A Different DOTA: Forgotten Game Modes of Warcraft 3

Joe Köller
Nov 19, 2017 · 15 min read

The rise of MOBAs as one of the dominant esports formats has brought a lot of critical attention to the origins of the genre, a history ranging from the Starcraft map Aeon of Strife to the Warcraft 3 map Defense of the Ancients and early commercial flops such as Demigod and Heroes of Newerth. I never got to play any of these early progenitors myself: Without a decent internet connection, my own experience with Warcraft 3 revolved around LAN matches with friends and the custom maps that came on magazine DVDs. So when a recent fit of nostalgia took me back to the game, I decided to use the opportunity to explore the custom map scene in more detail.

Returning to the game now — as an adult with a broadband connection, an appreciation for MOBAs (mainly League of Legends) and a brain easily overwhelmed by these hectic games — I quickly realized that I was less interested in the direct predecessors of today’s titles than in their forgotten contemporaries, adjacent maps that took the concept in different directions and experimented with other modes of play. Looking at these different types of early MOBAs, my mind immediately began itching with questions, wondering how the genre might have developed if a slightly different map had emerged as the most popular variant a decade ago. I started downloading as many maps as I could find and testing as many as I could get to run.

To reiterate, I’m not interested in retracing the lineage of DOTA for the hundredth time, rather the purpose of this post is to approach the genealogy of the MOBA genre from an evolutionary perspective, to consider some of its stranger branches and offshoots and speculate about an alternate history of game design in which the genre developed in a different direction than the one it ended up taking. With that in mind, let’s look at some of DOTA’s contemporaries, starting with maps that are relatively similar to our our current understanding of MOBAs and then moving towards the outer edges of the genre.

Screenshot from Footmen vs. Grunts

Footman Wars/Footman Frenzy/Footmen vs. Grunts

Maps such as these have existed almost as long as DOTA itself and were one of the most common variants of the nascent MOBA genre. The main difference to the original is that while DOTA gives you control of a hero and automates all other units on the battlefield, these maps give you control of both a hero and the surrounding troops, to varying degrees. Some versions of this game mode let you micromanage each individual unit, others only allow you to point them in a general direction, to tell them whether to retreat or advance. Unit production is still automated, but you can spend gold to unlock upgrades and different kinds of units that will spawn instead of/alongside basic footmen. This introduces an additional layer of team coordination, lest all players on one side invest into melee units that are countered by your opponents’ flying and ranged troops.

As harsh as it may sound: in itself, this branch of the MOBA genealogy is largely unremarkable. Since most of these maps still focus on using your hero and their abilities, moving and upgrading basic units ends up feeling more like a distraction than an additional tactical layer. Many creators compensated for this extra work by reducing the overall complexity of their map, resulting in a game mode that does a little bit of everything, but not all that well. However, this variant is incredibly important from a historical perspective because it forms the link between prototypical MOBAs such as DOTA and a variety of other maps that honed in on the new ideas introduced by Footman Wars and others, namely the concept of automating unit production, but putting the player in charge of choosing unit types and upgrades.

Castle Fight

This game mode appears to have developed sometime around 2007. It splits players into two teams and puts them in control of a worker who can place various buildings. Different types of buildings spawn different types of units, which automatically march down the middle of the map (the number of lanes varies) until they reach an enemy. Units also have different types of armor and deal different kinds of damage, so the main focus of the goal of this game mode is to observe what your enemy is sending down the lane and coordinate with your teammates to counter it effectively. Considering the number of available units and factions, however, things can quickly get quite chaotic, especially if a lot of players are involved.

Discovering this game mode ten years after its inception, the most fascinating thing about Castle Fight is the way it re-contextualizes certain aspects we associate with modern MOBAs.

First, there is the notion of picking appropriate counters to your opponents’ units. You can also see this component in modern MOBAs, but in the form of a single choice players make at the beginning of a match when they pick their hero. In high-level play, this pick-and-ban phase is an important opportunity to make strategic decisions, to secure your preferred team composition and disrupt the enemy team by banning or stealing important heroes. On a more casual level, however, the permanent nature of this decision can feel quite punishing.

If the hero you choose ends up having a bad match-up against your lane opponent, you basically just signed up to have a miserable old time for about 20–40 minutes. Even if you already know what your opponent is going to play and you know which hero works well against it, who is to say you know how to play that particular character? The expectation these games and their communities put on players to be both strategically flexible and mechanically proficient, by playing a lot of different heroes and playing all of them well, is a massive factor in the steep learning curve associated with MOBAs.

Given this, it’s interesting to see how Castle Fight allows you to respond on the fly to the unit composition the enemy team is sending down various lanes. If the first unit you produce proves ineffective, it does put you on the back foot, but it doesn’t put you at a permanent disadvantage. Just tell your team what’s going on and put down some different buildings in response.

Second, there‘s the way these games handle snowball mechanics. One way or another, all games on this list are about building an advantage big enough for it to start compounding on itself and grow into an insurmountable lead. In modern MOBAs, this can be achieved a number of different ways. On an individual level, one player may manage to kill their lane opponent, granting them gold, which they can invest in new items that make it easier to kill them again. On a team level, players may “feed” one particular teammate by helping them get kills or giving them free time to farm their lane.

However you do it, once you gain an advantage, it cannot easily be taken away from you. You don’t lose items or levels when you die. The enemy team may itemize defensively in response or eventually catch up to you in experience, but these things take time and careful play. Once you’re in the hole, a single mistake can be enough to dig yourself in even deeper. Modern MOBAs have some catch-up mechanics — from experience bonuses for underleveled players to the map layout, which gives defending teams the inside track for switching between lanes — but these are more suited for small course corrections than turning around an entire match.

These games sell a very particular fantasy of power and agency, so they always want the losing team to have a small chance to come back, but more importantly, they want to reward the winning team for their good play with noticeable advantages. Depending on what side of the equation you are on, that can be very satisfying, but it can also be very frustrating, especially if your opponents gained their lead in a fight on the other side of the map you weren’t even involved in.

Castle Fight matches snowball in a more literal sense: depending on how close to your base your units run into enemies and how much damage they do against your chosen armor type, groups of units can start building up in a lane. Once a bunch of units are together in the same place, they can easily deal with the individual enemy units marching towards them single file. They stop being a real threat and turn into an obstacle, delaying your advance long enough for more and more units to be rolled up into this ball until it finally crashes into the enemy base to wreak havoc.

The waves of units that charge at each other in modern MOBAs can build up in similar ways, but because they will run into either an enemy hero or one of the defensive towers along the lane, they are less of a threat and more of a resource. Knowing how to clear, trim, freeze, push, crash, deny, bounce and otherwise manage waves helps you farm gold and create map pressure, so these skills are fundamental to success, but they don’t directly win you the game. In Castle Fight, however, there are no heroes or towers to stop a push before it gets rolling. By the time it reaches your base, it can mean a lot of trouble, especially since the buildings that spawn your own units could be destroyed, putting you even further behind.

A rescue strike in action.

Because the lanes in Castle Fight are much more explosive than in traditional MOBAs, players are also given much more explosive countermeasures: once per round, every player can use a “rescue strike” — a devastating ability that destroys all enemy units in a large radius and grants you gold for each unit hit. Ideally, you coordinate with your team so that one person uses this ability when the enemy units are entering your base, just before they split up to attack different buildings. Once the bulk of your opponents’ forces are removed, your own troops, having a shorter distance to cover to get back in the action, will be able to group up and push back the other way.

In terms of catch-up mechanics, this one is a pretty bold choice. We’re talking about a game mode in which having more and better units is the only advantage you can ever build, and yet that momentum can be denied or even turned back against you in an instant. It’s not a perfect system by any means, but it’s certainly interesting to look at such an extreme solution compared to the relatively small counterbalances used by modern MOBAs. In combination with the fact that most games of Castle Fight consist of multiple rounds (the map will automatically reset without sending players back to the lobby), it gives matches an almost Tennis-like sensibility: both sides are expected to do well when they serve and you need a certain number of upsets to really get ahead.

Overall, the notion of influencing lanes only indirectly through upgrade and unit choices represents a drastic departure from DOTA and its focus on powerful player characters with direct agency, but the two branches still share a lot of common traits. We already discussed the focus on team coordination and countering the enemy composition, but even the builder unit players control retains many of the characteristics we associate with DOTA heroes. They may not be much of a fighter, but it still takes a lot of micromanagement to use them efficiently and place new buildings as quickly as possible before going back to repairing any that might be damaged.

Also, I didn’t know where else to fit this in, but there’s a Civilization-themed version of this game mode in which you upgrade units as you advance through the ages. I couldn’t get it to run though.

Wave Manipulators

This is a bullshit term I made up myself, so allow me to explain what it means: I propose this term to describe maps that hone in entirely on the concept of influencing lanes of autonomous units through passive upgrades (and maybe a few small active abilities). These maps do away with the need for team coordination or extensive micromanagement and instead ask players to focus on reading the map, anticipating problems and carefully timing their response in order to gain an advantage.

I choose this term partly because it’s a really cool name, but also because this branch never fully developed into a recognized map type. There is no prototypical version, no iconic and widely imitated namesake I could point to in this case, only a few sporadic examples that illustrate the potential of this game mode. Of these, the most sophisticated map I could find and the one we will focus on here is called Survival Chaos. This map has been around since 2011 and is still being updated today. There is evidence that the game mode itself is even older, since the creator of Survival Chaos cites another map as its inspiration, but I was sadly unable to track that one down.

Survival Chaos is a map for four people. Players are randomly assigned to the center of the left, right, top and bottom sides of the map and given a small base with a four defensive towers, one fortress and three barracks that automatically spawn waves of units and send them towards each of your three opponents. Lanes run along the edge of the map towards the players on either side of you and across the map towards the player directly opposite you, with an intersection in the middle that tends to turn into a big stalemate.

And that’s all there is to this map. You don’t move around any units, you don’t even build any new buildings, all you have to do in this game mode is pick the right upgrade for your current situation. At the start of the game, you choose a faction (or choose to play with random factions) and a passive bonus for that faction, such as stronger towers or additional units for one lane. Once the game begins, you can spend gold to upgrade all sorts of things. Your barracks can be upgraded to unlock additional unit types in that specific lane, your towers let you unlock special abilities for specific unit types and your fortress contains general upgrades for damage, armor and passive gold gain. It can also be upgraded itself to unlock defensive spells and hero units (which still act autonomously).

So with all these upgrades, how are you supposed to know which ones to get? Well, you can pro-actively invest into early aggression or a strong economy, but most of your decisions in this game mode really come down to keeping track of your opponents and responding to their actions. If an enemy is pushing towards you, you’ll probably want to upgrade the barracks in that lane to turn them around. However, you don’t necessarily want to do that right away. If a large wave has started to form, a few extra units here and there aren’t going to stop it. Instead, you generally want to let your opponent push all the way towards you and upgrade your barracks right as their troops reach your base. That way your buildings can soak up some of the damage, your towers can help out your units and your units can start grouping up and pushing the other way themselves.

The problem with timing this kind of counter-push as efficiently as possible is that you need to have gold ready for the upgrade when the moment arrives, but it would be a waste to set aside a bunch of money when it could be spent on other upgrades. Ideally, you want to spend as much gold as you can get away with, while making sure that you will earn back just enough, just in time to upgrade your barracks. That means estimating how fast the enemy will push, when they will arrive at your base, how much gold you will earn in that time — a level of foresight that takes time and experience and develop. The more you play this game mode and observe and analyze the ebb and flow of its lanes, the further ahead you will be able to plan. Your analytical skills go from being able to make an educated guess as to how long it will take a push you can already see to reach your base to having an intuitive understanding of how long it will then take your own counter-push to reach the enemy and trigger another push in response.

If you had only a single opponent to deal with, this would all be simple enough, but the real challenge of Survival Chaos lies in its free-for-all approach. Fighting against three other players doesn’t just mean planning on the fly in order to respond to pushes in three different lanes, it means doing so while trying to figure out which of your opponents is currently the biggest threat, based on limited information. You can only see what your units can see, so you don’t know a lot about what’s going on on the other side of the map. This makes it very important to pay attention to subtle clues, such as the notifications when a player captures a neutral building along one of the sidelanes, and more obvious tells, such as the notification for another player losing one of their buildings. At times, Survival Chaos can almost feel like a board game, where temporary truces are formed to deal with whoever is currently in the lead. This also tracks insofar as the map can feel quite unfair when you are targeted by multiple people at the beginning of the game.

Survival Chaos presents us with a radically different vision of the MOBA genre: it maintains the emphasis on careful observation and strategic planning, but removes the need for mechanical proficiency. Instead of charging into the thick of battle, players influence lanes and control the map from inside their own base. The real question for me, however, is how much further these qualities could have developed in a world in which a map like Survival Chaos had become the most popular variant of the MOBA genre a decade ago. In a world where our prototypical understanding of MOBAs is as a slow, cerebral game of observation and planning, what would the innovative new takes on this genre even look like? What would the world of esports look like? Can you imagine a stadium full of people, not cheering for ten players clicking 300 times per minute, but four players bent over a battle plan, stroking their chins?

The MOBAs That Could Have Been

As much as I enjoy thinking about the potential differences a small change in the history of this genre might have made, it would be too simplistic to suggest that these game modes are interchangeable and any one could have emerged as the dominant variant. It’s not difficult to see why DOTA, or at least a map like DOTA, was the version that popularized this genre and shaped our understanding of it: of all these slightly different branches, DOTA puts the most emphasis on direct agency and systemic complexity and mechanical mastery, qualities that gamers have been taught to prize above all else.

And these qualities can be legitimate design goals, but the problem is that our enjoyment of them is always at least partly aspirational: you might find yourself playing a strategically complex and mechanically demanding game not because you like the game, but because you like the idea of being good at it. Our intrinsic desire for self-improvement has become entangled in an industry that sells hollow progression and a culture that idolizes competition, and those who win in it. As a results, most modern MOBAs, being free-to-play games, try to sustain a broad playerbase to finance the game, while also supporting an esports community that demonstrates the competitive viability of the title. However, these efforts often feel at odds with each other: frequent updates, additions and balance changes keep the competitive scene from growing stale, but do the majority of players really benefit from there being quite so many changes to keep up with, on top of mechanics to learn and heroes to master?

At least, this has been my experience with the MOBA genre. I’ve always liked MOBAs on principle, but I get easily overwhelmed while playing them and end up feeling more self-conscious than entertained. So far, I have resolved this dissonance by watching competitive League of Legends games instead of playing the game myself. I resigned to seeing myself as an atypical MOBA fan, in part because I couldn’t imagine what a MOBA I would really enjoy might look like without envisioning a slightly worse version of existing games. Now that this look at the genre’s different branches showed me what could have been, I can’t help but wonder how many people among the millions and millions of MOBA players would have a better time with a slower, less complicated and less demanding variant such as Survival Chaos.

And that’s why I can’t stop wondering: What if the MOBA genre had developed a little bit differently?

Joe Köller

Written by

I used to write about games sometimes, now I make video essays I guess?

More From Medium

Also tagged Videogames

Also tagged Games Crit

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade