Curse

Beqxter · Nov 07, 2005, 06:01 PM // 18:01

So, I've decided to take up farming as a hobby between gvg matches and while waiting for chapter 2 to come out. I have a series of builds I'm happy with in various popular locations throughout the game, but to make it more interesting I thought I might track these things to determine A) which are the most efficient places to farm and B) does drop quality in fact diminish over repeated runs?

So I've started an excel spreadsheet that tracks the following per run:

Time spent
XP Gained
Gold earned (post selling)
Gold Drops
Purple Drops
Blue Drops
White Drops
Estimated drop resale value to players

But the trouble is I'm not totally sure what to DO with this information. I'm no math whiz, although I am enough of a geek to come up with an excel spreadsheet for this stuff. Can anyone come up with a formula that will ultimately deliver a number rating the effectiveness of a given run, so that they can be compared? I'm not sure XP should be a factor, and I worry that the whole player resale value is rather vague, but I'm sure the amount of time taken should factor in - for example, UW solo runs have the chance of being very profitable, as do SF runs, but they take considerably longer than a simple griffon run.

SantaClause · Nov 07, 2005, 06:45 PM // 18:45

Im no statistics major but I do use it in my job.

The first thing that I would do is apply SPC (Statistical Process Control) to the data you have already collected to make sure that the data represents the true output of the farming run. You can download free excel addins which will will perform this task for you. This will help you see if you happen to have a great run which is not typical and due to a special cause such as quest reward or duplicate green item drop.

Once you have a good representation of the data you can just do a simple calculation based on total gold produced over the amount of time spent. Some additional things not mentioned above that I would include in this calculation:

Subtract UW entry fee from gold produced
subtract salvage kit cost from gold produced (Some areas may not use)
subtract id kit costs from gold produced (Some areas may not use)
add time spent converting items to materials to sell to time spent

Maybe a math major could further help with Chi Square between characters or other hypothsis testing tools.