Mutual Random Could Be A Cool Differentiator For Stormgate's Ladder

Thinking through the pros and cons

A few years back, Age of Empires II introduced a feature to the ranked ladder called Mutual Random. The concept is pretty simple: players can opt-in to playing random against anyone else who selects the same, with civilizations revealed on the loading screen. If you match against someone who prefers to civ pick, you fall back to your own civ pick (which can also be random, if desired). The feature turned out to be very popular at higher levels, which got me thinking about how Mutual Random might work in a purely asymmetric Blizzard-like RTS like Stormgate.

A Good Idea

Mutual random is one of those ideas that makes so much sense when you learn about it that it’s hard to believe that it never shipped with any of the mainline Age of Empires games. It’s so popular that back when I did my first 1v1 ladder grind, I frequently got asked why I was civ picking - even though I was picking the worst civs in the game.

I think it works for a couple reasons. One, it’s somewhat impractical to fairly balance the 42 different civilizations in Age of Empires II. If we allow ourselves some time for game theory, we can see how this makes players reluctant to pick weaker civilizations by choice, because they assume their opponent will pick one of the meta civilizations. This is a self-fulfilling prophecy, because if they do happen to come across someone choosing a weaker civ, that person will now be inclined to pick a meta civilization on their next go around. On and on this goes until it’s Set mirrors on Watering Hole.

Mutual random fixes this up by guaranteeing both players the luck of the draw. Sometimes you get a good match-up, and sometimes you don’t - ultimately, because the strategic skill ceiling is high, you end up thinking more about the match-up and the map gen and how to play it out than you do about balance or design. And I think that’s a really good thing for the experience.

Two, Age of Empires II’s approach to asymmetricity is a good complement to mutual random. Civilizations share a common technology tree, so it’s not so difficult to switch from civ to civ to civ in a single session. But, civilizations are still sufficiently differentiated - through intentional unit and tech differences, unique bonuses, unique units, etc - to make playing random a meaningfully varied and interesting experience.

A great benefit of all this is that the developers of Age of Empires II can more freely try larger balance or design changes, because absent a really weak or really strong civilization, mutual random ensures that most people won’t be terribly affected by imperfect game balance.

… but would it work in a purely asymmetric game?

I think all things being equal, giving players more options is a good thing. Mutual random is nice because it’s an opt-in feature that doesn’t affect anyone else’s experience the way a new game mode might shard the ladder or a new gameplay option might offer an unfair advantage. You could just add it as a checkbox, which doesn’t seem like a whole lot of work, and let players decide for themselves if that’s something they want.

But having said that, I think it’s important that developers be opinionated about the way their game should be played, because I think it makes it more likely that players will enjoy it. And so instead of just saying “here’s a checkbox, feel free to click it”, I think we should ask ourselves whether mutual random is likely to be a good experience.

In purely asymmetric games, it’s not just the units or the tech trees that are asymmetric; it’s everything. Power spike timings are different. Ways of dealing with particular problems, like early anti-air or detection, are different. Comeback mechanics are different, basetrade mechanics are different. I would argue that even the way you think, the way you formulate the game state in your head, is different from race to race.

I think people generally underrate the degree of asymmetricity in games like Age of Empires II, because they don’t understand them very well. But I think it’s fair to say that it’s a materially different type of asymmetry than games that are designed from the ground up to be purely asymmetric; especially Blizzard-like RTS games on static maps which offer hefty rewards to sharp, specialized timings.

If you were to add mutual random to StarCraft II, for example, I think you would initially increase the number of cheeses and all-ins that you see on the ladder in a way that never happened with Age of Empires II, because with some exceptions that game doesn’t really have cheese. And I think this would happen because the races are so different that it’s quite hard to play them “correctly” without a lot of experience and practice, so a lot of people would just default to YOLO’ing across the map.

But I don’t think this would necessarily be a stable state. Just like the ladder starts out aggressive and trends toward defensive macro when a new patch comes out, I think the same would happen with mutual random, too. People would start to figure out how all the races play and learn how to defend cheeses and play more sustainable macro styles, because ultimately if they’re rolling random, so is their opponent. And while I think the mindset shifts would be a bigger hurdle than Age 2, I think people would become accustomed to it surprisingly fast because they know their opponent is dealing with the same problem.

I think one of the benefits of this is that it would lessen some of the racial tribalism I’ve observed in StarCraft ever since I started playing. One of the edge cases of people identifying with and cheering on one particular race is that it can contribute to feelings of animosity and “other-ness” toward other races. I can’t be the only person who’s noticed that balance and design discussions in StarCraft often devolve into accusations of bias based on the race that people play, right? When this happens, the game’s design starts to feel zero-sum, which I don’t think is the intent, nor do I think is the ideal way of enjoying the game. Allowing people to play random against one another would build empathy toward all of the races and allow people to take a step back and enjoy the game as a whole, instead of through the lens of one particular race.

… but could it really work?

I think I’m fine enough with the idea of mutual random working in a self-contained way; I think the risks are low and the upsides significant. I would guess that most ladder players who opt into this will end up enjoying Stormgate more than they otherwise would have, because I really believe an RTS is more fun when you perceive it as a whole game instead of from the perspective of one race.

I think the more challenging question is how this would work against players that specialize. Specialization is pretty advantageous in purely asymmetric games - sharper timings, sharper reactions, sharper everything, plus something of an unconscious mind meld with the philosophy underpinning the race. And in a game with a handful of very different races, a lot of players will choose to play just one.

Now - mutual random is opt-in, right? But odds are that given the much tighter balance constraints of an asymmetric game with a small number of races, there won’t be “meta races” to fall back to. Mutual random players would probably just fallback to whatever race they happen to prefer, for which they’ll have substantially less experience than the specialists they’re matched up against, which is going to make it increasingly harder for them the closer they get to the top. Which means that absent a competitive rule change or a fundamental design shift toward breadth of strategy and general skill, the top of the ladder will continue to be dominated by specialists.

I’m not sure how much of a problem this is; it’s the sort of thing that I feel should just “be handled” by a properly working MMR system, the same way it handles people being particularly bad at certain match-ups, or people trying new builds, or people having a bad day, or all the other types of natural unpredictable variation that happen on the ladder. More broadly, the way any RTS game is played at the top of the ladder is worlds apart from how most people play it - but that doesn’t invalidate everyone else’s experience, right?

And I guess that’s worth highlighting. Taking StarCraft II as an example, for most of the ladder - say, 80% of it - specialization has its limits. You can get well into Platinum and even Diamond before you start seeing proper build orders and timings, and even then they’re not necessarily correct or well-executed. I think many people don’t play the game deeply enough to necessarily benefit very much from specialization, but they do specialize enough to eat the cost of tribalism and narrowly perceiving the game’s balance and design through the lens of only one race. I think, for at least some of these people, mutual random would open up a fun and new experience.

Final Thoughts

I was surprised by how popular mutual random turned out to be for Age of Empires II. It really levels the playing field in terms of balance, and allows players to appreciate the broader complexity of the game instead of getting bogged down in design or balance complaints. The biggest aspect of this is empathy - players see the game wholistically because they have personal experience with all of the races. I think the opt-in nature of a feature like this means it could be a cool differentiator for Stormgate’s ladder. I hope the developers give it a shot!

Until next time,
brownbear

If you’d like, you can follow me on Twitter and Facebook and check out my YouTube and Twitch channels.