I started playing co-op actively in early-2018, and it quickly became one of my favorite game modes. Its design was utterly brilliant: an accessible and linear progression path, an emphasis on implicit over explicit cooperation, and strong, consistently structured missions, just to name a few great features. I liked it so much that I wrote an hour long in-depth review of the game mode:

Ever since the release of Tychus, I have grown increasingly uneasy with co-op’s design direction. These concerns have only grown over time and were cemented by the launch of Mengsk and the inclusion of a Brutal+ difficulty. I’d like to talk a bit about these concerns and explain why I don’t agree with the current direction of the game mode.

#1. Brutal+

A long-time feature request from the co-op community has been the ability to increase the difficulty beyond Brutal. Brutal had become much too easy for veteran players with maxed out commanders. There was significant demand for some kind of end-game content that enabled better replayability. Mutations didn’t really meet this need because they were more like one-off puzzles, and anyway they could only be credited for experience once per week.

I asked the developers about this at 2018’s Blizzcon, and here’s what they had to say:

What we’ve always said there is, it’s very easy to make changes to end-game such that we can increase the mobs, the number of units that spawn, we can increase the HP or damage of the units that spawn. But the challenge there is that co-op commanders are so powerful compared to their multiplayer counterparts that they kind of have, like, “delete wave” buttons no matter how many units there are or how much HP there is. So, we’re actively looking into how we can expand the difficulty of co-op but we’re thinking in more perhaps creative ways other than just scaling the power.

I’ve thought about this statement many times since I first heard it, and I’ve repeatedly concluded that it’s simply not true. While it is the case that techniques exist for any commander to solo stomp any mission, it’s never been my experience that a majority of players play the game in this optimized way. Aside from the new commanders – which we’ll discuss later in this video – solo’ing everything as Alarak or Raynor or Fenix or even Abathur is simply too difficult for most players. To me it’s undoubtedly true that increasing the difficulty in the straight-forward ways that were suggested would work quite fine to make the experience more interesting for veteran players.

Anyway, I think it’s fine that the developers chose another path, because there are many reasonable ways to solve the problem of co-op’s endgame. Here for example are a number of suggestions from Reddit. What I don’t agree with is the choice to leverage randomly generated mutations. The issue here is that many mutators hard-counter the design of individual commanders. This isn’t a problem in regular mutations because they’re designed thematically for players to figure them out and counter them. But in Brutal+, you have no idea what you’re getting until you launch into a mission. The first time I played I was given Fatal Attraction and Speed Freaks against a Zerg ground composition as Tychus. If this were mutation I would have just picked Zagara and steamrolled everything with banelings, but since I was Tychus, there really wasn’t much I could do except passively watch my partner carry as Zeratul.

Examples of this abound in the community: Nova with Kill Bots, Raynor with spider and widow mines, Tychus with Fatal Attraction plus Black Plague. The last game I played I chose Alarak, a relatively low mobility commander, and got Speed Freaks on Miner Evacuation, with Missile Bombardment to boot. I have no idea how the developers imagined players would react to this, but the result seems pretty obvious in hindsight: people queue Brutal+ and instant leave if the mutator plus commander combination seems unfair. This is not my idea of a fun or compelling end game. The idea that it’s designed for parties is similarly misguided, at least from my perspective. The last thing I want to do with a close friend is constantly re-roll mutations and argue about what’s fair and what’s not.

Really the larger problem with Brutal+ is that it misunderstands the demand from players. Brutal being too easy is just a symptom of the larger problem that co-op doesn’t have a satisfying end-game. Mastery levels aren’t particularly interesting and after you hit level 90 they become more or less meaningless. This is one of the reasons I was excited when they announced leaderboards as a top-line backlog item at 2018’s Blizzcon. It pairs well with a ton of great end-game ideas: daily challenges, weekly quests, percentile rankings, and so forth. It’s especially symbiotic with some kind of horde mode, which is great because horde is straight-forward to implement: improve enemy stats by some percentage after every wave, throw in some random events, and boom, you’ve created a thousand hour experience. It’s more complicated than that, of course, but not orders of magnitude more than what the team has already done.

This is one of the reasons that I don’t agree with the excuse that the developers are too resource-constrained to do anything but random mutators. There’s a lot of low-hanging fruit that they decided not to pursue here. Furthermore whether or not you have a lot of developers doesn’t determine whether or not something is a good design decision. The fact that random mutators is easy to implement is neither here nor there: the question to ask is whether it’s a good idea in the first place.

#2. The New Commanders

I split the co-op commanders into three groups: the launch set, consisting of Raynor, Kerrigan, Artanis, Swann, Zagara, and Vorazun; the expansion set, consisting of Karax, Abathur, Alarak, Stukov, Dehaka, Fenix, Han and Horner and Nova; and the post-revamp set consisting of Tychus, Zeratul, Stetmann, and Mengsk.

Both the launch set and expansion set of commanders feature clear trade-offs. Raynor’s army is powerful but melts when poorly controlled. Kerrigan’s army has incredible mobility, but it’s expensive and hard to replace. Alarak is among the most powerful commanders, but he has limited mobility and struggles against air-heavy compositions. Etc etc etc. All of these commanders are powerful when played correctly, but they all have weaknesses to balance out those strengths. This in turn rewards player skill while also encouraging implicit cooperation with your partner, a design decision I mentioned earlier as being utterly brilliant. Co-op works in part because it gives you the benefits of a multiplayer experience while limiting the tedium of having to explicitly communicate with your ally.

The post-revamp commanders turn this on its head by featuring almost no trade-offs at all. Let me set aside Stettman for a second, whom I’ll discuss in the next section. Tychus, Zeratul, and Mengsk are distinct from the other commanders in that they’re simply good at everything. They have a reliable early game, their average power level is quite high, their economies and start-up times are excellent, they have little trouble with any particular unit composition, and they’re extremely easy to play.

It’s just not that interesting to play commanders that have an A-rank at everything. It’s equally bad playing alongside them because they’re so easy to play that your partner ends up doing everything. I first noticed this problem with Tychus when I started leveling my commanders on North America, a region of the world where a non-trivial number of players seem to believe that their role in a game mode entitled “cooperative missions” is to attempt to solo every engagement on the map.

It seems to me that in this new design philosophy, instead of offering players interesting decisions in response to problems, the developers opt instead to remove the problem from the game. How does Mengsk decide between early game defense and early game economy? He doesn’t, because his workers and troopers are interchangeable, his supply depots are also bunkers, and he starts the game with a fully saturated main. How should players navigate Zeratul’s tech tree? Let’s just eliminate the concept of researching upgrades entirely. What heroes should Tychus select to best counter the enemy composition? Ah, I know, we’ll design Tychus and Rattlesnake to be a meta composition that kills everything, and then we’ll make the other heroes powerful enough that it ultimately doesn’t matter who you pick.

(Mengsk is a small improvement in this regard because at least he doesn’t have an instant army teleport button. But I think this is more because he is designed as a defensive commander who does not need a huge amount of mobility in the first place.)

The crucial point here is the elimination of interesting decisions and the ubiquitous streamlining of gameplay. Many of the launch and expansion commanders are also very powerful. But they still feature trade-offs, they still require the player to make decisions and have a game plan from start to finish. It’s fine to make one or two commanders an exception to this, but three out of the four released in the past two years is too much. It’s not unreasonable for commanders to have to choose tech paths or have to control armies or to simply have inherent weaknesses: it’s a core part of what makes the game mode interesting and what makes each mission stand out as unique.

#3. Stetmann

I don’t agree with Stetmann’s design direction for a couple of reasons. The first issue is with his progression tree, specifically with the way his level 15 upgrade changes how players are intended to open. With most other commanders, level 15 unlocks a significant and universal buff, like Raynor’s attack speed or Dehaka’s mutation bonuses. This is great, because it pairs well with everything players have learned from levels 1 to 14. Players get to re-visit the builds they used as they were leveling up and see how they’ve been augmented with the addition of mastery skills and the powerful level 15 upgrade.

Stetmann, as far as I can tell, is the only commander to not do this. Because Stetmann’s late game units are gas-intensive and because his early game units are not that strong before upgrades, players are incentivized to rush to Super Gary while they tech up and build out their army. But Super Gary is unlocked at level 15. This means all of the builds and openings players learned from level 1 to 14 are discarded at max level. This significantly detracts from the inclusion of a progression path in the first place. It misses an important design paradigm inherent to the other commanders: the progression path is not just a rewards loop, it’s also a teaching mechanism, it’s also building an investment in a commander by showing you how to play them.

Putting aside the progression tree, Stetmann’s larger problem is that he’s just boring; his design lacks a core creative hook that would incentivize players to select him. When you look at other support commanders like Karax or Swann, they have their own interesting features that make them cool to play on a stand-alone basis. With Stetmann, neither his units nor his upgrades are particularly unique relative to the other Zerg commanders. His main selling point are his stettalites, which are very tedious and far too mechanically intensive. The decision to have enemy AI prioritize them above virtually everything else means they’re not very reliable without Super Gary’s egonozone. This really should have been balanced closer to creep: helpful, but not necessarily essential to the commander, and certainly not requiring this many actions.

This is actually the main reason that I raise up Stetmann as a problem. I would be OK if the developers took a creative risk and ended up missing the mark, because at least we would all learn something about what works and what doesn’t work in co-op. But Stetmann feels uninspired; he’s a generic Zerg commander with a bunch of cool skins. And it shows in the data, too: I’ve played hundreds of games of co-op since Stetmann’s release, and I have run into exactly one Stetmann partner after his initial launch.

#4. Final Thoughts

When I did my in-depth review of co-op, I discovered that its popularity was not an accident. Co-op was successful because it made a lot of really good design decisions. The way the expansion set of commanders built upon these ideas made it clear to me that the co-op team had really figured something out here; I went so far as to say that this game mode is the future of the genre.

I would love to see the developers get back to what made co-op successful. Create interesting decisions for players to make, reward players for their skill level, put some genuine creativity behind the commander designs. Recognize the flaws in the end-game and address them at their root cause. I think if the team can move back in that direction, the game mode will have a brighter future than the one it has today.

Thanks for reading! I create analytical content for StarCraft and Age of Empires. I’d love it if you followed me on Twitter and Facebook to catch more of my work, including long-form articles and deeply researched, in-depth videos. All the best and see you next time.