Done, also has NWOBHM been added.

I don't mean to irritate you, but is it possible to split up the death metal chart into brutal death metal, melodic death metal, technical death metal and normal death metal

Would make it a lot more interesting, since Death Metal is an extremely wide genre.

But I started from 2000 on, to prop up my point.
Here's the complete thing. I stopped at 2005, for the reason I mentioned.
Not as much to say about it as the other genres but as it was asked for...

The MA search is unable to filter out "normal" death metal. Besides, many bands just under the "death metal" tag are more orientated towards thrash, grind, brutal death, etc.

About the decline in 2007 - is it possible that not all the releases out that year have been added? How many releases from, say 2006, were added 1 or 2 years later?

Yes.
http://metal-archives.com/board/viewtop ... 66#1052561

Your graph still has a problem with the year, oneyoudontknow. The last one is 2008 not 2007. I'd advice to make it stop at 2005 anyway. People will keep wondering about this drop-off.

Huh??? Check your statistics. The rise in numbers over the years is insignificant compared to the overall progression of the scene. From the current point of view there is a drop of 10-20% possible and the rise you refer to is below 1%; if you distribute the new added releases over the total amount of genres.

Even if I would cut of the 2008 year, there is a saturation visible! All graphs decline in numbers as well as in total growth and the trend will be reflected at the end of the year a falling graph; it begins in 2002 - 2004; depending on the genre.

Look at death metal for instance:
2007: 4526
2008: 3053

An enormous amount of releases have to hit the street in order to compensate the loss. Even if you take the 5800 bands without releases into consideration, then it is still the same: distributed over twenty years and at least eight genres.

Hum, no. It's more than 1%. Just look at my other post. The graph speaks for itself.
2006 releases rose 8,6% from 2007 to 2008.
2005 releases rose 5,3% from 2007 to 2008.
Over course in the meantime older releases also increased, but not as much.
I didn't question the saturation. But that huge decrease at the end has no significance whatsoever.

here is an updated version (data taken from today):



edit: genre names added.

edit 2:

these genres are now dealt with:
Black Metal, Death Metal, Doom, Thrash, NWOBHM, Heavy, Power, Speed, Gothic

What about Folk Metal?
That one might not have "peaked" yet.

there is no peak in 'folk'.
The problem is that this genre is very vague and a lot of distortion comes from the other ones; black, death etc.

How are you going to say that there were a thousand more thrash metal releases in 2005 than there were in 1989?

Are you sure this trend isn't just reflective of more small-time releases being noted on MA because of the internet?

the data can be discussed of course, but over the last ten years there has been a clear trend; due to the rise of the internet; which seems turn around now.

Whether the albums were released by short-living bands or not is something you can only find out be looking at each release and band of a year in question.

Why is this still alive? Has no one read Fanfarigoule's analysis? Can no one figure out that these charts are based on bad, premature data?

The data has been updated.
See here:
http://www.metal-archives.com/board/vie ... 05#1082205

and not much since I started this thread has changed... the same trend is visible. The data of the last five years has been changed.

So I guess you haven't read this? The analysis which shows you need to wait at least 2-3 years for more of the releases to be added before we start talking about a "drop off" in 2008?

Quality over Quantity.


1986:
1503 Releases (all genres, all formats...ep, demo, single, video, etc)
of those: 443 full-length albums

2006:
14005 Releases
4323 full-length albums


so in the span of 20 years, we have almost 10x the amount of stuff available to us. So even if there's a decline in metal, that doesn't necessarily mean a bad thing, sure not all of those 443 albums from 1986 are amazing albums, but in comparison a far greater percentage of albums were worth checking out compared to today's standards.

i don't see the decline as any sort of bad thing, especially since the decline does not seem incredibly drastic when put in comparison to how one of the biggest years in metal's golden age and the number of releases back then.

and I talked about the likeliness to see a saturation, see also the amount of bands added in recent months (a decline of 40-50%) in terms of the releases. Or do you see thousands of them appear magically out of nowhere or that bands keep them hidden from sight or huge metal labels appear from a place no one has known before? The MA is covering the scene pretty well and what Fanfarigoule pointed towards in his remark correlates with the downtimes this site had, as well as the amount in which this site is known in the scene; the former has decreased and the latter increased; result: more releases added over a certain period.

The amount of albums thrown out by labels are added often once they are out and even those bands that have their sites incompleted, do not have a graven impact on the overall tendency that is shown in the graphs.

In 2008 (and not even to the end of 2008) we saw over 1,000 2006 releases added, and almost 700 2005 releases added. I think it's pretty safe to say that there are thousands of 2008 releases your graph does not take into account, not to mention a significant amount of 2007 releases.


It's difference between a slight decline and a radical drop, at least. That's a pretty significant difference. I'll say it again, you're working with bad, premature data. Come back in, say, 2011 when more of the data has been collected.

Would it be too depressing to show Groove, Deathcore, and Metalcore on there?

wight_ghoul, come back when you have understood my last reply; which you seem not to have read, neither grasped the content.

Crick, I did this on Deathcore already, because I was curious whether the rise of the dm graph over the bm one was due to the amount of released deathcore albums; currently this is not the case. Yet this genre in particular is a borderline one, it might be that more than expected might be added.

I don't see how any of the vague assumptions you offered counter the data showing what Fanfarigoule called the "transient regime." Again, a year ago you could have made the same graphs and been talking about a decline in 2006 and a huge drop-off in 2007. But a year later these things were shown not to exist. You can assume that this year will be different for whatever theorized reasons, but that's simply not a prudent way to go about analyzing data of this nature. Keep talking about "peaks" all you want, I'll come back and join you in a few years when we have more data.

and what would your argument be, hm? Your reasoning is solely based on a short term analysis, that predicts the evolution would progress in the same way as before; pure linear thinking and without taking any outside influences like downtimes or correlations of releases added by users into consideration; very convincing; your dismissal of my points towards this aspect shows that you not thought about this for one second. Further is the only argument you bring forth contradiction without every having looked on the graphes.

on the topic:
The idea of this thread was to state the question whether there was a peak in the metal scene and there was; depending on the genre: dm and thrash in 2006, bm in 2005; heavy metal a little bit earlier; some other later; your arguments are a troll attempt which distract from the topic; fish is out, sorry... but it would be so fitting... only bones are left: <*-¦-¦-¦-¦-¦-><

By using your crude way of reasoning it should be as well as to discuss these years (2005-2006) now... so, you clearly see a peak. The decline in the dm scene starts in 2002, nearly keeps this way in 2003 and declines thereafter.

2001: 2234
2002: 2869
2003: 3397
2004: 4037
2005: 4453
2006: 4671
2007: 4563
2008: 3417 (year not complete)

01-02: +28,42
02-03: +18,4
03-04: +18,84
04-05: +10,3
05-06: +04,89
06-07: -02,89
07-08: -25,11 (so far; year not complete)

but perhaps the decline after 2006 can also explained by you with the graph by Fanfarigoule.

I don't know if this has been mentioned yet, but I'm pretty sure the number of 2008 releases will go up a lot in the next few years. This is because not all 2008 releases have been added to the database already, but that will grow in the coming years. So actually that's not a correct way to look at things, since the past years have had more time to have more released added to them. My point would be that the only reason there is a dip in 2008, is because of not all releases have been added yet, and not because there's actually a dip.

Have you? Please, explain to me how downtimes or "correlations of releases added by users"(?) will case the significant trend identified by Fanfarigoule to completely disappear in less than twelve months.


See, you aren't even taking the time to understand my "crude way of reasoning." Look at Fanfarigoule's observations. They pertain to this year, and show that 2006 releases are still being added at a rate significantly higher than releases for previous years. Wait six months, wait a year; when this rate becomes sufficiently similar to that for previous release years, then we can start talking about 2006 with relative certainty.

there is this thread which I created on the issue of bands without not all of their releases added. Currently there are slightly above 5700 bands that would apply to this rule.

see here:
http://www.metal-archives.com/board/vie ... hp?t=43308

I:
In case of two releases missing on all sites:
11400 albums -> 11 genres (those are listed in the genre field) -> 30 years (the span you have to take into consideration, especially as there are a lot of old and obscure bands whose albums still need to be added) = around 35 per genre and year.

II:
3 releases missing:
17100 albums = around 50 per genre and year

III:
4 releases missing:
22800 albums = around 70 per genre and year

to make it simple. From what I have seen on the sites, the reality is closer to the first scenario. The exceptions are the grindcore bands, but they have not been covered in my graphs so far.



Less downtimes -> more users (-> which attracts even more users)
Less downtimes -> more releases added (by more users; from more countries, more interests and therefore also a larger group of users who search for albums; we have here some folks that work themselves through the MySpace sites.)

as the downtimes decreased over the time, more users have created an account, submitted bands, have added releases and have written reviews. etc.

so difficult? look at the early days of this site.

The graph of Franfarigoule is just a reflection of this evolution of the site and not many releases have been added later, because so many bands appeard magically out of the Neverland.

Do you not understand? 2006 releases are still, this year, in 2008, being added at a rate significantly higher than previous years. This means that the site still hasn't caught up with the music that was released in 2006. Do you have data to show that the rate of addition for releases in later years is significantly lower? That's the only way I can see your post-2005 (hell, the 2005 rate is a bit higher so we could even say post-2004) being in any way reliable.

And what is your breakdown of unreleased albums supposed to show? Surely you realize that your calculations are massively flawed?

I do like the graphs, though the inevitable retrospective aspect of the list likely skews the data. The past is always constantly being updated.

I don't know if this has been mentioned, but have you considered (if at all possible) to compile data for the number of bands formed in each year? In that respect, the data would be more representative of the number of projects active in a year, since you're not just quantifying releases but quantifying bands themselves.

I would hypothesis, if you did go through with the above suggestion, that you would find that the number of bands formed in each year is increasing substantially, partly due to the internet and partly due to the popularization of certain 'fringe' styles.

Ideally, the most interesting graph to me is nearly impossible with MA right now. That would be to make a graph for each genre, and then outline the formation and the end dates of each band within that genre, so that each band would form its own line(s) of existence (2001 - 2004, 2006-2008). With enough bands, a line of best fit could be computed. But to do that would require a complicated computer algorithm and MA's database containing end date information.

Well you can add a graph which shows the new bands that have been added to the genre in each year.

The problems are:
- often you do not have the date of their founding; especially obscure ones
- you have no search function that deals with this issue; you would have to do this by hand ... over 64000 bands... good luck.

the idea is interesting though. You could create, perhaps even per country, life-span indices and compare them:

German black metal bands last longer than french black metal bands.

Certainly an interesting topic.

but without access to the database a hopeless attempt.


There is a certain phenomenon going on in the metalcore scene: a very short existence. There are bands that did not last even a year; this had been one reason to start on the 'missing release thing'.