RabbleFroth -> RE: Balance Tracker (5/1/2012 11:23:36)
|
I can talk a little bit about this. I do want to start though by saying the Balance Tracker system is not the only thing we use to determine what needs to be balanced or how to balance. We use a mixture of the tracker, personal experience from playing, feedback from players, and more to determine if we think changes need to be made. So first off, what exactly does it do? It tracks the overall win percentages of each class in 1v1 matches at 5 level intervals. It throws out same class matches (Blood Mage vs Blood Mage) because obviously that will always result in a 50% win rate and skews the data. This gives us a clear look at the overall status of each class. It makes it very very clear if a single class is doing much better or worse than they should be in each level range. Obviously, there are flaws with this system, and is the reason it cannot be the only method we use to determine how to balance. So what are the issues? * First off, it cannot differentiate between builds. If Mercenaries are doing terribly overall, but a single build is able to get 90% wins and only a handful of people are using it, the tracker will not show this. Once it catches on and becomes popular though, we'll see a spike in Mercenary wins and be able to fix it. This is why player feedback and personal play is important, so we can notice these trends early. * Secondly, it only works for 1v1. I'd like to implement a similar system for 2v2, but it's a little more complicated and I haven't been able to dedicate the time to do so. Luckily, proper balance in 1v1 helps with balance in 2v2. They're not equal of course, but related. * Finally, they're just numbers. At the end of the day, the goal is to make fights that are fair and enjoyable. Close balance helps with this, absolutely, but is not the only part of the equation. Skills have to feel good, different play styles should be available for different classes, etc. These are things an objective Balance Tracker cannot describe with a chart.
|
|
|
|