(8+7+7+8)/4=8,4

Written by PēterisP on 12.07.2022 at 08:52

(8+7+7+8)/4=8,4

Your calculation is flawed, because the only valid formula looks like this:

musical taste ≠ maths

In all seriousness, the fact that the mathematical mean of the rating breakdown doesn't have to match the overall score has been explained so many times that it's kind of tiresome.
What's missing in the individual ratings is the reviewer's personal opinion of how much they like the album. Everyone weights factors like production or originality differently, so the mathematical approach is simply nonsensical. Instead, it makes much more sense to simply read the review, because only there you can find out what the overall rating is really based on.

Written by Starvynth on 12.07.2022 at 10:46

Written by PēterisP on 12.07.2022 at 08:52

(8+7+7+8)/4=8,4

Your calculation is flawed, because the only valid formula looks like this:

musical taste ≠ maths

In all seriousness, the fact that the mathematical mean of the rating breakdown doesn't have to match the overall score has been explained so many times that it's kind of tiresome.
What's missing in the individual ratings is the reviewer's personal opinion of how much they like the album. Everyone weights factors like production or originality differently, so the mathematical approach is simply nonsensical. Instead, it makes much more sense to simply read the review, because only there you can find out what the overall rating is really based on.

The mathematical approach is not nonsensical, but of course it has to account for the weights on each of the four categories. In Peteris' calculation above equal weights were assumed.That is an assumption, which though might differ from the weights the reviewer had in mind, as you pointed out. But if weights are accounted for, the mathematical approach makes perfect sense to me.
That being said, I am not a big fan of the four breakdown categories in MS reviews in general. They should be omitted from all reviews, if you ask me.

Edit: Only now I realize that my post above was partly misleading. I said in Peteris' calculation equal weights were assumed. I did not mean to refer to the calculation that he wrote down. There obviously non-equal weights were assumed. What I meant was that Peteris probably assumed equal weights and thus stumbled over the overall rating attached to this review.

Written by Starvynth on 12.07.2022 at 10:46

Written by PēterisP on 12.07.2022 at 08:52

(8+7+7+8)/4=8,4

Your calculation is flawed, because the only valid formula looks like this:

musical taste ≠ maths

In all seriousness, the fact that the mathematical mean of the rating breakdown doesn't have to match the overall score has been explained so many times that it's kind of tiresome.
What's missing in the individual ratings is the reviewer's personal opinion of how much they like the album. Everyone weights factors like production or originality differently, so the mathematical approach is simply nonsensical. Instead, it makes much more sense to simply read the review, because only there you can find out what the overall rating is really based on.

I would not comment even if you had:
Performance: 10
Songwriting: 1
Originality: 1
Production: 1
and still would get 8,4.

But if none of the grades are above 8 then 8,4 does not make any sense to me.

Written by PēterisP on 12.07.2022 at 18:45

But if none of the grades are above 8 then 8,4 does not make any sense to me.

What if the categories with the ratings presented are incomplete from the point of view of the reviewer. Maybe the reviewer puts some weight greater than zero on some other categories not presented, say e.g. atmosphere.

Written by Redel on 12.07.2022 at 14:20

The mathematical approach is not nonsensical, but of course it has to account for the weights on each of the four categories. In Peteris' calculation above equal weights were assumed.That is an assumption, which though might differ from the weights the reviewer had in mind, as you pointed out. But if weights are accounted for, the mathematical approach makes perfect sense to me.
That being said, I am not a big fan of the four breakdown categories in MS reviews in general. They should be omitted from all reviews, if you ask me.

Dude, I love stats and numbers and I believe in maths, but probability calculation is a part of it. And the probability that there are two people on Metal Storm with an exactly identical personal weighting system of these more or less randomly chosen four criteria, each making up exactly 25%, is zero. Even the number of criteria is random, you could just as well include cover artwork and lyrics. To assume that the average of the rating breakdown must nevertheless correspond exactly to the overall score is therefore pretty naive, methinks.

By the way, the rating breakdown has never bothered me. It is what it is: additional information that is never essential, but often useful. For example, I am sure that it can be of use for an audiophile music lover to be able to filter out albums with abysmal sound right away, while the average user might be keen to know why one particular album can be fun despite of a poor production.

Written by PēterisP on 12.07.2022 at 18:45


I would not comment even if you had:
Performance: 10
Songwriting: 1
Originality: 1
Production: 1
and still would get 8,4.

But if none of the grades are above 8 then 8,4 does not make any sense to me.

I completely understand your point, but I personally wouldn't read too much into the overall rating, there are a few exceptions where I would give an album an overall high rating despite having all low scores on the quality breakdown, simply because of how much of an impact an album has had on me, or as the above comment says, there may be a special kind of atmosphere that I'm really drawn too, which would also result in me giving a high overall score.

Written by Starvynth on 12.07.2022 at 19:42

Even the number of criteria is random, you could just as well include cover artwork and lyrics. To assume that the average of the rating breakdown must nevertheless correspond exactly to the overall score is therefore pretty naive, methinks.

Totally agree. But still it is possible that there are reviewers out there who are able to pin down exactly which categories are relevant for them in a specific review and which weights they put on each of them. In this case I can see the mathematical approach of a weighted average work perfectly.
But since categories are largely random if reviewers cannot be forced to reveal all relevant ones precisely, plus to put an exact weight on each, I suggest to omit all categories in general, and just allow for an overall rating in reviews.

Written by PēterisP on 12.07.2022 at 18:45

I would not comment even if you had:
Performance: 10
Songwriting: 1
Originality: 1
Production: 1
and still would get 8,4.

But if none of the grades are above 8 then 8,4 does not make any sense to me.

And what if the two 8s the reviewer put on performance and production are actually an 8.4 each, since first digits are omitted, and songwriting as well as originality have each a weight of zero in the overall score?

Yeah guys... Start rating NA to avoid this kind of discussion.
Plus, maybe people are gonna actually resort to reading the review to monitorate understand the reviewer's opinion

Written by Karlabos on 14.07.2022 at 01:43

Yeah guys... Start rating NA to avoid this kind of discussion.

Way ahead of you, bud

Written by Karlabos on 14.07.2022 at 01:43

Yeah guys... Start rating NA to avoid this kind of discussion.
Plus, maybe people are gonna actually resort to reading the review to monitorate understand the reviewer's opinion

Read a review? Like... like... a really long tweet?