This is getting so ridicules. Here an excerpt of the current missions list available to me.
I leave this without further comments and just let the facts speak for themselves.
Yes, they do speak for themselves, but you are mistaken if you think those results are saying what you want them to say.
Whether the system is broken, with respect to the random selection of missions from a list of what exists for each grade mission, would require a lot of measurements. Something approaching 1,000 independent measurements would be a good start. There is a problem here, though: if there are 10 possible missions of a particular grade, the first choice has a 1 in 10 chance of any particular mission being selected. Since we can't get the same mission running more than once concurrently, the second mission to be selected would have a slightly better chance -- 1 in 9. The third, 1 in 8, and so on. The selection of missions in the same grade is not an independent process ... but to be honest, I don't think it matters much and don't know enough about "combinatorial statistics" to say for certain if it would.
There is a different issue, though, one that is most certainly intentional and one that deliberately enters bias into the system by design. The issue is this: the distribution of Companion Gift, Underworld Metals and Luxury Fabric missions, including the yield of the missions, is not even close to being equivalent across grades, even across factions in one particular case.
Let me give this some numbers. It is safe to say that in general, each grade of missions has a ratio of Gifts to Fabrics to Metals of roughly 2:1:1. Please note the use of "in general " and "roughly" -- this is not precisely true for both factions or within a faction for all grades. Now consider that two of the three Crafting skills that require Underworld Trading only need Underworld Metals. The deck is stacked roughly 3 to 1 against you getting a Underworld Metals mission. Even for Synthweaving, these crafters need metals for Heavy and Medium armor and fabrics for only Light armor, so they are getting screwed by design if they need metals.
At least, as long as the system randomly distributes missions these biases will hold true.
As it is, I'm trying to collect about 100 randomizations of my UWT list of missions -- again, they aren't really independent within
a grade, but between grades and randomizations they are independent measures -- which would mean a total of 3,000 data points (100 measures of six grades of missions taking five missions at a time). It's god-awful boring, but I hope to be done with it soon. There are some simple statistical tests that can be run that will show if mission distribution is biased or random. Using these, I hope we can get an answer on just how non-random our random system may actually be ... again, by BioWare's design.