Pet Classes in MMOs

It’s hard to balance pet classes in MMOs because of competing desires between different kinds of players, along with contentious power variations. So what I’m doing… well, let me step back. Here’s a quick primer on pet class balance! [Warning: this is a game-theory-heavy post. You can safely skip it if you don’t care.]

Action Economy

There’s a game-balancing concept called the “action economy.” It just means that you can only make a certain number of actions in a game, so you have to fit all your actions into your “budget.”

In an MMO this is incredibly apparent: if a fight lasts 30 seconds, and pressing a button takes two seconds, you have an action budget of 15 button presses. (Of course, all sorts of other things determine the actual number of actions, but you get the point.)

Pet classes break the action economy, which is why they’re hard to balance. The pet can perform actions at the same time that you are performing actions. So now either all of your actions after summoning the pet have to be sucktastic and useless, or your pet’s actions have to be useless instead. Otherwise you’re getting a double helping of awesome actions.

In games like DAoC, pets were clearly balanced to suck. They didn’t hamper balance much because their actions were pitiful — like doing 1/10th the damage you can dish out, and nothing else. That’s pretty easy to balance, but it’s not what pet class players are looking for! Players who are into pet classes want their pets to be powerful.

Balancing Pets By Splitting Up Gameplay Into Departments

In original World of Warcraft, they basically couldn’t balance Hunters while keeping them fun. For most of WoW’s life, Hunters have been very overpowered in soloing. And Blizzard’s response has been “eh, soloing doesn’t matter, because they’re not overpowered in groups and raids, and that’s what matters.” The idea here is that Hunter pets can act as tanks while soloing — effectively giving players double actions — but in raids and large-group encounters, the pet tanks aren’t good enough, so those actions are useless. So the tank pets get swapped out with with DPS pets that don’t do much else, and  this is the scenario the designers balance around — making it easy to balance again, because the pets are more like DAoC pets.

(I’m talking about early WoW, obviously. I haven’t played much in years, but as far as I can tell, post Cataclysm, their balancing plan has been to say “fuck solo balance, we’ll just make it insanely easy for all the classes, and then nobody will care enough to complain.” Which… is a valid approach for certain audiences. [Like me! -Sandra])

In a way, you can think of this design as Rock-Paper-Scissors balancing, where the categories might be Soloing-Raiding-PvP. If every class is good at one, bad at another, and mediocre at the third, then that’s a kind of balance. But we see the flaws with this approach pretty quickly. For instance, nobody wants to play a “mediocre PvP class.” They only want to PvP with top-tier classes. So a class that’s mediocre at PvP might as well be terrible at it: nobody’s going to use it anyway.

Even if this approach was bulletproof, I couldn’t really use it. My game doesn’t have PvP, and while I have group combat, I’m working hard to make that an extension of solo combat — so a group takes on a big mix of solo monsters, rather than tackling unique “group monsters”. I need to find another way to balance pets.

Do They Need To Be Balanced?

Before going much farther it’s worth asking why pets need to be balanced. Especially in this game, where there aren’t any fixed classes and anybody can learn a pet skill if they want to.

There are a couple of reasons for this. First is the usual reason for balance in this MMO: without some basic balance, experimentation is boring. If a pet skill is head-and-shoulders above the other skills, people will lose incentive to try other combinations – and that takes a huge part of my game away. Likewise, if pet skills suck, no one is going to try them.

Second, there’s a technical reason not to have too many pets: if every player ends up having two or three pets active at once, that adds additional cost to the servers. The server has a budget based on the number of monsters that need to be moved around the map at once. Pets are monsters, in this regard: the server moves them around, and has to make sure to move them in valid, sane ways. The more pets there are, the fewer monsters can be in a zone, because the pets are using more of the CPU. Not a huge deal, but it’s something to keep in the back of the head.

Third is that pets are contentious. Some players love them to death, but others are frustrated by their maintenance elements, or by how they get in the way all the time, or by how they obstruct visibility in groups. Pet skills need to be relatively balanced so that both these audiences can have fun. I don’t want pet players stuck with useless pets, nor pet-haters forced to use pets.

But I’m not saying they have to be perfectly balanced. They just need to be playing on the same level.

Dual Class Roles

The pet-balance dilemma is one of the reasons I used a two-combat-skill system for Gorgon. Every player can pick two combat skills at once, which means that their power comes from the sum of both skills. This makes it easier for players to understand what they’re getting into. If you’re a Pet Tamer/Fire Wizard, you can expect roughly half your power to come from Pet Taming and the other half from Fire Wizardry.

Put this another way: this determines a pet’s basic power level. At most, a skill’s pets can be half as powerful as a player.

This likely means they would make poor tanks in a game with hard tank/damager/healer roles. Because a tank who is 50% as good as another tank is actually not very good at all. That’s not the sort of role you can half-ass.

But Gorgon changes up those roles a bit. Here, you can be a useful half-assed tank because you will regularly be fighting two and sometimes three monsters at once, both solo and in groups. Having a pet to soak up one of the monsters’ attacks while you focus on another one starts to make more sense there.

“Balancing Via Tedium” Revisited: Maintenance Activities

One of my rules of thumb is that an MMO shouldn’t balance gameplay via tedium. By that, I mean a designer shouldn’t say “Well you could become overpowered by doing X, but you’d have to do X for 500 hours, and who’s going to do that?” MMO players, that’s who. Because of the competitive environment, a lot of them will do whatever it takes, and they’ll curse your name for “making” them do it, all at the same time.

But here we have a case where a bit of downtime activities can really help, as long as it’s not too tedious. Going back to the action economy, if you’re a Necromancer/Fire Mage, and your pet skeletons are doing as much damage as your Fire Magic, then what are you doing with the rest of your Necromancer buttons? If you also have lots of powerful Necromancy blast attacks, that’d give you too many damaging actions — you’d be overpowered.

Many MMOs take some of the pressure off this equation by making you spend some of your actions keeping the pet alive. You have to heal, or buff, or guide your pet, which takes away from the amount of time you could be pressing attack buttons. If this is done carefully, this feels pretty good: it’s not really tedium, it’s just maintenance. Which is all that a classic Healer role is. Some players hate it, others find it the epitome of fun, most are somewhere in between.

I’m definitely doing a bunch of in-combat maintenance for some of my pet skills. It works out great for Necromancy: the necro is constantly raising new pets, healing old pets, and throwing in a well-timed pet buff, all while using their other skill to deal damage. So they have plenty of stuff to do that’s entertaining.

But for single-pet classes like Animal Taming, this feels like too much maintenance. It’s just one pet, and players don’t expect Fluffy to die in every combat, so you’re not even going to be resurrecting him over and over. Unless you have to chain-heal the pet continuously, there aren’t a lot of fun maintenance tasks for you to do. One or two buttons, sure. But what do you do with the rest of your buttons?

I don’t have a perfect answer, but I’d flesh out Animal Taming with some utility powers instead of direct-damage powers. That way they’re probably a bit overpowered in terms of the sheer number of verbs they have access to, but not in terms of damage per second, which is the e-peen measurement of choice.

Out-Of-Combat Tedium… Err… *Maintenance*

But let’s talk about Animal Taming a bit more. Even if it’s not overpowered in terms of DPS, it still has more actions than other players, which gives it more versatility… and that’s not even accounting for the versatility of having a bunch of different pets!

This might lead other players to complain, “Why does Animal Taming let you be a damager OR an off-tank OR a crowd-controller? I have to level three different skills to be that versatile!” This balance issue isn’t about the number of actions in a battle, but about how often a skill is useful. Animal Tamers will likely find they have things to do in most every situation.

To counter that, we can just make it take longer, so that it’s more similar to someone who leveled up two or three separate skills. That would be extreme tedium in some contexts, but this is a case where there are (hopefully) people that are excited to do this extra work. For instance: pet training.

Suppose you have to play fetch with your new pet for a few hours before they will reliably attack your designated target in battle. The pet does cute things while you’re training, and there’s a little bit of decision-making involved: do you make the pet chase the far-away stick (so that their legs get faster) or the bigger stick (so their jaw gets stronger… or something)? Or if the pet disobeys, do you scold them, or just tolerate their behavior and reward them when they obey?

These decisions might only affect how the pet behaves out of combat, but it’ll still matter a lot to the sort of person I’m imagining getting the most out of Animal Taming.

I’ll also have Animal Husbandry, where you breed pets to get the ideal temperament and stats you want. (And sell the others as non-combat/weak-combat pets to other players, perhaps.)

And after a moderate amount of this stuff, we start reaching a certain audience pretty well, I think. At least on paper. Every design changes dramatically when it comes in contact with actual players. But I think it’s a good place to start.

All this to say that Necromancy will be added soon! And Animal Taming… is a ways off, because there’s a bunch of new ideas to implement. Stay tuned.

This entry was posted in Design, Project Gorgon. Bookmark the permalink.

14 Responses to Pet Classes in MMOs

  1. Pingback: Quote Mine | Kill Ten Rats

  2. Jason says:

    I’m not sure pet skills are so unique, from a balance perspective. Across the MMO universe, lots of classes gain various passive abilities that don’t factor into the action economy.

    Is having a pet really that much more of a difficult balance problem than heavy/medium/light armor classes, or classes with buffs that can be cast pre-combat?

    I suppose in some games, those two examples are relatively minor effects, while as you point out few people are happy with pets being a minor effect. In other games, though, armor classes and/or buffs are a huge deal–I’m remembering Anarchy Online’s buffs, which were enormous.

    Guild Wars 1 might be an interesting case in point. Most primary class abilities didn’t interact with the action economy that much. If Rangers hadn’t had such a great primary ability already in Expertise, they could easily have buffed pets and made Beast Mastery a primary ability.

  3. Brad says:

    As far as Pet Tamer goes, wouldn’t part of the action soak for the pet be keeping the pet motivated and buffing it during the activity?

    That seems like it would be the big difference between pet skills like Necromancy (where the individual pets are pretty much disposable) and Pet Taming (where the particular pet probably has a name, and took some effort to get). So to me, it seems like you’d want to favor “Discard the minion and make a new one” for Necromancy, and “Who’s a good dire wolf? Are you? Yes, you are!” buffing for Pet Taming.

  4. Eric Heimburg says:

    Well I wouldn’t say they’re the hardest balancing task, but the player expectations limit your options, which can make it tough.

    Never played much of AO. In most games with big armor differences, the armor categories are split by role, so you rarely have to direct-compare a robe and a plate mail because the same player couldn’t realistically wear both.

    But in a game without a heavy role focus, a big armor dichotomy is definitely hard to balance, especially if people with lower armor are expecting to have higher DPS in return.

    Buffs can be a problem too, but it’s usually pretty easy to subtract from their direct damage/healing/whatever to balance out the buff. When the buffs are too powerful for that, it does get hard. I have that problem with buffs in my current game, as many are very strong. Might end up using the Concentration idiom to help rein them in. Or just nerf the hell out of em :) pre-combat buffs aren’t really that fun anyway, so the easier fix is probably just to pare them back.

  5. Eric Heimburg says:

    Brad – hmm, I’d been imagining a lot of that motivating would be done between battles, as opposed to giving pets pep talks mid-fight. But you’re right that I can probably do that to help use up some actions, thanks! Though even so, there’s still probably a lot more actions to fill.

  6. Brad says:

    I guess my thought is that from a certain perspective the pet abilities are just window dressing. I don’t know, maybe that’s cheating. :)

    If you give the pet a base base amount of damage per attack, and then give the Pet Tamer a command which makes the pet perform some special attack or temporarily raise the pet’s damage per attack, it’s still really the player’s action that’s important but it’s expressed through the pet. So a player ends up with the choice between using their abilities to encourage the pet to do better in combat, or to use their other combat skills.

    Back before I burnt out on World of Warcraft, the hunter pets had species specific skills that mostly worked okay if you left them on automatic but there were a few that were really really useful if you clicked them you needed them. I think turtles had a ‘cower in your shell’ ability and some of the birds had a charge/interrupt that they could be commanded to do. But it gave hunters reasons to choose each of the specific pets, and but also gave them (more) reasons to manage what the pet was doing during the fight.

    But it seems to me that the big complication for something like Pet Taming is that you’re ending up with this parallel itemization (choosing and breeding the pets), and you need to make sure that the pets are an extension of the player. And one of the things I remember about WoW was that even though it put all of that emphasis on equipment stats, there wasn’t really much of anything for improving pet stats so you’d run into the problem of outgrowing the pet. I think they might have switched to giving pets a percentage of the hunter’s equipment, but it’s been a few years.

    Either way, I’m sure it’ll be interesting to see what you work out for Project Gorgon.

  7. Rakai says:

    Will beast speech have anything to do with training pets?

  8. TP says:

    As i wrote ingame, simply take a good hard look at Ultima Online… pre-trammel.
    Best animal taming game ever created, and i tried them all… :p

    But Ultima Online has 700 skillpoint cap, and each skill goes to 100 (back then), and with animal taming going to 100, and being able to heal them, with bandaging, taking 100… well that leaves less options open for other combat skills.

    In Project Gorgon, since u can get all… well… i find that path abit troublesome.

    Also in Ultima Online, if you had Fencing for example, to get maximum damage, you would also need to have Anatomy i think it was… which leveled via fighting, but was still needed to get the maximum amount of dmg.

    Same with Magery, would need Meditation for example, + another i cant remember.

    My advice, to balancing things like Animal Taming. Make a skillcap, allow us to level whatever combination we want, but make us choose out of XXX amount of skillpoints.


    Skillcap 700
    Animal Taming: 100 points
    Animal Healing/Veterinary (passive): 100 points
    Fire Magic: 100 points
    Lore (passive dmg buff to magic skills): 100 points
    Gather Skill: 100 Points
    Crafting Skill: 100 Points
    First Aid: 100 Points

    This is a tamer who can use some fire magic and still have a gather + craft skill, + some self sustain via healing. Now this is far from a perfect example. Imo there should be a meditation (power regen skill) ingame also, and Magic Resist skill. Which would have been replacing the crafting skills in this case, for PvP atleast.

    You could also ofcourse split up your skill points as you wish, going:
    First Aid: 50
    Armor Repair: 50

    And so on… aslong as you understand there is a 700 max cap, and if you hit that cap, you would need to “skill down” current skills to level new ones.

    Sorry for rambling… but Ultima Online did this absolutely perfect. Please take a look.

  9. TP says:

    Tamers are also the only ones able to tame Horses/Special mounts for people… making it an economy itself. In Ultima Online you could ride the normal horses with 0 Animal Taming, hence being able to purchase such horses.

    The skills you needed to be a pure tamer was:

    Animal Taming
    Animal Lore

    Allready taking 300 out of your 700 maximum skill points.

    I can recomend doing some research on this…

    Good old stratics have lots nice information still… even though i prefer the oldschool information.

  10. Steve says:

    I thought the City of Heroes/Villains game had just one of the best pet classes ever, the Mastermind. While you had some offensive powers, the majority of your abilities relied on summoning and buffing your pets (and teammates). Really well done.

  11. TP says:

    Never tried that game. But if i were to login to Ultima Online, and check my stables, i would still have my old White Wyrm… who would be 10+ years old. A buddy i have fought many evil critters with over the years, and a buddy i have saved more than once from evil PK’s trying to kill him.

    And a buddy i have logged in to heal in the very last moment from evil disconnects…

    I still remember lots of the adventures i have had and shared with my pets in Ultima Online, and THAT is what makes a good involving tamer game.

    In my humble opinion… physically shaking over almost loosing your valued personal pet… is something unique, and something only Ultima Online has managed to capture.

  12. what about having various “tricks” that you can command your pet to do to various effects. So basically them attacking normally is like a small Damage over time effect, punctuated with them distracting an enemy or some such.

    Perhaps even have combat maneuvers that are collaborations between you and your pet. I can see signalling your pet to stand in position to trip a foe you force to retreat.

  13. Lolin says:

    I still play Ultima Online and until recently as a tamer. I like the basics of thier system. The biggest problem our shard had with tamers is they became too strong in PvP. However, in some group activities, like spawns and peerless, pets very much needed, both for damage and tanking.

    Another issue UO had concerning pets was how many pets a tamer could control at once. They solved this by assigning a certain number of pet points a tamer had and each pet was assigned a certain number. So if a tamer was allowed 5, and a greater dragon was 5, that was the only pet the tamer could have. He couldn’t even ride a horse.

  14. TP says:

    I guess you play freeshard?
    Tamers are not strong in PvP. Just kill them.
    They have zero fighting skills.

    (Played on Europe, and was one of the more active PK’s in the game… tamers were food)