Official: DTF's song ratings 2009 RESULTS

[BlobVanDam]

Well I think the weighted method is more fair. A one minute song like Through My Words shouldn't affect the overall score as much as a 12+ minute song like Home or Finally Free. Especially when people tend to give the really short songs really bad ratings just because they're short.

[ariich]

But as I said, once you start trying to weight it then you can get too detailed, looking at sections of songs etc. For example everyone who rates SDOIT 10/10 for the full thing, I'm pretty sure almost all of them would NOT rate each section/track 10/10.

[BlobVanDam]

Well if they're rating it 10/10 if they don't think it deserves it, then that's their fault, not the fault of the system. The system is more accurate for an overall album score if it's weighted. I don't see how it's a bad thing at all.

[hefdaddy42]

Well I think the weighted method is more fair. A one minute song like Through My Words shouldn't affect the overall score as much as a 12+ minute song like Home or Finally Free. Especially when people tend to give the really short songs really bad ratings just because they're short.

I don't think it's more fair at all.  I don't rank how much I like songs by how long or short they are.  That's just silly, ariich's right.

[BlobVanDam]

Nobody said you should rank songs based on how long they are. You give a song the score it deserves.
Here's a scenario to explain it- Say you have an album with 2 songs. One of the songs is 70 minutes long, and everyone thinks it deserves 10/10. The other song is one minute long and nobody likes it and give it 0/10. Does that album deserve to get ranked as 5/10 based on an equal weighting score? Of course not. It deserves close to 10/10 overall.
A weighted score makes complete sense, and there is no reason not to do it.

[cygnusx1jg]

As the one that started this, I'd like to point out that yes, the gap between the scores changed, but the order of the albums based on where they finished rather than their score remained the same.  Thanks.

[BlobVanDam]

Now that I think about it, I completely understand where you guys are coming from (ariich, hef), although my preference is still weighted.

[FlashCE]

Take the Time is the best DT song dammit.

[TheVoxyn]

Great job ariich, thanks for compiling everything and working it out so well!

[ariich]

Now that I think about it, I completely understand where you guys are coming from (ariich, hef), although my preference is still weighted.

VICTORY IS OURS!

Obviously it comes down to preference, and I can see why someone likes the slightly more detailed one. I just think that going down that route would require even more detail to even mean that much, and so personally I feel that a simple average is more fair.

I can totally see the point in your argument though, and it's clearly the reason that Octavarium does significantly better with weightings.

[BlobVanDam]

I don't think any more detail is required for the weighted system to mean anything. If this whole rating idea means anything, we have to assume that people's ratings for each song is an accurate representation of how they feel about the song as a whole. Assuming that, I think that the weighted ratings for the albums are pretty useful, especially with Octavarium as you said.
But I understand that you and hef don't think duration should be taken into account, and instead essentially give the album an overall score based on the average ranking of the songs on the album, which also makes sense.
Just a preference of which figure you think more accurately sums up an album. Of course we all know that neither figure will truly represent how people feel about the album anyway, since an album is more than the sum of its parts.

[ariich]

Of course we all know that neither figure will truly represent how people feel about the album anyway, since an album is more than the sum of its parts.

Exactly!

This whole project was for songs, rather than albums.

[BlobVanDam]

We could do another smaller album rating thread. Same idea, but just give the albums an overall rating, then we rank the albums that way. That would eliminate any debates about which method is best, especially since neither of them is truly accurate anyway.

[ariich]

Yeah I was thinking that. Only problem is I can see quite a few people giving 10/10s for almost everything.

[BlobVanDam]

Well the same happened with songs too. And there will be people who give 0s just to be assholes too. It just goes with the territory when you ask an entire forum, I suppose. In theory the law of averages should even it out.

[orcus116]

It's also very interesting that the modern DT albums rate so highly, and Awake is in 6th

When people grow up and start listening to real music they'll understand the mistake they made by not listing the songs on Awake higher.

Orcus from 1 AM, shut the hell up.

[BlobVanDam]

Take that, orcus. You got told.

[EmilSumm]

Glad to see Black Clouds high and Failing Into Infinity low on both. Awake could probably be higher if Space Dye Vest wasn't there, so blame Kevin Moore.

[BlobVanDam]

I blame Kevin Moore every day.

[tri.ad]

Same old, same old.

[ariich]

And there will be people who give 0s just to be assholes too.

Did you just own yourself?

It's also very interesting that the modern DT albums rate so highly, and Awake is in 6th

When people grow up and start listening to real music they'll understand the mistake they made by not listing the songs on Awake higher.

Orcus from 1 AM, shut the hell up.

[BlobVanDam]

And there will be people who give 0s just to be assholes too.

Did you just own yourself?

Hell yeah biznitch.

[GuineaPig]

Looking over the list again, I'd say "Biggest Surprise" definitely goes to Blind Faith at #14.  I really like the song, but I didn't expect it to finish that high.

Also, shocked at how well rated "The Shattered Fortress" is.

[growly]

Fantastic list. Top 20 has 18 of my top 20 songs in it. Good to see all the different data

Biggest surprise for me was This Dying Soul at #55

I rate this song alot higher then this. But I guess it's not as popular as I thought!

[ddtonfire]

This is the trendline without WDaDU. It regresses.

[juice]

Yeah I was thinking that. Only problem is I can see quite a few people giving 10/10s for almost everything.

How about everyone just ranks them 1-10?  Which ever album has the most 1st place nominations gets the highest rank and the most 2nd place nominations gets the 2nd place rank and so on.  What do you think?

[Perpetual Change]

This is the trendline without WDaDU. It regresses.

After Awake, though, it inclines.

[BlobVanDam]

This is the trendline without WDaDU. It regresses.

If that's the gradient of their regression after 17 years, they're doing pretty damn well

[ariich]

Goddammit stop with the regression line! (Nick, look what you've started! ) In both of the graphs posted, the gradient is so low and the value of r^2 is also very low, which any mathematician will tell you shows that there is NO trend.

Yeah I was thinking that. Only problem is I can see quite a few people giving 10/10s for almost everything.

How about everyone just ranks them 1-10?  Which ever album has the most 1st place nominations gets the highest rank and the most 2nd place nominations gets the 2nd place rank and so on.  What do you think?

Yeah I thought about something like that as well, but it has the problem that the distance between albums is not taken into account. For example, for me DT have three 5-star albums. Say the next ranking was 3-stars (it's not, but let's be hypothetical); The difference between 2nd and 3rd place is the same as between 3rd and 4th in the rankings, but in terms of how much I actually like them, the differences should be completely different.

That was why I didn't do this song stuff as a ranking and did it as ratings instead.

[Scard]

That's a fairly accurate representation of my list, actually. Scarred should be higher, but overall I like it!

[ddtonfire]

Goddammit stop with the regression line! (Nick, look what you've started! ) In both of the graphs posted, the gradient is so low and the value of r^2 is also very low, which any mathematician will tell you shows that there is NO trend.

So they're really inconsistent is what you're saying.

[ariich]

Goddammit stop with the regression line! (Nick, look what you've started! ) In both of the graphs posted, the gradient is so low and the value of r^2 is also very low, which any mathematician will tell you shows that there is NO trend.

So they're really inconsistent is what you're saying.

No, all their albums are between ~7 and ~9 out of 10. Which is pretty consistent.

What I'm saying is that there is no upward or downward trend, and that there are fluctuations from album to album because obviously they won't all be identical in quality and popularity.

[Dream Team]

Really? That's sort of the case with me for ANTR, it just seems to go on for too long with no real interesting variation except the 'beautiful agony' section. But then I suppose I'm not really into the metal side of DT quite as much, despite the fact I think Glass Prison is a top fiver.

Exactly this. The last 8 minutes of ANTR has 1 Labrie vocal (the choruse). Give me a break.

[ariich]

Really? That's sort of the case with me for ANTR, it just seems to go on for too long with no real interesting variation except the 'beautiful agony' section. But then I suppose I'm not really into the metal side of DT quite as much, despite the fact I think Glass Prison is a top fiver.

Exactly this. The last 8 minutes of ANTR has 1 Labrie vocal (the choruse). Give me a break.

So? Stream of Conciousness has 11 minutes with zero Labrie vocal.

[tri.ad]

  I'm sure that Dream Team didn't mean it that way.