Wicc
Auto-translated
As a direct participant in many discussions regarding TE and VT balance, I can tell you how they went:
- Something is a bit weak with N, let's buff it by 20%.
- Won't it become OP?
- It might, then we'll cut it down to 10%.
There is no fundamentally different way in Heroes and there cannot be; experimentation is the basis of everything, only rough estimates are possible in one's head.
To understand the complexity of the task, try answering this question, for example: how many elves should Ivor be given at the start so that he has a 50-50 chance against Kirill on 6lm10a? Or at least describe the methodology for mathematical assessment.
Or an example from TE-VT — Brogh's OP status is well known; people would switch to him regardless of their Castle, sometimes even in the late stages of the game, but for the first 2 years of online TE (2006-2007), no one suspected this, starting with Vistan or Tazik. Experiments with Brogh began in 2008, a full transition to him happened in 2009, and he gained the status of an universally recognized OP in all of VT in 2010. Back then, admittedly, fewer games were played and there were no streams, so the evolution of strategies moved slower, but still.
Regarding "obvious Jeremy" — if you remember, when Cassie had 20-30 nymphs, she was considered the primary starter, although from the height of modern experience it's clear that Jeremy was already better back then. But the general prejudice against artillerymen was too strong: "well, and how are you supposed to take cons-hives-Dragon Utopia with them without a strike stack and off-def?" The fact that the cannon and log compensate for this only became clear after extensive testing.
Before HotA came to light, I said exactly this: "there will be no balance on the first attempt, so we'll do it by eye, and then fix it based on the results of tournament playtesting." And, by the way, note that for such a volume of innovations, the castle balance turned out very good, much better than in both SoD and VT.
