sentences. The PA is for Prog Archives, and the 1 is because I am number one (big The Prisoner fan).

I will start rating and reviewing some of my favourite things. I'm not interested in rating that which I don't like generally or from genres I don't appreciate. I don't think that rating albums from genres which I don't "get", am not that familiar with would be of value to myself or anyone. I value reviews and ratings more from people who can appreciate the idiom and are familiar with much related music. I do value the opinions on and the valuations of music, arts, sciences etc. from people who are well-versed in the subject matter/ genre/ field. I generally would rather a medical diagnosis from a medical doctor than an auto mechanic, and I generally would rather the opinion on the worth of, say, a hip-hop album coming from someone who appreciates and has explored plenty of hip-hop.

And, not surprisingly, I value those who have similar tastes to mine.

I don't care much about best lists unless they are best for me, and actually find ridiculous levels of focus on best lists, especially those that imply an unreasonable level of objectivity in the assessment. I have found it wonderful how well-rated so much music that I love is at RYM, and I have discovered so much great-for-me music through going through those charts (mostly custom fitted so maybe less surprising). It does attract many into quirky artsy music. I was listening to Kendrick Lamar's To Pimp a Butterfly earlier, which I do find to be a very interesting and eclectic album, and at times I found myself comparing it to black midi, which is popular at RYM. It took me a while to really appreciate black midi but I now love various albums -- Cavalcade is still my favourite. I could well see a big overlap in the audiences despite the differences.

Like many, I prefer to rate that which I like, and it is Rate My Music, which implies to me rating the music from my collection and the music I like rather than evaluating the music others like. Rate Another's Music (RAM) would be cool. The Tagline could be RAM, for the music other's wish to RAM down your throat. I have tried to make topics like that at this forum, but they did not work as well as I hoped they would. A lot of PA users are a very tough crowd. ;)

Edited by Saperlipopette! - June 07 2023 at 03:13