professional opinion

August 18, 2009

There’s a post up over at Wowenomics which speculates about the most profitable crafting professions and how that might have changed between patches. There is no armoury data that helps us relate professions to profits, but the post prompted me to have another look at the distribution of professions in general. This is the current state of play:

Patch 3.2 Patch 3.0.8
Mining 20% 20%
Herbalism 12% 13%
Enchanting 11% 10%
Jewelcrafting 9% 5%
Skinning 9% 13%
Tailoring 8% 9%
Alchemy 8% 8%
Blacksmithing 7% 7%
Engineering 6% 6%
Leatherworking 6% 7%
Inscription 4% 3%

There’s a lot less change in there than I was expecting, give that professions get worked over with the nerf bat just as classes do. I’ve marked the two most dramatic shifts: jewelcrafting and skinning. Jewelcrafting may indeed genuinely have been buffed to the point where people are more enthusiastic for it. The decline in skinning I’m not so sure about.

There has been a significant decline in some of the leather-wearing classes such as rogues and hunters; perhaps that has affected the crafting demand for hide. Certainly the classes that are flavour of the month want heavy metal and not the skins of small furry animals.

working class orc

July 24, 2009

Thanks to blog reader Grant, who asked for a report on races and professions. And voilĂ 

The numbers are percentages of the total race counts for toons that have at least one profession. So, to take the first number, 4.9% of all Blood Elves who have at least one profession have Alchemy as one of those professions.

Alchemy Blacksmithing Enchanting Engineering Herbalism
Blood Elf 4.9 5.9 11.2 7.9 10.8
Draenei 6 5.4 6.2 6.4 12.3
Dwarf 4.1 10.7 5.7 13.7 8.2
Gnome 5.5 3.3 10.4 20.9 10.1
Human 6.8 10.3 11.6 7.5 12.1
Night Elf 9.1 3.7 6.9 8 15.4
Orc 5.1 9.2 5.4 11.8 10.3
Tauren 9 6.4 4.8 5.7 16.8
Troll 7.7 3.5 7.5 11.3 13.2
Undead 7.9 3.8 13 10.6 13.6

Inscription Jewelcrafting Leatherworking Mining Skinning Tailoring
Blood Elf 4 8.2 4.7 23.1 10.8 8.6
Draenei 4.1 14 4.6 26.4 9.9 4.8
Dwarf 2.3 2.6 7.2 28.4 13.2 3.8
Gnome 3 2.7 2.3 22.7 7 12
Human 3.7 3.4 2.9 21.8 8.4 11.4
Night Elf 3.2 2.6 13.3 14.7 19.5 3.6
Orc 3 3.8 7.6 26.8 14.3 2.8
Tauren 3.7 3.3 12.6 17.7 18.8 1.2
Troll 3.1 3.1 9.3 19.1 15.6 6.5
Undead 3.3 3.6 3.7 17.1 9.2 14.2

Race, class and profession aren’t independent variables – class influence profession because of the different types of gear that can be worn, and race influences class because not every race can take up every class. Differences in playstyle (for example selecting a race for PvP) come into it as well. But there is more variation there than I was expecting, and I’m not sure I understand everything in those tables.

get a job!

January 16, 2009

well… two jobs in fact…

I’m struggling a bit with my professions reports. The database queries are correct, but the tables look odd because nothing adds up to 100%. The problem turns out to be a whole bunch of slackers who have fewer than two professions. I guess they’re all playing the AH instead… Or they’ve gone fishin’.

Out of a sample of 40,000 warlocks, for example, about 1000 have no primary professions and nearly 3000 have only one.

Tobold has a post up about a shortage of enchanters on his server. By coincidence, I’m data crunching professions at the moment, as I mentioned in my last post. It’s a simple matter to write a query that shows the overall popularity of each profession.

And voilĂ 

Mining 20%
Skinning 13%
Herbalism 13%
Enchanting 10%
Tailoring 9%
Alchemy 8%
Leatherworking 7%
Blacksmithing 7%
Engineering 6%
Jewelcrafting 5%
Inscription 3%

It will be interesting to observe the changes in this distribution over time. There is no doubt that professions are enlisted in the nerf wars, just like classes are. And they’re vulnerable to other economic forces too since decisions about what professions to adopt are not cost free even in a virtual economy.

Update: If you look carefully at this table and the one in the post below, you will see that the percentages don’t seem to match. This table has enchanting at 10% whereas the other table has the percentage popularity for the Tailoring/Enchanting combo alone at 9%. The answer is that a small but significant number of toons only have one profession (!!) and aren’t picked up by the combo query. Sigh. Some MMOs do have ’em…

invisible hand

January 8, 2009

I’m working on a new set of reports to cover some of the more general decisions that players need to make. Picking professions is a classic example. Professions can be played for fun and profit so it would be useful to know the most popular professions for each class; more useful still to know which combinations of professions are the most popular, since picking two that go together is half the battle, especially at lower levels when cash generation is usually the name of the game.

As people who work with database queries know, there are a couple of tricks to generating combinations from a relational database – we need a query which knows that “skinning/mining” and “mining/skinning” are not two different combinations. But once that problem is solved, it’s all downhill from there.

We end up with tables that look like this. This is the distribution of profession combinations across the total character base:

11% Skinning/Leatherworking
11% Herbalism/Alchemy
10% Mining/Blacksmithing
9% Tailoring/Enchanting
8% Mining/Engineering
6% Mining/Jewelcrafting
4% Skinning/Mining
3% Inscription/Herbalism
2% Skinning/Herbalism
2% Mining/Herbalism
2% Mining/Enchanting
1% Tailoring/Skinning
1% Tailoring/Herbalism
1% Skinning/Enchanting
1% Herbalism/Enchanting
1% Tailoring/Mining
1% Engineering/Enchanting
1% Jewelcrafting/Enchanting
1% Tailoring/Alchemy
1% Inscription/Enchanting

Combinations that have a popularity below 1% are not shown.

Useful information, but there is nothing too surprising in the table. Hopefully when the data is broken down across classes, some more interesting patterns will emerge.

(BTW I haven’t forgotten that the death knight stats pages aren’t up yet. There have been a couple of technical problems with my new stats queries… Should be fixed in the next day or so.)