Iskar: We look at the meta every day. In day one it was Bomb Warrior. We evaluated the population, win-rate, looked at matchups, got anecdotal data on how if feels to play or play against, then played it a bunch at high levels on the ladder ourselves. Then it became Rogue, then Druid, back to Rogue, finally onto a combination of those three classes and Hunter stabilizing at the top in terms of win-rate. All along the way we talk about what changes we would make to any individual class if it got to a point we thought it was necessary. With every deck we do this, even if we're not publicly discussing what we might change. What I can say, and have said in various publications over the last week or two, is that Rogue is the class we're looking the closest at. The reason for that is that even though the best Rogue deck tends to cycle between best and fourth best archetype, it's the class with the least amount of poor matchups. When a deck is 53ish% win rate but has only 1-2 matchups below 50%, there is a high risk of the meta stabilizing around that deck. On the flip side, when you look at a class like Warrior the matchups are a little more varied, with some matchups being around 45% and some around 55%. Polarized matchups get a lot of flack for various reasons, but a great deck having good and bad matchups is a big reason why the meta is able to shift on its own rather than stabilize in one spot.
Of course, it's not all about data. The reason we actually change things is because of how they feel. One of the factors that contributes to how something feels to play against is how long it's been around. For example, even if best deck X felt alright to be the best for a week, players probably won't feel like that if best deck X is still there 2 months later. That's where data comes in. We've been watching gameplay trends in data for 15 sets now and have a pretty good idea what a meta looks like when it's about to see a shift and what a meta looks like when it's completely stabilized. The part where data doesn't help is when we have decks that cast Shudderwock 15 times or have an indestructible Lifestealing weapon. Those changes are based around feel and are what we generally come to places like reddit looking for.
In terms of communication, we first have to get aligned internally. This subreddit finds out what we've decided to do when we get a patch date in place, the official messaging has been written, and it's been translated into all languages. This process usually takes a week or a week + a couple days. This is all to say, once we decide to do a patch, you will all know within about 7 days of the moment we decide.
懒人包:
他们一直在观望meta
改动的日期不知道 但一定会提早7天通知我们
改动的标準不是强度而是对上他们的feeling
如果要砍的话贼会是第一个
