I’m well aware of my readership numbers: nonexistent. I’m aware that this comes as no surprise, given that I don’t attempt to spread my writing in any way, shape, or form. Nevertheless, it brings me pain to have to make a point that I’ve already made, vis a vis, Overwatch 2’s new “Stadium” game mode–and, for more or less the same reason, the “Perks” system that predated it by only, what, a month or two?

I’m well aware that I’m anything but a typical gamer. When I play Overwatch, I pretty much exclusively play the Mystery Heroes game mode. That’s the one where you get a different character, randomly, every time you die. My numbers have only gone up since the last time I posted about this habit–and I don’t feel like remaking my infographic from a year and a half ago, so here you are:

As a person who, despite my best efforts, never really went away from Overwatch, it bothered me when I was grumping about the things Blizzard added to the Stadium mode, and someone simply said, “I’m having fun.” That’s nice; I never stopped having fun, but I feel like if I played Stadium, I would stop having fun. It is, in fact, substantially less fun–for me.

The real question is: how much of that is just me? I’d be dishonest if I said it wasn’t, if I tried to pose all of this as a purely objective argument. But as a person who has, over time, learned to control literally every hero in the game with a certain amount of finesse, Stadium simply piling too much more onto my plate is the dev’s way of saying “You are way too niche for us to care about.” And to a certain degree, that’s fair, and if this was the only argument I had against Stadium, it would be fair to completely disregard me.

Isn’t it funny, though, that I was thinking about this exact argument before Overwatch 2 was even released? I even posted it to this blog two and a half years ago, as Multiplayer, Content Grinding, and the MCV Problem. Granted, this ends up being slightly different, but… but it’s not, really.

The MCV Problem (In Short)

If you don’t want to read the above essay, the “Most Customized Variant” problem in gaming is essentially, “You aren’t supposed to like a level 0 character.” In a game where you transition from “fair multiplayer” where everyone is level 0, to “grind multiplayer” where it’s your goal and your job to get to some highest level X, it is literally the job of the game developers to dissuade you from liking fair multiplayer.

And I mean, literally it is their job. This is how they make money. It is their employment. If you like the game modes that they haven’t screwed with, more than the game modes they have screwed with, then they will be fired. I cannot stress “literal” enough. They have graphs, tables full of metrics. Job performance reviews are and will be conducted. If people like level 0 characters better than level X characters, people will be fired. Those people who will be fired are in charge of the game.

The problem is if the game was already good it’s difficult to make it better. And personally, I don’t see this push, nor the push when Overwatch 2 was still a concept, towards character co-op with levels, as really being about “making the game better.” Different, yes. I’m not too stuck on old ideas to see that Stadium takes inspiration from other games and tries to fold that into this design.

The problem is, simply and effectively, that it’s wrong.

The Emotional State Argument (Again)

I wrote about the Valve MOBA-shooter Deadlock last October–and I admit that I basically haven’t played it much since then. Whether or not the arguments in my blog posts are the real reason, it’s simply true that the game just doesn’t work for me. And frankly, Overwatch 2’s Stadium is Deadlock without the extra mobs, which means that not only did the devs not read my random blog post (which again, is only to be expected) but they basically went in the direction completely opposite than the one I recommended.

Anyway, I made the argument towards the end of a blog post about Deadlock’s use of character silhouettes, which was basically that the connection between character shape and character function is a reconnaissance ability meant to inform player decisions. In a game where characters are consistent, learning to use this recon ability means that simply knowing what character is approaching means you know exactly what to expect from them. If you use this information to your advantage, you feel as a player that you’ve done well, that you are mastering the game.

This becomes meaningless as a mechanic once you start injecting completely random chaos into character abilities, which is exactly what Stadium does. Oh, I suppose you could have an idea of the worst possible case for every enemy character, if they were at maximum level and you at minimum. However, in a game with 43 characters, as soon as you start to multiply those characters by adding variants, it becomes implausible to memorize all of the many variations you could be facing.

Hell, most people would say that what I do–playing every character in the game, never treating any of them as extraneous or worthless, trying to master every one–is implausible, unreasonable, unfun, and stupid. The Perks system alone multiplies that load by eight–the level zero character, two options for the first perk, and two options for the second. Stadium multiplies it by hundreds, or perhaps thousands–I have no interest in doing the math.

To be fair, they aren’t expecting the game to remain the same–they expect players to, I guess, “embrace the chaos” and just do the best they can in a world where consistency is completely out the window. And that, to be fair, is a workable game design.

It’s just kind of… contraindicated by the established fan base.

Now, gamers are fickle in the best of times, and Blizzard has done its best to drive away much of the fan base of Overwatch. It’s clear that nobody there knows what to do with the property, and even if I had my ‘druthers, there’s no guarantee that gamers as a whole would agree with my vision for adapting the game. So it’s worth taking anything I say with a grain of salt.

But to choose Overwatch has always meant choosing a game which values consistency in characters and mechanics. Part of the reason why people have been driven away from Overwatch is that adding new characters with new mechanics means adjusting the game balance by… a lot. Sometimes, a lot a lot. They’ll add a healer or a tank and the game suddenly stops being consistent. Oh, there’s some new meta, some new balance of mechanisms to get used to, but the old is gone–now you must embrace the new old, which will someday be gone again, just one more rug pulled out from under you.

In that sense, it becomes slightly more problematic when the “old” that is being gotten rid of, is the basic, unmodified form of literally every character. Oh, they’ll still exist. But Blizzard is literally paying people to make you like those characters less. They don’t want that to actually happen–they want you to keep playing old game modes, or more generally, play anything at all as long as it encourages you to keep playing the game–but they’ve decided that you must like something new more than something old, and since something old is good, they must convince you it is not good enough.

But even more than that, the Stadium game mode is going to wreck players’ understanding of characters, both as the one playing them, and as an opponent. Not completely–you still start every round with relatively low-level characters, but Stadium firmly starts you above level 0. You’ll get used to being able to hit harder, to have more health, to move faster, to have abilities that work in different ways. You’ll get used to literally every ally and literally every enemy having a bunch of chaotic modifications to their kit, and even if you understand the metagame, there will be people who modify it, who take their own path. And that’s good, in certain ways, but it’s not really great for a game whose existing fan base likes the existing characters.

If you move from Stadium to any of the game modes that made Overwatch famous, every character will feel underpowered, less flexible, less customized, less complete. That’s the MCV Problem again: the Least Complete version of the character is suddenly found lacking, even if you spent nine years coming to love them. It is the game developer’s job to make sure that the character you loved feels incomplete–in fact, they’re kind of doing a poor job of it (in my opinion) and that’s a large part of the problem.

It would be one thing if I genuinely felt like the old Overwatch was yielding to something better. Better things can exist. A version of Overwatch could be better than this. It doesn’t have to be my preferred version, or even remotely close. But I don’t think this is it. In fact, as I said… I’m not sure that they’re really trying to be better, just different. And if they’re sacrificing what’s good about the game in order to get “different,” they’ve basically just given up.

I’ve Complained, I’ve Suggested

To be clear, I haven’t simply argued against every change that Overwatch has made. I mentioned a couple times having a suggestion, and it’s still there on my blog, as the post Mechanically Addressing the Naïve Characterization. In that post, I talk about–shock!–adding new things to Overwatch while remaining consistent with the characters and the implicit narrative.

In short, it means taking what’s already established with the characters, and doing something with that. And while a game developer who lives buried in numbers and code may not see a significant difference between that concept and “buy a 10% weapon damage upgrade,” the difference does exist, and it is noticeable. Virtually everything I’ve seen in the Overwatch Stadium shop is, categorically, about tweaking numbers or screwing with game mechanics explicitly, that is to say, almost all of it lives within the non-canon gameworld rather than the implicit narrative.

(Why is the game world non-canon? Because characters die, because there can be more than one of them, and more broadly, because it’s a game with game-like physics in it. That, in itself, isn’t a huge problem, even if I wrote a blog post after Sombra’s rework in 2023 and explicitly called it “The Fundamental Nonsense of Multiplayer Shooters“. Games by definition are non-canon–if what you did was canon, you could only play the game once, and whether you won or lost, you’d just have to accept the result. Which, being honest, would be kind of a lousy game.)

Now–there are games out there where the developers play around with in-game numbers and mechanics without trying to make it model something real, at all. I’m not really deeply enough into game publisher statistics (not at all, actually) to tell you that those games failed. But what I will say about that style of game is that it is, definitively, common. It takes a rare lead developer to find and maintain the kind of ludonarrative consonance that Overwatch once displayed. And whether or not I’d succeed, if given the lead of a project like Overwatch (I absolutely do not have the credentials to do that, and am not seriously asking), what I would do would not be a common, number-tweaking modification of the game.

If I feel like Overwatch Stadium is a step “backward”, part of that is because I’ve already decided what a step “forward” looks like to me: consistency, and game mechanics that encourage players to find and exploit that consistency. OW Stadium’s “embrace the chaos” message is a signal that the people currently in charge have no intent of making a consistent game mode. I can argue that they’re wrong, but it’s simply fact: consistency is not their guide, something else is. Attracting attention, maybe, but I won’t speak for them. I’m sure that in the end, they’re just trying to find “the fun”.

As a person who never thought Overwatch wasn’t fun, I suppose I’ll just never understand them.