Icefrog's approach to balance in Dota, and later Dota 2, has been a rather interesting one -- one that I would call ingenious as well. So for this post we're going to talk a little bit about what I call horizontal balancing.
Dota 2 is a competitive PvP game at its core, with currently 117 different characters to choose from. Each of those 117 different characters bring something unique to the table, and thanks to how they're balanced there's a time and place for each of them. Some characters, called heroes in game, have a slightly higher winrate than others, but it's always in the scope of having an overall 60% winrate instead of something absurd like 80%.
Keep in mind that this level of balancing is actually a very difficult concept to execute, as it's easier said than done. One of the difficulties being the instinctual human tendency to hammer down the parts that stick out. By that, I mean if you have something in your game that's been deemed as overpowered, or too strong to be fair, the instinctual response is to nerf, or reduce in strength, the overpowered aspect until it's fair. For example, you have a character or a class running around killing every enemy player with one shot, so the first response is to nerf that character's damage until they're roughly doing the same damage as everyone else. The game is now safely "balanced".
The unfortunate side effect of this is that it makes each player option just a bit more homogeneous. The player's choices start meaning less.
In comparison, horizontal balancing would take that same oneshotting class, let them keep their oneshot capabilities, but make the windup to doing that damage take a ponderous amount of time. The oneshot class keeps their separate identity, and they're still good at what they do particularly if no one notices them winding up their big strike, but they're no longer good in every single case, and that is what you're aiming for.
Eventually with everyone having their own identity and point in which they're useful, very small changes to the environment itself or natural player tendencies to counter the current popular classes/characters will force those same overplayed or overpowered characters out of the meta, with no drastic number changes needed whatsoever.
There are sometimes you do need to scale back direct numbers of overpowered things, but not to a very large extent. Which is fortunate, because players can get very upset over what they perceive as huge nerfs to what they currently play.
For a topical example, since there's been a hubbub about Classic WoW for a while, let's talk about World of Warcraft as it was when it first came out, oft dubbed Vanilla. Some of the many reasons why WoW did well was because the team put an absurd amount of passion into the project, the game released at a good time, and there was a lot of hardcore fans eagerly anticipating and building up hype to explore the world they loved from the Warcraft RTS series.
Another reason why WoW did well was because each class felt very unique from each other. There were a vast amount of things that only one or two classes did that each was a completely different experience, and had their own identity. Blizzard's method of balancing over the years and expansions lost a lot of that identity. Shamans use less totems, warlocks less dots, druids stay in one form all the time, there's a litany of stuff.
Absolutely, there was a lot of stuff that was removed from WoW because it honestly wasn't fun, and I think that Blizzard was right when they said, "You think you do, but you don't," when people asked for the Vanilla experience again. However, that's because Vanilla WoW wasn't balanced all that well. It's unfortunate Blizzard balanced the way they did, because what people actually wanted was a time where their class actually felt like it meant something, their item choices felt like it meant something, their every choice in game felt like it meant something.
And for Classic WoW, that's something horizontal balancing could fix. Unfortunately they won't do that since they're aiming for a strict recreation of how WoW was back then.
Great Post. It will be interesting to see how popular classic will be over time. Blizzard mentioned the possibility of adding new classic content, between Naxx and BC, depending on demand.
ReplyDeleteI honestly think Classic will be a financial success, but it'll never come close to recapturing the numbers of the original release. There will very likely be a solid coterie of hardcore fans playing it into the ground and keeping it afloat, but it won't spread out much from there. And that's just fine, because that's pretty much what they want.
DeleteCompletely didn't see Blizzard mentioning the possibility of adding new classic content, but the more of a financial success it is the more I can believe it happening. It's really savvy of them to do so, and also really cool from a player's side.
> Blizzard's method of balancing over the years and expansions lost a lot of that identity. Shamans use less totems, warlocks less dots, druids stay in one form all the time, there's a litany of stuff.
ReplyDeleteVery much so. Spent way too much time in Vanilla, and also spent a lot of time on the boards defending the class as it was vs. how people wanted the class to be.
Every expansion leveled out the class to be more like the rest, and took away the gameplay I enjoyed about it.
That leveling out of classes is exactly what I'm talking about. So many developers do that style of balancing, but Blizzard and WoW just happens to be one of the most visible because WoW's been such a touchstone.
DeleteWhat would've been cool would have been to address balance issues by first addressing hardware issues like increasing/uncapping debuff limits, maybe tweaking ignite mechanics, and then seeing how players went from there. Maybe start phasing in the changes they made to prot paladins in TBC, which were great changes and still felt very unique to paladins, and gave the players more options instead of pretty much only using warriors for tanks.