Best Ever Albums

Re: Rating vs Ranking discrepancy

SuedeSwede — Sat, 28 Sep 2013 21:05:00 GMT

Author: SuedeSwede

Posted: 09/29/2013 01:05

I just put the album where I think it deserves to go in the chart and then rate it dependent on where it is ranked, for example, if it is sandwiched between two 85's it'll get an 85, if it is between an 85 and an 80, I'll decide whether I would give it a strong 82 or a weak 82. Anyway, I only do ratings based on other albums - my favourite albums get 100 (I think I have 4 or 5 100 rated albums), and then it just decreases from there, there really is no basis to mine.

Anyhow, it's been said before, a low ranking in a chart (something like 80/100) will still give it 21 points, but a rating of 10/100 will decrease its rating - people have albums low in their chart even if they don't like it so much.

Complicated shiznit though.

Re: Rating vs Ranking discrepancy

Anonymous — Sat, 28 Sep 2013 16:45:42 GMT

Author: Anonymous

Posted: 09/28/2013 20:45

I've mentioned this before, but the album ratings I give are always unpredictable. I have 90 and 95 albums in between 100's. In fact, Nevermind is in between some 100's and I give that album an 80.

Re: Rating vs Ranking discrepancy

Necharsian — Fri, 27 Sep 2013 18:23:39 GMT

Author: Necharsian

Posted: 09/27/2013 22:23

tekin wrote:

Yeah there are albums with 85 and 90 on my overall chart but there are 100s that i didn't include; that's basically because my ratings are mainly stupid and change a lot perhaps with every listen.

Haha ya my rating pattern is pretty dumb too. I started just giving 100s to anything on my overall chart, so if you see a 100 from me its likely been on my chart at some point. But pretty much anything from 80 up could potentially be there id say.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 18:15:02 GMT

Author: Anonymous

Posted: 09/27/2013 22:15

tekin wrote:

Simple: you have low ratings which decrease the overall rating but low rankings increase the overall points.

Yeah, this especially comes up with the year charts. Should a user include an album on their 1950 chart even if they didn't like it? I don't know if digital recordings even exist for 100 albums from that year.

I generally don't put albums on a chart unless they're at least above average, but I don't know what others do.

Re: Rating vs Ranking discrepancy

Dingerbell — Fri, 27 Sep 2013 17:30:22 GMT

Author: Dingerbell

Posted: 09/27/2013 21:30

tekin wrote:

Yeah there are albums with 85 and 90 on my overall chart but there are 100s that i didn't include; that's basically because my ratings are mainly stupid and change a lot perhaps with every listen.

Couldn't have put it better myself. I rarely change things like my decade charts of ratings, because it's too much effort.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 17:25:40 GMT

Author: Anonymous

Posted: 09/27/2013 21:25

antiloopje wrote:

Maybe it's time to turn this thread around: are there albums you rated high, but didn't include in your charts? And for what reason? Personally, my chart is a dull copy of my ratings, but I'm interested in your criteria.

Yeah there are albums with 85 and 90 on my overall chart but there are 100s that i didn't include; that's basically because my ratings are mainly stupid and change a lot perhaps with every listen.

Re: Rating vs Ranking discrepancy

Dingerbell — Fri, 27 Sep 2013 16:28:30 GMT

Author: Dingerbell

Posted: 09/27/2013 20:28

antiloopje wrote:

Sorry for the double post, but I've just thougth of some more possible reasons:

Maybe it's time to turn this thread around: are there albums you rated high, but didn't include in your charts? And for what reason? Personally, my chart is a dull copy of my ratings, but I'm interested in your criteria.

OK Computer. No idea why I've left it out, but it's just a gut feeling which I couldn't possibly explain.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 16:27:55 GMT

Author: Anonymous

Posted: 09/27/2013 20:27

antiloopje wrote:

If you want to know my way of rating a chart, I don't go by favoritsm but rather by actually judging the production, lyrics, how well it fits the genre(s), diversity, vocals, rhythms, instruments, everything. I don't write it on a sheet of paper, but in the end it all comes down to what place I feel the album best deserves over a certain album and under another. I try not to let my personal interests get in the way to include me genres into the chart, and to explore more genres such as shoegaze and rap and electronic. It may sound objective, but in the end it's still how I view the development of the album. So it's not objective and it's still my opinion.

I have a separate chart for actual favorites, but it's very short because it's hard for me to pick favorites. That's why I go by judgement. Also because I want to better myself as a critic.

I have quite a few of the top 40 on BEA in my chart, but it all just adds up to what we like/deem worthy.

Re: Rating vs Ranking discrepancy

antiloopje — Fri, 27 Sep 2013 16:22:24 GMT

Author: antiloopje

Posted: 09/27/2013 20:22

Sorry for the double post, but I've just thougth of some more possible reasons:

* the overall (ranking) chart does take "outside opinions" into account (from all the different magazines), but these people don't rate the albums => higher rankings for the "critically acclaimed" albums

* About those 160 people (1): I assumed all charts give points. This isn't true for custom charts (although these charts still count as a chart)

* (2) How stupid of me, saying "those 160 people won't matter". They DO matter, for the Jazz albums for example, who have only little ratings, those (average(actually ranging between 300 and 28 )) 160 people make up to 35% of all raters. That's 35% of the raters that give 0 points. To be honest, this explains a lot: both albums with high & with low scores have a relatively high percentage of people that don't include the album in their charts.

*

tekin wrote:

Simple: you have low ratings which decrease the overall rating but low rankings increase the overall points.

That's very important indeed, I feel stupid now.

Maybe it's time to turn this thread around: are there albums you rated high, but didn't include in your charts? And for what reason? Personally, my chart is a dull copy of my ratings, but I'm interested in your criteria.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 16:14:18 GMT

Author: Anonymous

Posted: 09/27/2013 20:14

Simple: you have low ratings which decrease the overall rating but low rankings increase the overall points.

Re: Rating vs Ranking discrepancy

antiloopje — Fri, 27 Sep 2013 15:52:04 GMT

Author: antiloopje

Posted: 09/27/2013 19:52

Quote:

I thought the reason that there may be discrepancies between an album's rating and its rank on the overall chart is simply that the overall chart is calculated based on an album's frequency/positions on critic and user charts, and not based on its ratings. So, even if an album consistently gets a 95/100 rating, for example, but is not on anyone's charts, it wouldn't be on the overall BEA chart. I think it's a pretty simple explanation. Am I wrong in this?

Naturally that's true, but it doesn't explain why people rate it high, but don't include it in their charts.
(As meezus said, the point system seems to be well designed, I don't care for the maths behind that)
I'm more interested in the albums that rise/fall a lot in the rating/ranking stuff. For example, I find it very interesting that some of you guys don't include jazz albums in your charts.

But, that can't be the explanation for everything: on average, an album has 160 more ratings than charts. There is no correlation at all between this difference and the score of an album (see below). (both Kind of Blue & Nevermind have >160 while in the court of the king has <160 ). For most albums, these 160 people probably won't make the difference.

So back to the type of albums that change positons heavily: ... I can't really find a line in it. Jazz rises & so do lesser known albums by big bands (animals, mtv unplugged in NY) Some critically high acclaimed albums drop (nevermind, the wall, london calling ...). That's as far as I get. It appears as if some people like the first type, but don't dare to put them in their chart. And, they don't really like the second group, but feel like they must include it in their charts. However, that's not really the impression I get from the BEA charts. You guys seem to follow your own opinion, not the stupid rules described above.

FYI: I've made a bigger list of albums in the rating top 51, sorted by a score that includes how high they are on the ranking chart & how big the difference between the ranking & rating is.

Rising Albums (better Rating)

A love supreme - John coltrane 6723
In the court of the Crimson King - KC 6240
Bringing it all back home - Bob Dylan 5664
Kind of Blue - Miles Davis 5425
MTV Unplugged in NY - Nirvana 5050
After the gold rush - Neil Young 4375
Hunky Dory - David Bowie 3920
Pink Moon - Nick Cave 3818
Sticky Fingers - The R Stones 3527
The Black Saint & the sinner lady 3000
animals...
marquee moon ....
blood on the tracks...
closer (joy division) ....
Remain in the light 2500

Falling Albums (better ranked in the charts)

Nevermind - Nirvana -4656
Is this it - The Strokes -4152
Stone Roses - Stone Roses -3400
The wall - Pink Floyd -3382
London Calling - The Clash -2835
Pet Sounds -1500
Kid A...
In the aeroplane over the sea ...
exile on main street...
The White Beatles ...
The Queen's Dead (Smiths)...
Loveless (my bloody valentine) -1000

formula : (200-Ranking)*(Rating-Ranking)

Re: Rating vs Ranking discrepancy

meezus — Fri, 27 Sep 2013 15:00:00 GMT

Author: meezus

Posted: 09/27/2013 19:00

JMan wrote:

Of course, the highest rating an album can get with 1 vote (thanks to the way ratings are calculated) varies from 78-80 depending on what the average site rating is at the time. The highest an album can get with 50 votes is around 85. Really, a large number of votes is required to get past 90, and even then not all of the user ratings will be over 80. I don't know how the formula works entirely, but I think even with a few lower ratings it is possible for an album to one day get a perfect 100. It may take thousands of ratings, though.

Yeah, I admit I've never really understood (and still don't) why an album or chart's average rating always seems to skew so much lower than its individual ratings. But this site seems to be so logical and well-built, so I just trust it and try not to think too much about it. 😜

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 14:49:41 GMT

Author: Anonymous

Posted: 09/27/2013 18:49

meezus wrote:

Re: Rating vs Ranking discrepancy

meezus — Fri, 27 Sep 2013 14:43:01 GMT

Author: meezus

Posted: 09/27/2013 18:43

I thought the reason that there may be discrepancies between an album's rating and its rank on the overall chart is simply that the overall chart is calculated based on an album's frequency/positions on critic and user charts, and not based on its ratings. So, even if an album consistently gets a 95/100 rating, for example, but is not on anyone's charts, it wouldn't be on the overall BEA chart. I think it's a pretty simple explanation. Am I wrong in this?

Edit: Nevermind... I thought the OP was asking a more technical question. I re-read it and realized it's more theoretical.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 14:33:18 GMT

Author: Anonymous

Posted: 09/27/2013 18:33

Let's get back on topic. You see, most of the albums here on the site are not based on what the average person buys, but rather what music buffs like. Nirvana is a combination of both, selling more than 20,000,000 albums and being a top 10 album on just about every website. But the CRITICS love it, so people try out these albums that critics seem to love. But there's always a difference of opinion, so you never know how well recieved an album is until you calculate the average ratings.

Most of the highest selling albums are acclaimed, but not necessarily in top 100 charts, such as Bat OPut of Hell. 40,000,000 copies and yet the album is usually in the top 300 or 400. Cher's Believe was not well recieved but it sold 20,000,000 because of the hits, especially the title track, and Cher's large following. Shania Twain has the highest selling country album of all-time, really for that one hit called "Man I Feel Like a Woman." It got only decent ratings.

Anyway, I don't think I added much to this topic, but things vary. Sales ≠ Popularity ≠ Rating.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 14:27:07 GMT

Author: Anonymous

Posted: 09/27/2013 18:27

Dingerbell wrote:

I, personally, always felt it was very difficult to compare them to rock albums.

I find it difficult to compare just about any two great albums. 😜

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 14:25:32 GMT

Author: Anonymous

Posted: 09/27/2013 18:25

Dingerbell wrote:

I, personally, always felt it was very difficult to compare them to rock albums. It's, for me, like comparing a dessert to a main course when thinking of your favourite food. It's just too difficult.

It is, but if you don't include Jazz on your chart you're killing it on this site. Every little helps.

Re: Rating vs Ranking discrepancy

Dingerbell — Fri, 27 Sep 2013 14:19:50 GMT

Author: Dingerbell

Posted: 09/27/2013 18:19

JMan wrote:

Also, excluding jazz is a weird rule for charts.

I, personally, always felt it was very difficult to compare them to rock albums. It's, for me, like comparing a dessert to a main course when thinking of your favourite food. It's just too difficult.

Re: Rating vs Ranking discrepancy

Anonymous — Fri, 27 Sep 2013 14:03:26 GMT

Author: Anonymous

Posted: 09/27/2013 18:03

Dingerbell wrote:

Some people exclude jazz albums from their charts. I did- although it's recently been changed.

The albums that tend to be the highest positions on the main chart are those which have the most listeners- and therefore, more people hear them and so, they are more likely to have them on their chart.

This means the overall chart is more popularity than quality, as it rewards those albums with the most listeners.

Exactly, a popularity contest. Also, excluding jazz is a weird rule for charts.

Re: Rating vs Ranking discrepancy

Dingerbell — Fri, 27 Sep 2013 13:58:08 GMT

Author: Dingerbell

Posted: 09/27/2013 17:58

Some people exclude jazz albums from their charts. I did- although it's recently been changed.

The albums that tend to be the highest positions on the main chart are those which have the most listeners- and therefore, more people hear them and so, they are more likely to have them on their chart.

This means the overall chart is more popularity than quality, as it rewards those albums with the most listeners.