Mightier than all the beasts in all the lands.
Roleplaying games are complex beasts. Even as sprawling simulationist games like 3.5 D&D and Shadowrun become less common, systems must have rules to cover a mind-blowing number of situations. With so much to cover, it’s no surprise that a few scenarios have slipped through the cracks. Game after game struggles with the same scenarios, making them seem intractable. They’re not, but game designers need to be aware of them so the scenarios can be eliminated before a system hits print, and GMs need know about them in order to head off trouble before it erupts at the table.
1. Reaching High Level
In roleplaying games, it’s nearly axiomatic to say that high-level play is fraught with difficulty. Third Edition D&D was notorious for the hours it took to get through a single round of combat with a party higher than level ten, and Pathfinder has largely inherited that reputation. Fourth Edition solved the problem with rigid advancement rules, but the solution was so extreme it created a whole list of entirely new problems. Fifth Edition is better than Third, but 20th level spellcasters still send GMs scurrying behind their screen.
The problem is hardly limited to D20. Legend of the Five Rings (L5R) and Shadowrun are both free from dice-pool systems that work well with beginning characters but slowly lose coherency as the PCs get more experience points. In both games, high-level characters must keep track of an overwhelming number of bonuses.
In games like Mage: The Ascension, powerful characters gain so many abilities they can easily destroy the plot. A mage with four or five dots in their spheres of magic has so many options it’s nearly impossible for the GM to take them all into account. A session of Mage will come to a screeching halt as one PC uses their time travel, complete control of space, or powers of resurrection to make all the GM’s prepared material worthless. “What’s that, you awaken the enemy’s building and ask it to crush everyone inside? Yeah, I guess that’ll circumvent everything I had planned.”
This is unfortunate because high-level characters are necessary for certain stories. Running a game about the fates of powerful nations is difficult when the PCs are barely out of protagonist school. Plus, when you buy a game that says the maximum level is 20, it’s reasonable to expect you can actually play at level 20 without everything falling apart.
So why is designing a system for high level play so difficult? Often, it has to do with play-testing priorities. Roleplaying games are usually produced on paltry budgets, so the low- and mid-levels are given the most attention. Games might ship without the designers having any idea how they play at a high level.
Scale also contributes to the problem. The difference between a rank one samurai and a rank five samurai in L5R is huge. In D20, the gap is even more pronounced. High-level D&D characters are like unto minor gods when compared to first-level beginners. When dealing with such a vast spectrum, it’s difficult for designers to balance both ends. If they make the game work at a low level, the high-level titans will be so overpowered as to make GMs weep. If they do the reverse, beginning characters won’t be able to accomplish anything at all.
How to Address It
For the designers, the ideal solution would be larger play-testing budgets, but that’s unlikely to happen any time soon. More realistically, designers can reduce the difference between low- and high-level characters. Don’t eliminate it – characters should still be able to progress – but move the two ends closer together. In Mouse Guard, for example, a patrol leader rolls more dice than a tender paw but not so many more that the leader feels like they’re from another plane of existence.
For GMs who want to use high-level play, the best option would be to pick a system like Torchbearer or Burning Wheel that handles it well. If that’s not an option, then at least you can make your life easier by restricting what material players have access to. It’s difficult enough to handle an archmage with six arete, worse if they have every splat book ever released to pick options from. Beyond that, just know the system really well and be prepared to do a lot of math.
2. Using Downtime to Gather Resources
Many roleplaying games feature abilities where time is the primary constraint. Mages in Mage: The Awakening* can use rituals to accomplish amazing feats, but doing so takes hours of work. Feruchemists in the Mistborn Adventure Game* can store vast amounts of power in their metal jewelry, but doing so means they must spend long stretches of time virtually helpless. Sounds like an effective balancing factor, yes?
It is – so long as enemies are beating down the door. PCs are unlikely to stop and perform a five-hour ritual in the middle of a gunfight. But in most campaigns, characters have long stretches of time when no one is trying to kill them. This downtime often happens between sessions, but it can happen at the table when the GM says, “And we cut to a few months later.”
Why does it matter, you might ask, what PCs do during their downtime? Isn’t it the action-packed stuff that we care about? Yes, except that what happens in downtime doesn’t always stay in downtime. In both Mage and Mistborn, characters can turn their downtime into permanent bonuses for later. Mages can ritual up bunker-busting lightning bolts and store them with time magic or use enchantment to create long lists of magic items for themselves and others.*
In Mistborn, feruchemists can use downtime to store so much power as to make themselves unstoppable. They can also use the downtime to acquire even more metal jewelry, as the rules for doing that are quite vague. In either game, the only options for a GM to stop this is to spontaneously generate enemies that attack any PCs who misuse their downtime. That’ll work once or twice, but soon the PCs will figure out what’s going on, and they’ll resent the GM’s transparent tactics.
In these games and others, designers don’t seem to understand that PCs are not in danger every moment of their lives. Unless the campaign is a Lord of the Rings style trek, or the sessions occur back to back like episodes of 24, there will be periods when nothing is happening. If the PCs’ abilities let them use this downtime to their advantage, they will exploit it eventually, and often the game will suffer for it.
How to Address It
As a designer, think more like a power gamer. Any time you put in an ability that’s limited by time, imagine how that ability would function given infinite time. With a mechanic like spells per day, the results are harmless. Wizards cannot build up extra spells slots from those they don’t use. The issue comes with any ability that generates greater returns the longer it’s used. When you find one, it’s time for a redesign.
When GMs find such an issue, their best option is to attach some kind of cost to the ability that can’t be mitigated by time. In the Mage example, I often require PCs to fetch rare and expensive reagents before performing any ritual. If they want to brew up a super lightning bolt, they’ll have to venture into the spirit world and recover Zeus’s sandals first. That way the use of ritual magic is a reward for doing something cool, rather than an incentive to camp out in their basement for months at a time.
3. Players Cooperating With Each Other
Often, a system will seem perfectly balanced so long as it’s one PC against the world. But then you add more PCs, and things get complicated. In Changeling: The Lost, players have access to pledges, which are a kind of supernaturally enforced contract.* Pledges can be anything from a pact of friendship to a lifelong vow of vengeance against litterers. No matter the specifics, following a pledge grants powerful benefits, often in the form of permanent bumps to a character’s resources or other abilities.*
When dealing with an NPC, this is fairly balanced. Swearing to the Freehold Lord that you will be their lifelong ally might grant you several ranks in Fencing, but it also gives an NPC control over your actions. That’s not a decision to be made lightly. Who knows what the Freehold Lord might do with your oath of loyalty?
This isn’t a concern when you’re dealing with another player character. If you swear the same oath to another party member, you can get the benefit without worrying your oath will be abused. If it is, then you’re dealing with a PVP situation, which has its own list of problems.
Every changeling PC is surrounded by three to five other changelings with whom they can form pledges without fearing the consequences. The bonuses generated by this mechanic can be game breaking. Suddenly, every PC at the table is overflowing with money, allies, weapons, and anything else the pledge can grant.
This problem arises in social dynamics as well as game mechanics. Vampire: The Masquerade is meant to be a game of intrigue and plotting. Vampires scheme against each other and launch plans that take decades or centuries to bear fruit. When playing a vampire, you can’t trust anyone. Well, anyone except the other PCs. Again, unless the group is willing to delve into PVP, every vampire PC has a cadre of fellow bloodsuckers they know can be trusted.
That’s one reason so many Vampire games devolve into undead action movies. When the PCs can present a united front, they don’t have as much motivation to invest in all the wheeling and dealing the designers intended.
How to Address It
Designers must remember that the relationship between PCs will always be different than the relationship between a PC and an NPC, or even between two characters in a novel. Out-of-game social dynamics necessitate this difference. To some extent, most PC groups will act like a single unit as much as they act like individuals.
Armed with this knowledge, designers have two choices: They can either create abilities with the understanding that PCs will use them in cooperation with each other, or they can create a game that supports PCs treating each other the way they would an NPC. The first option is much easier. For designers who want to go the second route, I recommend looking to the Song of Ice and Fire game for inspiration. While it assumes the PCs are members of the same family, the rules allow for a more antagonistic relationship without devolving into the nastiness of traditional PVP.
GMs who are already holding a copy of Changeling or Vampire must take a more pragmatic approach. For mechanical issues, the solution is similar to dealing with downtime abuse: attach a cost that discourage rampant ability sharing within the party. The cost shouldn’t be too high; you don’t want to make it impossible for PCs to work together, but it should be enough to make players think. In the Changeling example, I impose a -1 to a PC’s maximum glamour* for each beneficial pledge the changeling holds. PCs can still make pledges with each other, but they’re less likely to go overboard with it.
If the issue is social cooperation, as with Vampire, then I recommend encouraging minor differences in the PCs’ goals. Nothing too drastic, you don’t want the party at each other’s throats, but enough to keep them slightly wary of each other. For example, you could encourage PC Megan to think that feeding from the blood bank is a good idea, and PC Tamil the opposite. That difference is unlikely to cause a meltdown, but it will give them something to think about. Of course, some groups can’t handle any kind of PVP, and you’ll need to judge if your group is among them.
4. PCs Getting Their Own Minions
From stormtroopers to orcs, minions are a tried and true tool in the GM’s arsenal. That’s all well and good, but what happens when players get their hands on a few? Madness.
Across multiple systems, the rules for PC minions are broken as hell. I’ve spoken before about Retainer merit from New World of Darkness (NWoD). In short, it allows players to get a second character who’s better than their original. The Familiar ability does the same thing in various editions of Mage. NWoD also has a mechanic for training an infinite number of attack dogs, but I suspect that’s just an error the editors missed.*
Over in Pathfinder, the Leadership feat can bring games to a crashing halt.* First, it grants the PC a follower only two levels lower than they are, which is an immensely powerful resource. Make the follower a cleric or other support class, and the party’s effectiveness greatly increases. As if that weren’t enough, Leadership also grants dozens of weaker followers. Just keeping track of them all is a nightmare. The rules for determining the followers’ stats are super vague, but if the player can justify them being wizards or sorcerers, the party will have a deep reserve of magic missiles to draw on.
Some argue that PC minions are balanced because they can be killed or otherwise removed from play, unlike core PC abilities like skills and attributes. That’s true, but it’s a nuclear option. When players spend points on a minion, they view it the same way as spending points on their sword skill. If you use your power as the GM to kill off a minion you think is too powerful, the player is likely to see that as a serious breach of trust, and that isn’t the kind of dynamic you want in your game.
Minions are also difficult from a story perspective. As the GM, you’re expected to roleplay them like any other NPC, but the player controls their actions. You also have to account for them in every interaction involving their player, which can get tiresome. You can’t just say that they were separated from the party, because the player will cry foul. They paid points for that minion; it’s not fair of you to just remove them. The minion ends up requiring even more work than a PC, but they don’t contribute anything to the story.
How to Address It
From a design standpoint, a PC-controlled minion should rarely have their own character sheet. Instead, they should provide a specific bonus, within whatever their purview is. Chronicles of Darkness* did a really good job of this with its version of the Retainer merit. Instead of a second, overpowered character, Retainer grants a loyal NPC who’s good at one task as mandated by their profession. This is useful to fill a gap in the PCs’ skills and hardly overpowered. Torchbearer has a similar system. PC minions only give a helping die.
If losing a minion is on the table, create a built-in way to replace them. In World of Darkness, where minions are bought through character points, this should happen automatically after a session or two at most. In a game like Torchbearer, where minions are acquired with money, all a PC needs to do is drop some more cash.
Story-wise, GMs should make it clear to the group that they may end up separated from their minions temporarily if the story calls for it. Players who know what’s going on won’t be indignant when it happens, and losing access to their minions for a few scenes can be a good balancing factor for something that might otherwise be overpowered. To save themselves some work, GMs can also recruit a second player to roleplay the minion, rather than doing it themselves. This takes some responsibility off the GM’s shoulders, and lets them focus on running the game.
5. Characters Hyper Specializing
If an ability in World of Darkness gave +10 dice to Stealth rolls and only cost one merit point, we’d all know it was overpowered. Fortunately, designers can usually spot these overtly overpowered abilities and remove them before a game ships. A few slip through but usually not too many. Far more pervasive is the problem of many different abilities, each reasonably priced, giving bonuses to the same roll. If a PC takes all of these abilities, they become hyper specialized.
A lot of systems have this problem. In Changeling, it’s easy to stumble into a huge dice pool for Stealth completely by accident. Anyone who plays a Darkling and takes the corresponding Contract* will end up with more sneak dice than they know what to do with. In Spycraft’s first edition* it was possible to stack feats so that a level 1 character could easily make DC 30 skill checks. Of course, these characters couldn’t do much else, but that didn’t stop players from building world-class hackers right out of the gate.
A hyper specialized character isn’t over powered; they’ve spent the commensurate amount of points for their ability. However, they do create a poor play dynamic. Any roll in the character’s specialty is almost certain to succeed, while rolls outside of it are likely to fail. Players don’t like being asked to make a roll they’ll almost certainly fail, so the GM feels pressure to only ask for rolls in the PC’s specialization. GMs want their players to have fun, after all. But then the GM must either artificially inflate the difficulty or just let the PC succeed on nearly every roll. The first option damages the credibility of the rules, and the second one is boring for everyone.
When a PC is super good at one skill, every problem looks like it can be solved by that skill. Clever players can come up with all sorts of creative ways to leverage their stratospheric rolls, and heaven forbid they’re hyper specialized in a broad skill like Persuasion.
Players typically expect high rolls to equal high results, and usually that isn’t a problem. But when a PC is hyper specialized, their rolls can get so high that giving a commensurate result would destroy your game. “Yes, I know your character can easily hit the top of the difficulty chart with their Baking skill, but I’m running out of Legendary feats of cookery to give you.”
How to Address It
As is often the case, the best way for a designer to root this problem out is to think like a power gamer. They must look at each skill or attribute in their game and imagine how to take it to ungodly levels. If the designer isn’t a power gamer themselves, there’s usually at least one floating around the office or friends group who will happily take the game apart. Once the problem skills are found, the best solution is to cut down on abilities that give bonuses to the same roll. We only need so many ways to get an extra die in Flower Arrangement. They shouldn’t eliminate specialization entirely, which would destroy PC individuality, but they should keep it from reaching game-breaking levels.
As a game master, your best bet is to limit how high the skills or abilities can be bought in character creation. Like the designer, you aren’t trying to eliminate specialization; just keep it to a reasonable level. The exact limitations will vary from game to game, and you’ll need to determine how high to set it yourself. I’ve found a good benchmark is to cap abilities when they can easily hit twice whatever a system’s average difficulty is, but your mileage will vary.
6. Giving Players Too Much Responsibility
For a long time, the standard model of roleplaying has cast the GM as some kind of solitary figure on the mountain, dispensing narrative as a prophet might dispense divine wisdom. The GM tells a story, and everyone else is along for the ride. This idea is starting to fade with the introduction of more player-focused games, and that’s a good thing. Roleplaying stories should be collaborative; otherwise the GM would be better served writing a novel.
Unfortunately, some games have compensated too far toward the opposite extreme. These games severely curtail the GM’s authority over the story, trusting the players to fill in the gap. Burning Empires is a prominent example. In this game, the GM’s only meaningful distinction from a normal player is that they get three characters instead of one. Otherwise they have to follow all the same rules as everyone else. They use the same restrictive scene-building system and must roll the same way to bring in more NPCs.
On paper, this might seem like a great way to take collaborative storytelling to the next level.* In practice, it causes a lot of problems. It turns out that when you give five or six people equal control over a story, what you usually get is a mess. Instead of combining into a beautiful whole, each player takes the story off in their own direction with whatever makes sense for their character. The narrative completely falls apart. Since no one has the authority to shape the story, there’s no foreshadowing and no setup for dramatic reveals. Worse still, because the GM can’t easily bring in more NPCs, they have no way to provide a challenging opposition if the PCs are more formidable than planned. Entire campaigns can be wasted defeating paltry enemies over and over again, and that gets boring fast.
Primetime Adventures (PTA) doesn’t explicitly require this style of play in its rules, but the author clearly intends for it to be played that way. The GM is meant to create an introduction scene, then sit back and let the players take it away. The result is almost always the same: a lot of silence and confused looks.
Unless your players are a professional improv troupe, it’s unrealistic to expect them to collaboratively fill the role of storyteller. Some players have no interest in actively shaping the story; they show up to be entertained. Others have ideas but don’t know how to communicate them well. But no matter how inclined towards storytelling players are, they’re hindered by the basic setup of a roleplaying game. In RPGs, players take on the role of a single character. They become that character’s advocate. For a lot of players, this is necessary to get into character. It’s very difficult to move from that point of view to the big-picture thinking necessary to craft a story. That’s why the GM doesn’t have a character of their own.
How to Address It
Designers need to learn that player agency is good but it must be used to complement GM storytelling, not replace it. Despite its potentially treacherous advice, PTA’s rules are actually a good place to look for inspiration. PTA allows anyone to create a scene and gives the GM explicit control over the game’s antagonists. That way, if players want to actively participate, they have the tools to do so. At the same time, the GM is free to fill in the gaps, create enemies worthy of the PCs, and frame scenes that bring any wandering plotlines back together. If PTA isn’t the perfect balance, it’s damn close.
For GMs, this is an inherently difficult situation. How do you address a problem that explicitly takes away your power to fix it? This is one of those times when you just have to lay down some game-master fiat. If you’re playing a game that ties your hands, simply inform your players that you’ll be ignoring those rules. Chances are they won’t care. Please note, rules that empower the players to shape their own story are not inherently restrictive to you. How do you know the difference? Quite simple: one type of rule lets the players do something; the other forbids you from doing something. It’s only the second variety you must look at askance.
Roleplaying-game designers have trouble with these scenarios because they’re extremely difficult to handle. Everyone who makes a game would love it to run perfectly at high level, but getting it to do that is another matter. However, if we want RPGs to continue improving as a medium, recognizing the problem is a necessary first step.
Treat your friends to an evening of ritual murder – in a fictional RPG scenario, of course. Uncover your lost memories and escape a supernatural menace in our one-shot adventure, The Voyage.
I’m not sure I agree with #6. There’s been a lot of great GM-less systems out there, such as Fiasco. I feel you just need to set the expectations away from the traditional structure and for more collaborative play.
GM-less story games like Fiasco or Kingdom certainly offer a very different experience to the standard RPG setup. It’s not an experience I particularly care for, but they are up front and explicit about it. Games like Burning Empires keep the GM around, but as some kind of withered appendage, unable to fulfill their function.
I would say that 2 and 3 are actually faces of the same problem: mechanical benefit “purchased” with non-mechanical costs. Players are getting a well-defined, often quantifiable advantage in exchange for something that often doesn’t impact the game. They don’t have a finite amount of time to spend or a penalty for their relations with party members. Those costs are not coded as part of the game system, they don’t feature on character sheets or adventure design and end up being completely unpredictable. The amount of downtime available isn’t defined by the rules, so it depends entirely on the story being told – but it gives a pretty well-defined advantage that is not dependent on the story. The GM could, theoretically, take that into account for his adventure design – but the rules just don’t cover it. It is a non-mechanical cost.
The same problem happens when the only restriction or cost is story-based – if magic is overpowered but practitioners are hated, for example. Of if to get some powerful ability you need to find a special mentor. Or even narrative demands that could be fulfilled on a backstor. Those aren’t things the rules actually control, so they end up either being non-issues or just “this doesn’t fit in the game”.
Obviously, non-mechanical circumstances are an important part of the game. They enrich the story and ties the abilities with the game world. They just can’t be used as costs for mechanical benefits.
I’m in total agreement. That’s an elegant way of putting it, Elda.
With regard to Kingdom, what is interesting is that it, in effect, splits the GM role between the in universe roles in the community. Which works extremely well to simulate the way in which power struggles in political bodies actually play out.
Another related idea I had was a bit unusual. It is to have the GM play the campaign itself as if it were a character. They would still have major NPCs involved at different points, but the main enemy would be the campaign itself. The idea is about building enough of an advantage to take out the campaign in some fashion, depending on the nature of the threat. What would be interesting is that it could work for any genre. It could have a real value, if the GM were playing something like the Empire in Star Wars, or it could have an abstract one, like in the case of a mystery in which the mystery itself would have mechanical values beyond that of the actual killer. It could also represent more exotic threats like a disease, a la Pandemic(the board game). The key issue would be a way to reduce the mechanical value of the big picture problem to the scale of individual antagonists in a way that doesn’t produce all that much actual work for the GM. This is something that I don’t actually know of any system that does.
The idea was largely based on the fractal concept in Fate, in which any element can be modeled as a character. In particular, the main objective would usually be about creating invoc able aspects that can be used to attack the opposing campaign and destroy it in a single attack.
Elements of this also occur in resolution mechanics of something like Fiasco, or in a very similar also GM free form that works against an outside threat in Our Last Best Hope, which is largely about a team of heroic scientists preventing disasters, often involving heroic sacrifice to deal with mistakes. It has also been used for superheroes similarly to The Avengers, which could be generalized into a heroic suicide mission with soldiers, or Inception, which could be generalized to a generic heist. I wonder if there is a way to adopt that sort of mechanic to a GM without weakening them to the point of uselessness in the fashion of Burning Empires.
I’d like to add another to the list.
So many games list Charisma, Intelligence, Wisdom, etc. as stats, yet aside from derived skills, mechanics, etc. they are not very descriptive of how the character will be played. I’ve had players with terrible scores manage to squeak by with superb role-playing and problem-solving, making these stats largely irrelevant.
A solution would be to simply re-brand the attributes to thinks more strictly in-universe. Change Charisma to Appearance. Change Intelligence to Education. Change Wisdom to Perception. Basically, make sure the mechanics don’t contradict something that a player already has.
My favorite example of #5 is the third-level half-elven bard in D&D 3.5 with an 18 Charisma, the Negotiator and Skill Focus: Diplomacy feats, and six ranks each in Bluff, Diplomacy, Knowledge (nobility and royalty), and Sense Motive. Between skill synergies and racial, Charisma and feat bonuses, she gets a +23 modifier to Diplomacy checks — before she casts Eagle’s Splendor on herself. Forget the rest of the party; that bard can go solo adventuring and turn every single sapient creature she encounters into an ally — which quickly leads into problem scenario #4, too.
Hah, and they said the bard was underpowerd!
#5 is just as big a problem for beginning characters. One unfortunate incident from my gaming years:
“We’re playing Shadowrun”
“Cool! I’m in the mood to play a decker.”
“We’re playing a running shoot ’em up, no decking required. Someone give this chick a leather jacket, she has no armor whatsoever . . . !”
A bit of metagaming by the GM in the form of pre-game advice would have saved everyone a lot of hassle.
>player agency is good but it must be used to compliment GM storytelling, not replace it.
Should that be “complement”, or is player agency supposed to make GM storytelling feel good about itself?
Nah just a typo.
It seems like you like Mouse Guard. Or Torchbearer. Which is okay, but I don’t hold them as examples of “doing things better.” I mean, Mouse Guard and Torchbearer fail some very basic tests for me.
In Mouse Guard, even in published stuff, players have to score these “token of player agency” like I would prefer to call them where players have to disadvantage themselves to earn the right to do things after the mission they might find important themselves but weren’t part of what the Guard deems your mission. Want to spend a side trek to find that family heirloom for that family? Better lose some ties to the GM so that the system allows you to. No in-game reason. No actual justification, except that the game designer seems to love it when players have to make some token sacrifice to be allowed some agency.
Then Torchbearer, the supposed genre emulation of the old school gaming that makes every dungeon a nail biter – but can you run a city adventure in it? Also, the challenge of scoring some candles… and oh yes, the death spiral of conditions vs having to go back or starve and sleep in the streets. As a genre emulation it completely fails. It’s a completely self-contained game that emulates a particularly hateful OD&D DM’s game… Besides loot being so important the first published module is not centered around loot at all, it seems more like beating up a big baddie. Personally, I think it’s messy.
Then there are the “guess my next three moves” mini games and the predilection of resolving every possible conflict in game with a series of dice rolls, whether it’s trying to trick a dragon or combat with a troll, together with randomized HP. And then you negotiate who dies if you dare initiate a kill conflict against anything that isn’t a measly kobold.
And still Burning Wheel has hordes of fans. Sigh. Unless you think a particular aspect of the game is praiseworthy, it makes me wonder what makes these games repeat examples? They seem to be hellbent on taking agency away from both GM and players and putting everything on a ticking clock in case of Torchbearer. Everything.
I think Mouse Guard can be enjoyable enough, but Torchbearer I think is a screwjob…