Actually I studied programming at university and I know a fair bit about game coding. While I have never programmed any MMO games (I give you that) the company I work for as an IT tech has several quite sophisticated proprietary programmes that I have helped develop and modify. I can make reasonable guesses based on the algorithms I've been working with so far. And it is as I have said, I know this for a fact: the more paramateres a program has to analyze, the more variables we need to input, the more expensive, difficult, complex the program becomes. Even a simple data analyzer program that only measures two values (bought/sold) becomes significantly more complex if you add in just one more parameter (faulty products.) But enough about me, I don't want to bore you with my personal life.
Expertise in itself is not enough to be a deciding factor when trying to make balanced matches. Valor in itself is not enough either. Average mod levels and armor rating is not enough. Team composition is not enough in itself and never mind that it can't distinguish between tank/dps and heal/dps variants of the same AC. And what about hybrids?
You'd need a value (preferably hidden) that takes all the abovementioned into account. And of course, implement safeguards for the inevitable attempts to circumvent the system. Sure it can be done given enough time and money, no one is doubting that, other games have developed similar systems after all. But what until that joyful day? Leave PvP as it is, in this miserable state? I don't think so.
Let's examine queue separation as an interim solution (that is UNTIL WE GET MATCH MAKING) that measures two values. The person who presses button 1, gets into group 1. The person who presses button 2, gets into group 2. What happens when a premade is trying to synch solo queue? 2 people will be placed on each team if it's a same faction match. 2 will be placed on one team and 2 will wait for the next match if it's an opposite faction match. And you know what? You can even allow solo players to press button 2! That way all the masochist, self-tormenting fakirs who want to face certain death for faster queues can fill out leftover spots in premade groups.
So yeah, once again, match making all the way. But until that happens let's not sit back comfortably and sip Corellian Ale but do something that curbs the rate at which we lose people.
Well, I certianly admit I made a false assumption about your background. Sorry about that.
I still don't think Bioware can afford to make a short term fix like seperate queue's considering the potential blowback for little gains. (Most important, how would the group bracket backfill, or fill for incomplete warzones?)
Another interesting thing I found, look at this quote talking about the plans for ranked in 1.2:
The Ranked Warzone matchmaking system will match full groups of eight against one another, solo players against one another, players with similar ratings, etc. However, in order to ensure that players won’t have to wait too long for a match to occur, the system will become more ‘flexible’ over time. Solo and group queued players might be pulled together after a while if needed to launch a match. If a match cannot be made for extended periods of time, then Normal and Rank queued players (except those who queued as a group of five or more) might be pulled together to launch a ‘mixed’ match. Players will always be notified when a mixed match is made on the Warzone enter prompt. The rewards received in mixed matches will be based on how the player queued (type of commendations, rating adjustment if any, etc.).
While I found no mention of a ranked matchmaking system actually being implemented in the 1.3 notes or it's article, it appears there may even be ground work alread designed for what has been suggested. Essentially a split solo and group queue, yet with the ability (and flexibility) to cross fill after a short amount of time.