Hey everyone, you all probably already seen the advanced class distribution graph shown at the guild summit (the number markings were not too accurate and has being corrected for use in the inforgraphic below). Alratan @ sithwarrior.com suggested to me that it might be nice to have a relative advanced class ratio based on that – in an infographic format. So here is the result of that idea!

Relative ratio is calculated by dividing that class’s % by the % of the least played class (i.e. vanguard or scoundrel). So since 10% of the overall SWTOR population plays Sorcercer, that would be 10%/3% = 3.33.

## 19 replies on “SWTOR overall advanced class distribution”

THIS is but another example of why Dulfy is the best SWTOR site period. Humorously enough, sithwarrior is my #2 without a doubt.

hehe ty! 🙂

If you look at the source then Scoundrel is actually a bit lower then Vanguard. So Scoundrel should probably be last 🙂

good catch, edited! 🙂

Nitpick: still says Vangaurd as lowest class in relative ratio header.

Awesome work! 🙂

Great work Dulfy! But there is a typo in the Sorcerer name 😉

hehe I will get that edited soon 🙂 thanks for noticing it ! 🙂

[…] Klasse im Spiel so schlimm kann es doch nicht sein, wie selten man auf diese Klasse trifft (Spass) SWTOR overall advanced class distributionDulfy | Dulfy Sabo hat genug andere Probleme nur weil der Opener manchmal derbe burstet ist das finde ich noch […]

[…] http://dulfy.net/2012/03/15/swtor-overall-advanced-class-distribution […]

[…] over at Dulfy.net took data from an advanced class distribution graph shown at the Guild Summit and converted it into […]

Since you stated that the smaller number to right is the population playing that class in percentage, i would like to point out that:

10+

9+

9+

8.5+

7.5+

7.5+

6+

6+

5.5+

5.5+

5+

4.5+

4.2+

4+

3+

3=

98.2

That means there is a 17th class, undiscovered so far, that has been played by 1.8% of the population.

hehe, it is probably just rounding error 🙂

/facepalm

since there are numbers with .5 and .2, that means a 1 digit precision, that have not been rounded it means that the rounding must have been made for 0.01-0.09 digits that multiplied for 16 classes makes a total rounding of 0.16-1.44, assuming every rounding in excess and for every class.

It would still imply a deficit of 0.44. Since for the relative ratio a 2 digits precision has been used, it makes sense to think that the same precision should have been used for both the sums, making the mistake even bigger.

But the problem is that if this is a statistic you cannot have a sum inferior to 100% because it would mean that you have lost part of your sample during the measurement, just like if during an interview at some point some people refused to answer to your questions. In this case there is only one question so those cases would be completely discarded and the result would be again a 100%.

Those numbers don’t make any sense and i am sorry for the person who arranged them, that should go back to school, and for the “ingenuity” of the person that posted, since she decided to blindly accept them without checking.

Does this take into account characters *before* advanced classes?

I dont see Sith Inquisitor or Sith Warrior on this list, and I can attest I still have 4 of each spread out on servers that are still on Korriban.

My mistake… nm

There is a typo on Operatives (4.5%). The figures are rounded to 0.5, not 0.1 as you state, and the rounding error is within margins.

The thing you don’t get is that when you “try” to use statistics to prove things the first thing you have to mention is the dimension of your population or at least of the sample you are using with a nice number on top or under the table.

This is missing entirely.

When you have an estimation of how big your population is then you decide the kind of rounding you are going to use otherwise people who read it cannot evaluate how precise is your rounding.

Moreover rounding isn’t necessary for small groups, like your obviously is since it is only made of those guilds that joined the survey, beacuse, when you analyze small samples’ distributions, low represented classes can be more affected by rounding than higher ones. Once again this cannot be said since it is not indicated the exact number of people surveyed.

Third, margins? what are these margins if you so far have not given any way to analyze your data?

If at the beginning i was thinking about an error i guess it is quite obvious that those data are severely artifact, most probably in an intentional way.

Next time try not to use statistics, when you still need to learn basic maths.

I’ve heard statistics that can’t possibly be measured to .1 precision and that are about video games are a big deal and require paragraphs talking down to people on the internet. Well done, I’m sure you’re very proud.

[…] are among the lowest played advanced classes (according to an albeit slightly outdated census here). So Bioware decides to make the majority of the rewards only usable by 25% of the population, […]