Balance before release, not after.

morph3us

The current approach to ensuring adequate play balance
I'm wondering if the devs are devoting sufficient time to ensuring that a character is properly balanced before they are released on a two weekly cycle. I could be wrong, but it seems to me that even though the devs are spending much longer than two weeks planning and creating these characters, if they're releasing characters at fortnightly intervals, it doesn't leave a great deal of time in ensuring proper play balance. This problem becomes magnified as more characters are added to the pool, as the potential for unforeseen game-breaking combinations escalates.

Post hoc rebalancing isn't a good solution
The current approach appears to be, after an initial short period of balancing pre-release (as we saw with Falcon, Psylocke, and Prof X), to watch how the new characters play in the wild, then rebalance them after a short to prolonged period of time. There are several problems with this approach.

1. Overpowered characters tend to get rebalanced relatively quickly, since overpowered characters either warp the meta or "break" the game 2. Underpowered characters tend to languish for a lengthy period of time, since all that happens is they get underutilised, which has fewer implications on gameplay 3. Post hoc rebalancing undermines your player base's faith in spending money (or time) in accumulating covers. A 4* cover costs approximately $20-25, a 3* cover approximately half of that. Given that the only end game in MPQ at this stage is building characters, post hoc rebalancing disincentives your player base from buying token packs and covers.

To draw an analogy with Magic: The Gathering, when a new set of Magic cards is released, they are final. There's no instance of WotC changing the text on a card, because the initial pre-release balancing run wasn't correct. I regard the business model of MPQ and MtG to be reasonably similar. You can obtain rare cards via lucky card pack pulls or you can buy them outright. Same goes for rare covers in MPQ.

Post hoc rebalancing needs to be less common in MPQ, and to be of smaller scale.

I appreciate that rebalancing is necessary on occasion, and it's good for the game on the whole. I'm not going to comment on the specific rebalancing that's been done at this time, since it might all actually turn out to be reasonable.

What I would ask for though, is that the devs spend more time pre-release ensuring characters are balanced, rather than after release. MPQ is built around character accumulation and development; constant post hoc rebalancing is undermining that model. If slowing down the release cycle gives you more time to ensure characters are properly balanced, I think we'd all appreciate that.

ynglink

While this seems like a good idea initially, what about buffs? Unlike a card game, you can update characters to make them better rather than Re-releasing them in a variant form to update them. Same can be said about nerfing a character vs. cards. Look at a Card such as Ancestral Recall or Black Lotus. There are several throwbacks or "fixed" versions of these cards down to even replicating the name in a way (see Ancestral Visions and Lotus Bloom). Another thing you need to look at is that this isn't a card-type game. There are no "ban" lists to help balance things as well.

So I'm okay with characters getting buffed and nerfed. People complain about 4Thor getting nerfed, but her best friend got the opposite treatment, so there's that.

morph3us

Fair point. I don't object to buffing and nerfing philosophically, it's just that it doesn't make sense in the context of this game's model. As far as we're aware, the game's main revenue stream is from roster slots, token pack purchases, and cover purchases. Given that it's entirely built around the characters, any rebalancing affects the value of the digital goods. As such, it's unsurprising that we're seeing a litany of complaints about some of the rebalancing.

You make a good point about "ban" lists. It would make more sense to have a little more forethought into character balance pre-release, then multiple formats to provide more variety in character use.

With regards to the throwback/''fixed' versions of broken MtG cards, I would say that, yes, that is essentially a nerf/buff, but it's also a different card. If you have a Black Lotus, for instance, you still have it, it hasn't been changed. Admittedly, you can only play it in a very restricted number of formats, which is why your point about multiple formats is a good one.

fmftint

That would require extensive internal testing and thought on character interactions. They clearly do neither

Unknown

There is no game that can just test internally no matter how rigorous to catch everything. To quote something from a WoW dev, 'there are at most hundreds of us versus millions of players, many of them are considerably better than any of us at our own game. It'd be weird if the players can't figure out things we haven't figured out.' The problem is the timeframe for balance, which is fairly lacking. The competitive nature of this game also makes damage control difficult. If every event is non competitive, like say Gauntlet, at least the AI isn't going to complain about how Iron Fist is overpowered. But in a competitive event, obviously whoever doesn't have the latest overpowered guy and ends up doing worse is going to blame that guy, whether correctly or not, as the cause. This doesn't mean an event like Gauntlet solves the balance issue, but you'd have a lot more time to compensate if events are more like Gauntlet.

morph3us

You're right, it's not feasible to completely test for every combination and permutation internally, but it does seem like any testing at the moment is reasonably abbreviated. That's probably related to the small dev team, and therefore limited resources. I'd prefer they spend less time developing and releasing new characters, and more time testing them before release. Halve the release cycle to monthly, and that frees up a lot more time for internal testing.

The difference between something like WoW and MPQ is that WoW is selling you a subscription model for the gameplay, and in game content. You're not actually paying to develop your characters as the endgame.

I absolutely agree about the timeframe for balance, and the competitive nature of the game.

Just to reiterate, I'm not diametrically opposed to rebalancing characters after release, I just don't think they should be released in such a state as to require significantly major reworks later.

GuntherBlobel

One thing they could do is preview the powersets a bit more in advance, so that they aren't scrambling to fix problematic character interactions like with Prof X's (or too late like with IF's ).

The other thing they could do is just get rid of purchasable Ability Upgrades altogether, but I suspect that players would dislike that even more than periodic nerfs. But that's been my personal belief since the Ragnarok nerf and the reason I have so rarely bought them.

Lastly, I think the 4* Thor nerf was less about her being OP, but more about Ragnarok's charged tiles. With 4* Thor's initial specs (OP or not), they could never create another character that created , , charged tiles without unbalancing the game. A lot of people want Gambit to have charged tiles. The Devs shouldn't have limit themselves for all future characters based on previous decisions that made sense at the time. I think they need the freedom to change existing characters. Again, this is why I think they shouldn't have sold specific covers in the first place.

babinro

I think one of the more positive changes of late has been releasing the character stats PRIOR to release as was the case with Professor X and Iron Fist.

This lets the player base speculate on what might be overpowered for the dev team to potentially address prior to launch. It already worked with Professor X and we saw an Iron Fist change quickly (where most of us would have expected it to take a minimum of 5 months).

Ideally it would be amazing if D3 could just launch all characters perfect from the start but all companies out there have trouble with game balance. Mistakes are made even in truly competitive e-sport titles like StarCraft 1 & 2.

Giving the player base character information as early as possible NEEDS to continue for every release. The sooner the better. Take advantage of our feedback even if you choose not to use it.

NorthernPolarity

I think previewing a character 2ish weeks in advance and seeing what we have to say would go a long way in making sure especially imbalanced characters aren't released (cough IRON FIST cough). Sometimes we'll get it wrong, but when everyone seems to universally agree that something seems OP (IRON FIST THREAD), then its a pretty good sign that things should be changed.

TaoSpoons

morph3us wrote:

You're right, it's not feasible to completely test for every combination and permutation internally, but it does seem like any testing at the moment is reasonably abbreviated. That's probably related to the small dev team, and therefore limited resources. I'd prefer they spend less time developing and releasing new characters, and more time testing them before release. Halve the release cycle to monthly, and that frees up a lot more time for internal testing.

The difference between something like WoW and MPQ is that WoW is selling you a subscription model for the gameplay, and in game content. You're not actually paying to develop your characters as the endgame.

I absolutely agree about the timeframe for balance, and the competitive nature of the game.

Just to reiterate, I'm not diametrically opposed to rebalancing characters after release, I just don't think they should be released in such a state as to require significantly major reworks later.

I disagree with this post on a couple of points:

1) Re: WoW to MPQ comparison: the difference you highlight is in the payment model supporting the game. The games themselves bear many similarities to each other. They both include an experience gain/level element and they both lack a traditional "End" condition. Their challenges in the arena of game theory are pretty much the same.

2) Additional time testing won't always catch the kinds of problems* that stem from character interaction balance. Sometimes, you need live data to confirm or disprove a particular problem condition and even then, a short timespan can lead you to bad conclusions. The longer you gather data from a production build, the more accurate a picture you get and the better your final decision will generally be.

*I don't necessarily agree that the recent rebalancing is a problem, but that's a discussion for another thread.

morph3us

TaoSpoons wrote:

1) Re: WoW to MPQ comparison: the difference you highlight is in the payment model supporting the game. The games themselves bear many similarities to each other. They both include an experience gain/level element and they both lack a traditional "End" condition. Their challenges in the arena of game theory are pretty much the same.

I think the payment model is relevant in this case, however. In WoW, the payment model is a subscription supporting the game as a whole. In MPQ, the payment model is directly tied to the characters, particularly in the case of purchasing covers for upgrades, which is where a lot of the complaints are emanating from. As such, I do think major post release rebalancing poses an issue with regards to the payment model. Players are drawing a direct link between their expenditure of HP on covers and the recent rebalancing of specific characters.

With regards to additional time spent testing, I agree that it won't catch every single problem, but if we use IF as a specific example of a character that was inevitably going to have to be rebalanced, that's the sort of problem additional time testing might have assisted with, given that he's been rebalanced only 2 weeks post release.

morph3us

GuntherBlobel wrote:

Lastly, I think the 4* Thor nerf was less about her being OP, but more about Ragnarok's charged tiles. With 4* Thor's initial specs (OP or not), they could never create another character that created , , charged tiles without unbalancing the game. A lot of people want Gambit to have charged tiles. The Devs shouldn't have limit themselves for all future characters based on previous decisions that made sense at the time. I think they need the freedom to change existing characters. Again, this is why I think they shouldn't have sold specific covers in the first place.

No question that's why she was nerfed. By the same token, it reveals a lack of planning on the devs part. They've released one character with Charged Tiles, and then realised after release that the mechanic was causing issues for future characters. Wouldn't it have been better for them to plan a number of characters who were going to used Charged Tiles in one batch, to examine the interactions prior to the release of the new mechanic? They don't have to release them all at the same time, but it would make sense to do the Charged Tile balancing for the entire hypothetical group of characters all at once.

TaoSpoons

morph3us wrote:

I think the payment model is relevant in this case, however. In WoW, the payment model is a subscription supporting the game as a whole. In MPQ, the payment model is directly tied to the characters, particularly in the case of purchasing covers for upgrades, which is where a lot of the complaints are emanating from. As such, I do think major post release rebalancing poses an issue with regards to the payment model. Players are drawing a direct link between their expenditure of HP on covers and the recent rebalancing of specific characters.

That link is a false one.

Every cover in the game is available through continued play for theoretically zero monetary cost. Anyone who spends money to purchase hero points to in turn purchase covers doesn't buy power; they buy convenience. What a lot of people are really complaining about is that the window of convenience was shorter than they expected. However, there was never any guarantee on the length of their convenience benefit anyway, so their complaints are moot.

tl;dr: Anyone who plays a game involving a roster of characters that are: a) balanced against each other and b) ever increasing should expect their expenditures in those characters to depreciate due to balance shifts in addition to traditional depreciation-over-time.

morph3us wrote:

With regards to additional time spent testing, I agree that it won't catch every single problem, but if we use IF as a specific example of a character that was inevitably going to have to be rebalanced, that's the sort of problem additional time testing might have assisted with, given that he's been rebalanced only 2 weeks post release.

I wasn't saying they needed to spend additional time testing. I was saying that internal test data only goes so far. I know from my work that there are problems which only show up or are demonstrably problems in a live user environment. It's quite possible D3's test data showed Iron Fist was powerful-but-not-unbalanced, but the live data told a visibly different story.