Jman5, on 12 September 2013 - 12:37 PM, said:
The developers do not nerf and buff things based on who cries the loudest on a forum. If they did, the game would be completely different. Ghost Heat and 3PV would have been out ages ago if this was the case.
The developers look at what people are saying, they look at their own server-wide data, and they look at potential changes coming down the pipeline that could impact said-weapon (like fixing HSR, or the introduction of a new mech), and finally they look at whether it's fun. The community is part of the process, but we are only one small piece.
Your opinion that UACs are fine is no more valid than my opinion that UACs need some tweaks or a reworking of how the weapon works. If you disagree with the developers, make an argument based on reason and evidence and not: "It's fine because I say it's fine and if you disagree, you're wrong!"
I am not sure I agree with you that the Devs never balance weapons based on input from the forums because when it comes down to it, perception is much stronger than reality.
Case in point is the Type 59 in WoTs and if you have ever played that game you know that it is considered the most Overpowered and unbalanced tank in the game, even after several nerfs. In any case, there was a dev post once discussing why they had to remove the Type 59 from sale. They specifically said that according to all their data, the Type 59 was an average performing tank at best. The key is is that "ACCORDING TO ALL THEIR DATA" the Type 59 was average.
However, player perception was that this tank was far from average so all it did was generate complaint after complaint of its Overpoweredness and everyone with the spare money to buy one was doing so to the point that the game got to be 7-10 Type 59s per side. Therefore WG had to actually stop selling these premium tanks and commit to some nerfs to keep them from being so popular. Basically due to mass hysteria, an average performing tank by all the data WG collected had to be made unavailable. The point is, there is definate precedence for balancing based on feedback, not actual data if for no other reason than to fix player perception.
I think the same thing happens in MWO more often than people think. Though alot of people will disagree, take the PPC for example. Everyone precieved them as overpowered and it got to the point 80% of the player based mounted as many as they could on every mech they owned. This created issues beyond the weapon being balanced or not as far as the actually backend data was concerned including alot of negative press when they were struggling for market share. Therefore they had to nerf their popularity which they accomplished in an astounding fashion. I think it when from 80% PPC usage to like 10-20% PPC usage in just a few days. PPC problem goes away, negative press diminishes and none of it has anything to do with balance.
This scenerio may or may not have happened (though based on my testing it is 100% accurate) but you have to admit, it would be an easy way to solve a balancing problem that was occuring just before the official launch of the game.