Disclaimer
This is a long post. It addresses some core game design theory. It provides examples. It attempts to lay out a case regarding PVP and PVE and whether or not those two can be balanced together. If you don’t want to read a long post, move on. Otherwise, I welcome honest thoughts and constructive feedback.
TLDR
I don’t believe PVP and PVE can both be balanced in the same game without balancing around PVP as the primary design goal. Diablo is a PVE game. If you want the nuances, read further.
I’ve read many posts from people asking for PVP in Diablo 4, most from old school players who want a return to D1 and D2 days. Frankly, I have my suspicions about their motives, but because it is a common request, I thought about it for a while. How do you balance it? How do you make it sync with the PVE part of the game? What makes a good PVP game?
Fairness
Successful PVP games are fair, but fairness is a somewhat nebulous concept. Defined here, it means that each player has the same opportunity to win and the same tools at his disposal. When this is the case, the skillful use of the game mechanisms decides who is the winner. Think about checkers or chess as a good example. Each player has the same number of pieces which can make the same moves arrayed in a mirror image across from each other. From a developer’s standpoint, if they make changes, they make them equally to one side and to the other, and then, they only have to match players based on their skill levels to ensure that each player has an equal chance to win.
A Good Gaming Experience
I think it’s probably important to take a moment to define this as well. The term gets thrown around often and it’s usually taken to mean “I subjectively enjoy the game,” and that is a very difficult standard to define concretely. Psychology tells us that children will continue to play a game as long as they have between a 30-70% chance to win the game. Less than that and they see playing as futile and become discouraged. More than that and they get bored because the game is too easy.
While this is helpful, in practice, modern gamers are a little more demanding. We have statistical logs, either in-game or via 3rd party websites. We’ve come to expect MMR or ELO skill-ranking and matchmaking systems as industry standard for a PVP game, and we expect them to be good enough to provide us “fair” matches against equally skilled opponents. We measure that by expecting a 50% win rate, meaning you have an equal chance of winning or losing, and we tolerate maybe 1-3% variability with that, skewed towards winning. By that I mean, no one complains if you’ve got a 60% win rate, but you’ll hear howls from those with 40% win rates. And of course, because these systems are designed around a Bell curve, for every player getting a 60% win rate, there will be someone else at 40% whose very angry about it, so statistically we want as tight a peak on that Bell curve as we can get so the maximum number of players feel that they have an equal chance to win.
Further, we’re sensitive to winning and losing streaks. This is probably a result of the gambler’s fallacy that states that a streak of bad outcomes increases the chance that your next game will be a win. If we start losing 5-6 games in a row, we start suspecting the system is rigged against us and making us lose. While streaks are not avoidable in a matchmaking system, they happen and players will react negatively to long losing streaks because we perceive the immediate past much better than long-term trends. Consistency of experience needs to be a design goal.
Another issue is timing. In a boxing match or MMA fight, both fighters are called to the center of the ring then given the signal to fight from the referee. It is illegal to hit an opponent before that signal or after the bell has rung. You have to be ready for the fight. I point this out because much of the PVP experience in many games is ganking. You specifically pick a weaker target, hunt it, and then strike when it is preoccupied with something else. That is positively infuriating for the player getting ganked. Any fair PVP system needs clear rules for players to willingly engage in PVP combat and to declare they’re ready for a fight.
Team dynamics are also important. Diablo 3 allowed teams of four to group together. D4 will likely do the same. But 2 on 1 fights aren’t fights. They’re grossly unfair. Solo PVP systems have to be separated from team PVP systems, and teams have to be matched based on number of players. Otherwise, you’ll have packs of friends grouped up and coordinated on Discord or some other voice chat going around ganking smaller parties or individual players. As we’ve established, getting ganked isn’t fun for the guy getting ganked. Ganking is a tactic specifically designed to take minimum risk or skill on the part of the gankers and to exploit that unfairness. If teams enter combat, they must be paired against other teams with the same number of players and the same skill level of those players. Calling in your grandmaster friend to beat up on brand new players isn’t fun or fair.
To provide a good gaming experience in a PVP game, many factors need to be assessed and balanced: player skill, the toolkits available to the players, the number of players on each side, and the timing so both sides are aware when to fight and when to hold fire.
Complexity of Uneven Player Toolkits
I started with the example of checkers or chess, but PC games aren’t so simple. Warcraft 1 and 2 had two different factions with two different toolkits. Starcraft had 3. Warcraft 3 had 4. WoW’s PVP experience has a seperate toolkit for each class and each specialization within that class for 36 seperate toolkits. HotS now has over 80 different heroes you can play. Further, they balance both 1v1 and various teams. At release, we know D4 will have 5 classes.
But it’s more than balancing just 5 classes. We have to consider the various power systems that will be in play. A lvl 20 player will beat a lvl 10 player every time. Most games with levels are designed around giving substantial power boosts at each level to make them feel meaningful. Power growth is logarithmic, not linear. The difference in power between a lvl 30 and lvl 40 character may be an order of magnitude. While that’s perfectly fine in a PVE environment where you just add a zero to monster hp pools and damage, in PVP, it means the lvl 30 player always loses, and never has fun. And that’s not the only power system. Diablo revolves around gear as a power system, and it presents the same logarithmic power gain as level systems. A player perfectly geared in min/maxed end-game gear will always beat someone in leveling gear who just hit max level. Sockets, enchants, charms, or whatever systems they decide to implement will all add complexity that has to be balanced.
When there are only a few different toolkits to balance, it is possible to achieve close enough to parity that they effectively function like checkers or chess, and thus games will be decided based on skill. Starcraft 1 is an good example. However, the more variables you allow, the less individual player skill matters, and the more the various other power systems within the individual toolkits become the deciding factor, as we illustrated with levels and gear.
To provide a good gaming experience here, you have to finely balance the individual toolkits available to players so that skill becomes the deciding factor, but you also have to finely balance the various other power systems you add. Your matchmaking algorithm now has to calculate player power and player skill and match on two parameters instead of just one. And it’s not a small task. The algorithm needs to hit a very high standard for accuracy as we’ve seen above in order to provide a good gaming experience. But now, we’re asking it to take into account the huge range of variability in player power which grows on a logarthmic scale, in addition to skill. Ideally, you’d remove either the gear or the level power systems or both to remove the complexity and make balancing easier.
Reconciling PVE and PVP in the Diablo universe
It is important for Diablo players that it retain the core elements that made it what it is. It is based on tabletop D&D, where players choose a class (toolkit), and then go out into the game world on adventures or quests, killing demons, scrounging items from the world, and gaining power. That power comes from levels, stat point allocations, gear, and gear augments of various types. The power growth is non-linear, so that players feel a sense of power gain and progression as they level up and gear up. The itemization is the primary source of power gain, which is a three factor system itself, gating power behind item level, the item quality (magic, rare, unique/legendary), and any gear augment system that may be included. This is core to the Diablo experience, and PVE players would demand it, for if this were not present, it wouldn’t feel like a Diablo game.
And therein lies the problem with implementing PVP into a Diablo game. Diablo has never been designed to function as a properly balanced PVP game. It deliberately avoids doing the things that must be done to balance PVP games in order to give the player a sense of power gain and progression as he fights tougher and tougher demons. The demands for build diversity require developers to build in a wide variety of player power systems, which become increasingly difficult to balance to parity. In addition, the developers have flirted with the idea of infinitely growing power systems, which are impossible to balance in PVP. In a PVE game, these things are design goals, while in a PVP game, they are things to be approached with extreme caution if not avoided entirely.
This dichotomy of design goals creates a huge dilemma for designers who want to address players’ request for a balanced PVP experience while also providing a classic RPG PVE experience.
A Non-Exhaustive List of Possible Solutions
It is not constructive to simply outline a problem and leave it like that. The goal here is to solve problems, and there are a couple of ways we can address this dichotomy. I’m going to go through the options as I see them. Each has pros and cons, and ultimately assessing those pros and cons is going to be a subjective determination. My goal here is to spur discussion about what the community actually wants and what we’re willing to do without, which hopefully the developers can find value in.
Remove PVP Entirely
This is certainly an expeditious solution. It allows developers to focus purely on the PVE experience and gives them wide latitude to design their power system(s) in a wide array of possibilities, opening the door to potential build diversity. While great for the PVE experience, it would greatly disappoint the PVP players.
Balance for PVE First
This is the classic Diablo universe solution. PVP is allowed, but it is simply tacked on to a system designed purely around the PVE experience. It is grossly unbalanced with a handful of builds which simply dominate and comprise the only viable means to PVP. Players who specialize in PVP exploit hostility rules to gank lower level or otherwise unprepared players, creating a toxic experience. A handful of PVP players enjoy themselves. Everyone else avoids PVP entirely because it is unfair. It has the advantage of providing a great PVE experience, but the PVP experience is poor.
Attempt to Balance PVP and PVE Simultaneously
This is a solution the WoW team would be excellent to consult on. The arena experience is key for them. Skills have modified function in PVP settings. A parallel gear and leveling system for PVP can be created specifically for PVP players. The game is built to be flexible enough for players to choose PVE or PVP or to do both. However, it is very labor intensive and having a dual set of parameters for each skill which varies by target is difficult to learn for the players and difficult for the development team to maintain. The demands on the development team mean both the PVE and the PVP experiences would be suboptimal as the team is forced to make constant revisions and compromises between the two.
Balance for PVP First
Given the PVP has the more stringent requirements for balance, the team could choose to build the classes and skills around a PVP experience first, taking into account balancing classes against each other, giving them a robust toolkit to allow them survivability, but also a range of attacks from melee, to middle, to long range, with various levels of cleave/splash/AoE, etc. A leveling and gearing system would be built into the PVP-based structure where levels were earned or lost based on wins or losses, and gear could be ground specifically designed for the PVP experience. CC is heavily regulated in this system, since it is easily exploited in team environments to single out, freeze, and delete enemies. Once the game is built and designed around PVP, then the PVE experience is created for players to adventure in secondary to that, while retaining the core toolkits constrained to the balance parameters of the PVP experience. In this version, a player could easily go back and forth. Development effort is not excessive. The PVP experience is good and properly balanced, but the PVE experience suffers as it is constantly compromised in deference to the demands of PVP.
Separate the Two Experiences Entirely
In this solution, D4 continues to be designed as previous Diablo games, as a PVE game first. However, the developer team takes into account the demand for PVP and creates a PVP mini-game of sorts with its own rule set separate from the PVE experience. You wouldn’t take your fully-geared character into PVP. You’d take an archetype into PVP and play with a PVP-balanced version of your character’s class. For example, you’d have your fury dual-weilding Barbarian in PVE, but you’d play a generic barbarian archetype in PVP with a toolkit specifically balanced for PVP. Your barbarian would be the same as an opponent’s barbarian in PVP, even if their PVE ones were wildly different. This is a bit of a lazy solution, but it’s a lot easier to create and balance than attempting to fully integrate the whole game.
Conclusions
PVP and PVE games ultimately have two different and opposite sets of design goals. This creates a dilemma for designers where they are forced to design either towards fewer variables (PVP) or more variables (PVE) and compromise between those two creates the potential for a substandard experience in both realms of play.
The best compromise solution I can see is one where PVP becomes the primary balancing parameter and PVE is compromised around that. I don’t like the mini-game concept since no one wants to just play an archetype. They want to play their character. Balancing for the more stringent requirements of PVP first makes it possible for a good PVP experience, at the cost of the PVE experience which is compromised in the name of balancing PVP. Ultimately, the PVE experience will feel highly constrained and un-Diablo-like, and that makes sense because it effectively recasts Diablo as a PVP game instead of a PVE one. The core experience of Diablo becomes the “afterthought” and I don’t think anyone would be happy with it.
Ultimately, I believe that balancing the two in the same game is a futile effort, given this dichotomy. I’m not opposed to PVP in principle, but I am opposed to a fundamentally flawed PVP system that fails to address fairness and provide a good gaming experience. I am open to solutions that I haven’t thought of. Diablo is a PVE universe and it has always been designed around that. PVP was tacked on as an afterthought, exploited, and became a toxic experience which is why it was effectively pigeonholed to Brawls in D3 which no one plays. Given that PVP has always been the secondary and less popular of the two priorities in Diablo, if I had to make a choice, I would remove PVP and keep a great PVE experience intact.
As always, honest discussion and constructive feedback are welcome!