The skinny of it is this:
Perhaps instead of large patches which changes to content/game mechanics in which much can go wrong that happen infrequently (approx 1x a month) and usually cause nothing but derision in the community...
How about smaller updates MORE often. Not with everything mind you, but smaller things. Like missile damage. Your data mining probably produces more information of value in one day from player statistics than a whole month of play testing by internals. Moreover, look at the missile fix:
Issue was complained about for months by players.
Testing grounds released.
Testing grounds allowed players (not devs) to find faults in game splash damage mechanics in short order.
Your response: Waited a bit and did your first hotfix... which I liked that it was a hotfix. I did NOT like that it turned SRMs into ineffective paper balls unless you're using bugged streaks, bugged LRM+artemis, or reasonably functional SRMs plus artemis.
Now everyone gets panicky when they get egg on their face. It happens. And you wanted to prove you could make a big change quickly. Bravo! Now's the time to start introducing *INCREMENTAL* changes to the Missile damage until you can find the sweet spot. Same with other components such as machine guns, flamers, thermal vision, etc! Your vast pools of data are just begging to be mined. So mine them! This is the essence of a beta - to fine tune these mechanics for release. The faster you do them, the faster you can move on to other things! Use a more efficient process and reap the rewards of increased productivity

TL;DR: Smaller game balance updates, more often instead of larger updates less often.
Edited by Gevurah, 23 April 2013 - 09:37 AM.