I have been working for some months on a tactical RPG, and the question of pacing multiplayer fights was one of my first tasks. Unfortunately, it looks like the subject “pacing multiplayer games” is not really discussed (I have searched the GamaSutra archives but didn’t found anything).
So, here are my thoughts, based on other games.
The ideal curve
The ideal tension curve for a multiplayer combat follows the traditional curve of pace / tension / narrative stuff, composed by a sum of small rises and declines that go into a great ending climax.
This is the only curve we have to know :-)
This curve is very classical: it is basically the same we find in every solo game. In a multiplayer game, each rise corresponds to a decisive clash, while declines are moments where players retreat to lick their wounds or manage secondary conflicts.
In a solo game, this curve is controlled by different means: story, screenplay, difficulty, etc. All these techniques are predefined and relatively easy (*cough*) to handle by designers.
In a multiplayer game, these points are not necessarily available and we have to find other ways to obtain the same results. Taking references from competitive multiplayer games, I have tried to extract the most used techniques to create that pacing curve.
Evolution with player’s abilities
During a game, each player’s abilities get stronger and more intense. This evolution assures both a renewal of the game experience and a constant growth of intensity.
Player’s abilities increase during the game.
Pace and intensity follow the player’s abilities.
Here are the three main ways to improve the player’s abilities:
The panel of strategic choices widens. Players have more alternatives in their actions. If some strategies counter other strategies, this can create really interesting mind-game dynamics.
The effectiveness of these strategies. New strategies are more and more powerful. This point is a prerequisite to push players to adopt new strategies.
The general power of the player’s avatars. As it implies no strategic or tactical change, this is probably the least interesting method. An important point to take note of: during their progression, avatars should always earn more destruction power than survivability (evolution should enforce the Entropy, see below).
Example - Counter Strike:GO
The Shop of CS:GO
Each round, the player earns money they can spend to buy weapons.
Weapons are to be considered as strategic options (as in “a submachine gun gives tactical advantages that differ from a shotgun”). By accumulating money, the player widens their panel of strategic choices.
Also, there is a power rank for the weapons (assault rifles are deadlier than simple pistols, etc.) A rich player has more effective strategies at their disposal.
At last, the bullet-proof armor directly increases the general power of the player’s avatar, independently from any strategic consideration.
Example - StarCraft 2
During a SC2 game, the player develops their economy and their tech tree. Economy allows the player to create more and more units. Technology unlocks stronger and stronger units.
Units are to be considered as tactical choices. During the game, the player increases both the number of his strategic options and the effectiveness of these strategies.
Evolution with steps
Generally, players’ abilities don’t evolve linearly but are marked with steps. The player becomes suddenly more powerful because he unlocks a new skill, gets a new weapons, etc.
These power steps are interesting for different reasons:
They break linearity, establishing pace and creating tempo.
The player has a big incentive to take action at that moment.
If steps are predictable, an experienced player can adapt their strategy according to the steps of their opponents. That creates a very interesting layer of yomi.
In a confrontation game, we have to compare the curves of the different players. Steps are generally not placed at the same timing, because of the asymmetrical design of units/races/character’s class/etc.
Steps bring moments of tension, during which one player has a temporary advantage.
If we compare the two curves, steps in the power curve correspond to peak in the pace curve.
Example - HearthStone
Flamestrike is a mage card that deals enough damages to clear the opponent board after turn 7. It means turn 7 can be considered as a step up in a mage player’s power curve.
Furthermore, a seasoned opponent can predict and counter Flamestrike by avoiding playing creatures at turn 6 (or playing only creatures that can survive FlameStrike).
Example Heroes of the Storm
In HotS, when a team reaches level 10, all its players unlock their ultimate ability; the team gets a huge power leap at that very moment. This rule dictates a specific pace: if your team has reached level 10 and your opponent hasn’t, you have to attack because you are sure to win any team fight.
Some strategies have emerged to use this mechanic, for example getting to level 10 before your opponents at the very same moment as a strategic objective spawns on the map. These strategies and pace peaks are indirectly generated by the notion of steps.
Here, the blue team making the most of its level advantage to create a pressure the red team cannot contain.
Evolution of resources
In some games, the evolution of a player’s abilities relies on one or more resources. In the previous examples: money for CS:GO, Mana for HearthStone, minerals / vespene gas for SC2.
Generally, the player harvests these resources and spends them to acquire power. So, there is an indirect link between resources and pace curve, and that is why it is important that resources grow during the game (to ensure a growth in power = growth in tension).
Depending on the games, the player can influence the acquisition speed of resources.
From the previous examples, HearthStone has the simplest curve:
The player’s abilities increase linearly with his mana pool.
The Mana pool increases each turn.
Two classes of characters can affect the evolution of the curve: the druid can win Mana points and thus accelerates the curve; the warlock can sacrifice some Mana points (to play some more powerful creatures) and thus slow down their curve.
When changing the Mana curve, players are taking risks that create tension peaks.
Except for these two characters, the resource curve in HearthStone is strictly linear and independent of the player’s skill.
CounterStrike : Global Offensive
CS:GO offers a little more complex system. Players earn money mostly by taking down opponents. Thus, skilled players get more money than their opponents; with this money, they can get powerful weapons that makes them more able to kill their opponents.
This is a super slippery slope, so designers have integrated a second rule. The amount of money also depends on the weapon used to kill the enemies: the weapons that are the riskiest to use give you the most money (killing a guy with a knife makes you a rich man while sniper rifles give near to nothing). This rule serves as both a risk/reward mechanic and a way to reduce snow ball effects.
The whole system is interesting as it adds strategic choices to the game. During tournaments, it is pretty common to see a team save up its money for a round (and lose) to ensure it can purchase better equipment during the next round. These moments also create tension peaks.
StarCraft2 offers another system. In SC2, player’s abilities depend directly on the number of harvesters and bases. The SC2 economy is built on a notion of investment: a player who spend their money on harvesters will get stronger abilities later in the game but they are immediately vulnerable.
The player’s abilities in SC2 rely mainly on strategic options (short term vs long term) ; making theses choices creates tension.
The list of examples above is far from exhaustive.
Controlling the pace of the game thanks to a resource evolution is super interesting, as it is transparent for the players and it can bring specific strategies. However, this method can easily turn into a slippery slope, so one needs to be careful when using it.
Evolution of Level Design
Level Design can also influence the pace of the game. Two main ways of doing it:
The removal of Safe Zones
The creation of Key Zones
The removal of Safe Zones
Each multiplayer map is composed of Danger Zones (in which players die more often) and Safe Zones (in which players are less exposed and, err… safe).
Depending on the game, we can measure the danger more or less easily: for a FPS, some tracking devices highlight the places where characters dies. In that type of games, the pressure is directly linked to the presence and the size of Safe Zones (more Safe Zones equals less pressure).
Example extracted from Halo 3.
Deadly areas are highlighted in red.
Some games remove Safe Zones from Level Design during the game. That’s the case of MOBAs. In a MOBA, Safe Zones are highly defined by the turrets. Turrets are progressively destroyed, reducing the Safe Zones to nothing.
Map of Heroes of the Storm : Sky Temple
At the beginning (1), a blue player can move freely in the blue Safe Zone.
During the end game (2), every turret has been destroyed, thus Safe Zones barely exist anymore. Every part of the map is dangerous.
Some interesting points about MOBAs:
To win, players have to reduce their opponents’ Safe Zone.
Thus, to destroy the furthest turrets, they have to go far away from their Safe Zone.
The evolution of the Safe Zones depends entirely on the players’ actions.
Each turret’s destruction is a step in the pace curve.
Each bump corresponds to a turret’s destruction.
The creation of Key Zones
Some part of the map are more important than others (because they contain objectives, power-ups, etc.). In the rest of this article, I will refer to them as Key Zones.
Key Zones generally constitute dangerous parts of the map, but the opposite is not necessarily true: dangerous areas are not necessarily important (as an example, a full-open space can be a very risky place in a FPS, even if doesn’t include any objective).
The importance of each part of a map is therefore measurable with tracking tools.
This map of CoDMW2 (aka Call of Duty Modern Warfare 2).
Red is not death, but activity (= time spent at a point).
In some games, the Level Design is static and Key Zones are predefined.
But in others, new Key Zones spawn and disappear during the game. This is far more interesting as it creates tension peaks (generally, players are attracted by Key Zones). For instance, in multiplayer games, players are driven to fight over them.
There are basically two ways to handle these modifications:
Arbitrary: evolutions are predetermined by the Level Design. We, as designers, have a very good control on the course of events. With this method, we can create the curve we want.
Organic: evolutions depends on the players’ actions. As this method leaves some pace control up to the players, the curve is far less accurate and detailed.
Example - Heroes of the Storm
In HotS, objectives appear regularly on the map. Taking objectives is crucial as it generally deals tons of damages to enemy turrets. The spawn of any objective is often followed by team fights.
The blue circle is a Key Zone:
when a player stands on the circle,
the pylon shots on enemy turrets.
In HotS, the Level Design dictates how and when Key Zones appear. Each of them provokes a temporary rise of pace.
Example StarCraft 2
In SC2, the harvesting of minerals and gas tend to exhaust them. That forces the player to build new bases and harvest resources further away. If we consider bases as Key Zones, it implies two things: while they look for new bases, players are pushed against one another (which is a source of tension). Also, the territory they have to defend progressively become bigger and bigger (it also increase the tension).
Here, the progressive exhaustion of bases push players against each other.
In SC2, the modification of the game space depends on the player’s action. The spawn of Key Zones doesn’t depend on some arbitrary rules but on the economical progress of players. This dynamic ensures a constant rise of the tension. However, it is far less precise and doesn’t provide tension peaks as the previous example.
SC2 pace curve: a chaotic one that depends on players’ action.
Comebacks & Entropy
In multiplayer games, pace also relies on two main notions:
The constant possibility of a comeback.
A player should always have a slight chance to win, even if they failed at the beginning of the match.
The existence of an entropic dynamics.
The stakes must increase from the very beginning until the gg.
These two themes have been largely detailed in that super good post from Hannes Rince. Even if these concepts don’t necessarily need new mechanics, we should still check each of them as they can mess up the pace curve pretty badly.
Let’s try to summarize a many pages post into three points (*cough*):
A game shouldn’t be played if a player is sure to lose.
Stakes should always rise with time.
Any advantage a player can get at the beginning of a game should stay marginal, or should bring the end of the game.
Example - Mario Kart
In Mario Kart, stakes are rising during the race. If the player has an accident during the first turn, he can always come back. On the contrary, falling or having an accident during the last turn generally brings defeat.
Example - Heroes of the Storm
The sanction for dying in Heroes of the Storm grows with the time. When a character dies during the first minutes of the game, they revive pretty quickly and it doesn’t give much experience to the opponent. During the end game, a dead character has to wait more than one minute to revive, which is generally long enough for the opponent team to take the win.
There is also a risk when stakes are too high. Too much danger can create a turtling dynamic during the end game (the SC2 community reminds with pain that 2h+ game ; players were afraid to fight for 120 minutes).
Turtling has a big effect on the pace curve.The tension curve, that until now followed the danger curve, dissociates into two:
the danger stays high, with high stakes.
tension and interest are reducing, slowly but surely.
Turtling separates tension and danger.
To create a tension curve in a multiplayer game, we have differents means:
Players have to unlock new strategies. These strategies should be more effective and/or their avatar should become more powerful with the passing of time.
Steps in the progression curve are a good thing. Especially if they are predictable for the opponent.
If as a player you need resources to progress, you should need more of them the more time passes. If these resources can be influenced by the players’ actions, one should avoid any slippery slope dynamic.
In Level Design, Safe Zones have to be cut down over time. The game needs to be more and more dangerous.
Regularly, some Level Design elements can appear to create Key Zones that in turn generate tension peaks.
The underdog should always have a chance to come back in the game.
Stakes have to increase during the game, without pushing player into turtling.