What can Blizzard Learn from MOBA Balancing/Design - Page 3

On September 07 2013 00:19 Plansix wrote:

He is also wrong, Balance patches come every 6 months or more. Not every week. The weekly patches are for store updates and sometimes really minor balance changes.

On September 07 2013 00:26 TXRaunchy wrote:
TL;DR When one unit gets used too much, Blizzard nerfs it.

On September 07 2013 00:25 Crownlol wrote:


Well, at least we know the first letter of the new Terran unit in LotV.

The MellBat, of course.

On September 07 2013 00:28 Crownlol wrote:


Yes, and no. You're not including the meta changes of implementing new, very strong champs. Some champs are immediately played at the top level as soon as they release: Vi, Kha'zix, Rengar, Elise, Diana, Zac, Draven, Thresh.

On September 07 2013 00:30 DinosaurPoop wrote:


5M is WAYY too much production

On September 07 2013 00:30 Plansix wrote:

He is talking about Dota 2, not LoL. League nerfs the shit out of everything that is good.

On September 07 2013 00:32 DusTerr wrote:

not if it can be made from reactors!

On September 07 2013 00:33 Crownlol wrote:


Well then, I have nothing to say about DotA2 since I don't follow it (played it a bit, actually I think I'll switch from LoL).

On September 07 2013 00:37 Plansix wrote:

The recent trend in patching in Dota 2 is to patch about every 6-8 months and to only provide minor nerfs to dominant heroes and buffs to unloved heroes. It basically keeps the games sort of stable, but encourages new things in the game play. It has lead me to believe that nerfs are bad, buffs are good. If you are looking to give it a spin, let me know. I am terr-bad, but the game is more fun with good people.

Jayomat: People need to understand that every RTS "gets figured out" at some point. The pros are playing all day and find ways to abuse certain styles. To think it is realistic that you can "just play as random as you please" and it will all be feasible at the highest level is pretty naiv. If you have a limited set of units, maps and abilities, you have a limited set of things to do. Yes the meta changes and evolves, but there will always be dominant build that rule over most others. Not saying that's good, but that's what it is folks.

Larse: You are right to some extent.

But the pragmatic solution is to increase that "ceiling of diversity", so that this "figuring out" process will be long enough for the life time of the game. Theoretically, I think it lies in the field of Combinatorics.

I have asked an exact same question as yours to the dota2 community, and I got some interesting answers about this. If they are not over-optimistic, their general answer is that you may not be right: http://www.reddit.com/r/DotA2/comments/1k1lxl/a_question_from_a_starcraft_fan_in_esports_do/

Jayomat: you really cannot compare SC2 (RTS) and DOta2/LoL (MOBA) in that regard. Reading the first couple of most upvoted comments, the repeating "facts" are these:

There are MILLIONS of permutations of viable hero compositions. On top of that, item and skill builds are flexible and situational.

You won't find optimal builds here, there's just too much combinations.

Strategies vary team to team. There are so many possible compositions, interactions, and variations that you'll see quite a bit of diversity in DOTA. On top of that, you have a slew of different item builds from game to game, and the laning setup can be extremely diverse.

I mean it's pretty obvious what their simple argument is: the variables are in higher numbers. So many heroes who can pick so many items. Combine that with the fact that you play in teams of five. I mean come on man, isn't it obvious to you that they say the same thing which I said, just with higher numbers?

The viable strats in SC2 are limited. Of course there are always situations where a crazy strat might work (GSL, BC+Hellion+marine vs zerg on 2base). But you cannot compete in a consistant manner in a 1v1 game with builds like these. They are TOO situational, and SC2 in general is very predictable. If there was no pro play, sure, it would take quite some time to figure stuff out. But with the pros playing, you see what is viable at the highest level after 1 month. If MMMM is the safest way to win in TvZ, that's what people will do. If the amount of dota2 heroes would be decreased to 10 with set combinations and only 20 items, you'd see the same happen there.

You cannot simply add more and more units/skills to a RTS game like SC2. The balance is already a "clusterfuck", how do you think adding 10 units will play out? It was always the same, C&C, WC3, AOE and SC. All have "best builds" after X amount, and only "mindgames" and "metaing" bring diversity, not the game itself.

Larse: I see your point and I said you are right to some extent. Almost all RTS games have this problem.
But I am more optimistic as I think there are pragmatic solutions to reduce the degree of this problem (since it can't be eliminated in practice). Increasing the amount of units is an obvious solution, though you said it will cause balance problem. It doesn't need to be 10 new units at the same time. But if you gradually add units as the game staggering, it will change the stagnation and the impact on balance will be more modest. At least, the impact will be much smaller than the impact of an expansion.

Also, making the unused existing units more viable is another obvious solution. Blizzard has a more conservative approach to this as they are very careful about balance. But my personal opinion is that they can be a little bit more progressive in changing the less viable units in the game.

Also, I feel that your "moba can do it but RTS can't" sounds more like an excuse for me to not innovate and progress.
Sorry if the tone sounds rude.

Jayomat: I agree that adding units gradually and changing the unused ones would "help", no doubt. but that doesn't help all that much because most people play to win, especially pro gamers. you will witness how they figur out the stuff that works. also, the units will be too similar at some point. yeah sure you can think of many crazy, very specialized units with spells'n'stuff, but in the end what do you end up with? basically WC3 with more eco, or dota2 without eco. I don't even know man, I want SC2 to be "a better game", whatever that means, but I think we also must understand that the "ever evolving" never "stagnant" RTS game has yet to exist.

Larse: There won't be one as theoretically the optimal and near optimal builds will be figured out eventually, unless the game is designed by mathematic grandmasters for a long time.

As I said, all that matters it's a pragmatic solution to reduce the degree of this problem so that in the life time of a game it won't reach that point permanantly. In other words, every time it reaches that point and people are getting bored, the designers should implement changes to disrupt the optimality and thus naturally reinforce the emergence of another optimal one. Once this new optimal one gets figured out and they are getting bored, do it again, until the game eventually get old enough to die out.

The approach of expansion is the same way at the fundamental level, but they do it as a big chunk while my solution is more gradual, though admittedly it is close to the tattoo model in this subreddit.

On September 07 2013 00:12 thezanursic wrote:
I'll try to explain the problem

BW - TvZ MM
WoL - TvZ MMM
HoTs - TvZ MMMM

The problem

On September 07 2013 00:26 FLuE wrote:
Sometimes I wonder if they should focus less on balancing and more on creating a constant state of flux. The game gets boring when the meta settles. It becomes very cookie cutter. Balance actually isn't exciting. The game has been fun when certain races or builds are a bit OP because it forces the other races to try new and think outside the box.

It would be cool if after each WCS they did major patches, added abilities/removed abilities, buffed and nerfed certain units. Then gave everyone 2 weeks to practice with the changes and the new season starts up. That is what makes the game so fun to watch after the initial WoL release and HotS. The uncertainty. Ultimately you keep the core of each race the same but it would be a change. Instead of trying to achieve perfect balance instead achieve good balance with variety that then rewards mechanics, decision making, and adaptability.

The game these days stagnates very quickly. With replays, team houses, forums, build order programs people find the ideal builds to fast and then don't try anything new. This would be a way to create some new play each season.

On September 07 2013 00:59 vthree wrote:


Actually, BW did this by changing up maps and that help the meta evolve or new tactics came about. But in sc2, high ground advantage is minimal and not many map control units (a lot of matchups don't even require map control). So most maps play out the same.