Progression by the Classes

Sadly, class performance is the sort of data that is out of anyone's scope to accurately catalog unless Anditu (I think it was for ACT guy) snuck in a tracker to collect the log data of eveyone who uses ACT. I would love to, don't get me wrong, but that'd require the setup of every raid, who plays what class in every raid, their player stats / gear, and probably logs from the past 2-3 weeks worth of raiding at least. With that and enough time (looking at months here as a time scale) you could accurately forecast the performance of every class pending no major changes go into the game.

On its own, yes that is the face value of this information. To the right people this is a tool to further their own research, pose questions to be answered, data to compare to other data, and insight into which classes lead to successfully killing various mobs.

You can use it as evidence in a spam-email to a dev saying 'RAGAGAGAGAH! Y U NO LIKE [insert inevitable 'Beastlord' here] ?!' with the hopes of furthering your cause to get an issue with any Beastlord-based classes resolved... You could use it as research for your own guild, as a pointer that you probably want an extra templar around for the hard fights... You could print the charts out and make origami animals I suppose (take photos please!), or even find it interesting to see just how many people are raiding your class.

Honestly and seriously though, you are right: this data alone doesn't answer many questions, but it does offer clues and additional data for research purposes. But what do I know, I'm just a llama!

Excellent work Katanallama!

If you ever do this in the future, you may want to separate out the utility classes from the "dps" classes.

When I do that, the archetype percentages change to:
Utility = 29%
Mage DPS = 15%
Scout DPS = 15%

As always, your best chance of getting a seat on a raid is to be a bard or enchanter.

I think you're wrong about channelers. The channelers make up 2.4% of the kills, but they make up less than that of the total level 100 population. (They're 1.7% of the level 100 population on Permafrost.)

Even beastlords appear to have an easier time getting a spot than shadowknights and paladins. The only tank that appears to have greater representation in raids relative to its representation in the level 100 population is the guardian.

Well, how often each class is invited to participated is directly linked to how they perform. Performance isn't the only factor, but it's the major one, so while you wouldn't get 100% confidence in the correlation, it's pretty close. Availability would be another major factor but performance is a big determining factor in availability too.

You can also loot at the upper echelon of difficulty for a better indicator, as the fights killed by the top end guild are less likely to have low-performers.

I really like this way of looking at things.

I'd take it a step further even and in addition to looking at the total level 100 population, i would look at the "total raiding population", which we could probably baseline by taking the stats from the first mob in Molten Pools, since it's fairly safe to say that "Everyone that raids has killed first mob in Pools". From there you can look at things like "of the total raiding beastlords, how many are used on T2 encounters?" and stuff like that.

Based on your suggestion, I looked at Katanallama's data again. This time I did two comparisons:
Comparison 1 - Teraradus, Crumbling Icon, Perador, Barin/Morin, Soronigus
Comparison 2 - Rhag'Zadune, Protector of Bones, Villandre V'Zher

Comparison 1 vs. 2 (Archetypes)
Fighters = 11.3% vs. 9.0%
Priests = 30.7% vs. 33.0%
Utility = 28.2% vs. 30.6%
Mage DPS = 14.6% vs. 14.2%
Scout DPS = 15.3% vs. 13.2%
It appears that high-end raids take 2 tanks, 8 healers and almost 8 utility per raid.

Comparison 1 vs. 2 (Selected Classes)
Monks = 2.0% vs. 2.0%
Bruisers = 0.7% vs. 0.7%
Every other fighter class was less used in Comparison 2, though berserkers were still more used than monks, and all were used more than bruisers. Looking at overall level 100 population, it looks like crusaders fare very poorly.

Templars = 4.3% vs. 6.6%
Wardens = 5.6% vs. 4.4%
Channelers = 2.4% vs. 2.8%
Level 100 wardens significantly outnumber level 100 templars. And channelers are surprisingly popular.
Necros = 3.4% vs. 2.7%
Warlocks = 3.9% vs. 4.5%
Conjurors fare almost as poorly as necros. Wizards fare almost as well as Warlocks.
Brigands = 4.2% vs. 4.9%
Assassins = 3.6% vs. 2.6%
Beastlords = 1.0% vs. 0.2%

I was a little surprised to see more than 1 brigand per high-end raid. I was more surprised to see less than 1 assassin per high-end raid. The beastlord numbers, while sad, reinforce peoples anecdotal reports.
Coercers = 7.0% vs. 7.0%
Illusionists = 6.3% vs. 7.1%

High-end raids aren't quite taking 2 illusionists and 2 coercers per raid (though they are getting close).
Dirges = 8.7% vs. 8.2%
Templars = 6.2% vs. 8.3%

It appears that high-end raids go with 2 dirges and 2 troubadors by preference. Other raids will take more dirges because dirges significantly outnumber troubadors.

As others have been saying, this is a starting point for further study. It has been my experience that these kind of numbers reflect people's perceptions of how well classes perform in raids, rather than the reality. In many cases, a particularly successful raid guild, or a particularly skilled and/or geared individual can greatly influence these perceptions, even though the success is not due to the class, but other factors.

Someone takes a beastlord to a raid?

As far as channelers, they are quite good, but the number of people playing them well seems to be low overall.

It is hard, if not impossible to draw any real conclusions about class performance from this data. It reflects (a) what is popular to play (b) what is easy to play, resulting in more of that class, (c) availability - Channelers for example are in very short supply.

Add to that perception, which is a big factor - Bruisers reputedly suck, so there are few of them, and few end up in raids. Perhaps they really DO suck, but how much is fact, how much is perception?

As far as Templars outnumbering Inquisitors at the high end, there is too little data to draw much of a conclusion - perhaps only 3 guilds are raiding that zone, and they happen to have a lot of Templars.

Lot's of interesting data, but it raises as many questions as it answers.

The wheel is round.

So basically you posted a graph on which people will cookie cutter their class to fit YOUR needs in order to get a raid invite.

This data is trash, completely and totally worthless and serves no purpose but to put a biased spin "This is HOW is MUST be done" elitist mentality that seems to infect guilds that want to raid end game in easy mode rather than actually work at it.

Just another post of Our way or the Highway

Of course the first 2 named on each zone will be inflated due to bot farmers that run it constantly with their bot raids.

The data is not biased, the observer (you) are. Data is data, it has no opinion. What you do or what you make of it - is the opinion part, and I totally disagree with what you said (my own opinion).

Thank you Katanallama for doing this analysis and the attractive graphs. You've clearly explained what the data is (and is not!) and set the expectations. All this is is a list of which classes are succeeding in raid content. No more no less. It doesn't say what anyone can or can't do, only what is being done. Yes there will always be those few who are drinking the haterade and think that data like this are the reason they don't get invited to do X.

Any player can be exceptional, even in an unbalanced class, and good raid leaders will let anyone try out and judge their performance on the merits rather than going on preconceived notions. If someone wants to make themselves known as a fantastic Beastlord, there's nothing stopping them from doing so except the reality check of the class's current deficiencies.

As for Solo content, all 26 classes are more than fine and have the necessary tools to do well.

While I agree with where you are going, there are cases of classes that are just simply broken. No matter how fantastic a player is running a beastlord, he'll still be easily out parsed and performed by an exceedingly average player on any other dps class.

The data may not be a direct indicator on what classes are broken, but it does provide some additional supporting material to the argument that a few classes need to be addressed.

Well either Rangers are really fun to play that we're willing to put up with having no utility, or we're really good at bribing raid leaders, or Rangers really have been tuned up to be pretty good these days. At least one of those things is right.

To be fair, in my original post I did tell people to make their own observations *shrug*, I actually refrained from making any observations of my own in the original post for the very reason of avoiding any kind of personal agenda. I should have used huge, bold letters on the disclaimers I guess in my original post.

Anyways, I don't mind negative responses too much, gotta expect it time to time!


And as some are finding out, this data only answers one question *evil cackle*. Wasn't intended to answer all the questions, only add to the pool of observable data for research. Give validity to some claims, add data to the investigation, pose questions to solve, spark conversation and so on.


Anyways, here's one of my super secret macro commands for lag:

Code:

/cl_min_sleep_ms 2500

Sorry Feldon

You neglect the time-honoured ritual of ranger sacrifice.

This is of course a lie.

Good raid leaders build a raid with the classes that are currently OP so you can clear 3 or 4 raid zones in 4 hours.

I disagree 100% with this comment. Some raid leaders may do this.. I don't think they are necessarily 'good'. Certainly nowhere near the 'best' by any measure I'd use.

Ive never seen you do anything other then trolling and shout "elitist" at ppl. Stop trolling this thread where kat has obviously put down effort and free time to present these statistics for ppl who might be interested.

Those who can, contribute to the community. Then there are those who don't contribute and just exist to tear others down.

It is an interesting read. There have been a lot of complaints about scouts loosing raid slots to mages, but these numbers don't support that at all. While it could be off due to people opting out it should be "close enough" to give a rough idea. And the rough idea says scouts are getting fair face time.

Now individual classes, that is another story. One I'll leave to others to dissect.