Five Roleplaying Mechanics That Must Go

Some of these mechanics are at the heart of their game.

Roleplaying design isn’t easy, as we can see from all the bad mechanics that make their way into our favorite games. Some mechanics aren’t inherently flawed but are merely implemented poorly. Other mechanics have no redeeming qualities and should be dropped completely. It’s that second category we’re talking about today. Designers use them despite their many problems, and someone has to speak up!

1. Static Difficulty Numbers

Cover art from Call of Cthulhu, featuring a scary castle and the RPG's title.

In the majority of systems, the difficulty of a roll depends on what the character is trying to do. The GM either consults a chart in the book or assigns a number on the fly. Landing a passenger plane in rough weather will be easier than flying a fighter jet down the Grand Canyon at combat speed. Two factors affect the probability of success: the task being attempted, and the character’s skill. An average pilot can probably manage the first task, but only a highly trained expert can pull off the second. This method is straightforward and effective.

But some systems are different. In games like Call of Cthulhu and Apocalypse World, the difficulty of a task is always the same. The only factor affecting the probability of success is the character’s skill. The immediate effect of this mechanic is to give the game a surreal feeling, because characters routinely fail easy tasks and routinely succeed difficult ones. For a lot of players, this breaks the suspension of disbelief. They have a hard time enjoying themselves in a world that seems to operate on such alien and comical laws.

More subtly, static difficulty numbers strongly discourage players from diversifying their characters. In most games, getting a few ranks in a new skill means the character can have a decent chance to succeed at easy tasks. But with static numbers, every task has the same difficulty, meaning skills with only a few ranks are useless. At the same time, being really good at a skill means the PC can achieve just about anything they like, because even the hardest tasks have the same difficulty. The end result is a party full of characters who are only good at one or two things, which they almost never fail at.

Finally, static difficulties take away an essential GM tool: making the villain more threatening. From a fiction perspective, we all know that defeating the main boss is harder than taking out mooks, but it doesn’t feel that way if everything is the same difficulty. Normally, GMs increase the difficulty of climactic rolls to represent how much the PCs must struggle to get what they want. With static difficulties, that is impossible. The big boss isn’t any harder than the campaign’s first encounter was.

The upsides offered by static difficulty numbers are minimal. They relieve the GM of the responsibility of assigning difficulties, but that becomes second nature to most GMs pretty quickly. It’s often easier to calculate the probability of success with a static difficulty, but players don’t need to know their exact chances every time. These fringe benefits are not worth all the trouble that static difficulties cause.

2. Special Abilities That Aren’t Special

Cover art from Legend of the Elements, featuring an Earthshaper and a Watershaper.

Most systems have some kind of special abilities characters can take outside of standard skills and stats. They go by many names: feats, advantages, merits, etc. But no matter what they’re called, these special abilities let a character do something they couldn’t otherwise do. That’s why the abilities are special, after all.

Or at least that’s how it’s supposed to work. Sometimes, a special ability does nothing other than give a character permission to do something they could already do. For example, in Legend of the Elements, the Hunter gets an ability called Stealth, which lets the character roll to sneak past their enemies. Except, shouldn’t all characters have the ability to roll to sneak past their enemies? Moving quietly is a pretty basic concept. Not everyone is good at it, but they understand the basics.

Legend of the Elements is hardly alone in this. In the New World of Darkness, characters have to buy a special merit if they want to disarm their opponent. Without this merit, even a character with max ranks in Melee, a true master of swordcraft, has no concept of knocking the other guy’s weapon out of their hand.*

No matter which system they come from, these “special” abilities have one of two effects on the game. In the first scenario, the GM just lets everyone roll to use stealth, disarm their enemies, or whatever task is supposed to require the special ability to perform. At this point, the player who bought the special ability will be understandably upset because they wasted their points.

In the second scenario, the GM follows the letter of the rules. Only the PC who buys Stealth can try to sneak past people, and only the PC with Disarm can try to take away the enemy’s weapon. At best, the other players will chafe against this restriction. At worst, they’ll turn on the player who bought the ability as the cause of the problem. “I could sneak just fine until you bought an ability that says only you can sneak!”

Special abilities should give characters new and exciting options. If they don’t do that, there’s no reason to include them.

3. Harmless Knives

Burning Wheel cover art, featuring a wheel and lots of fire.

When determining how much damage weapons do, it seems natural to use size as a guide. If getting hit with a small sword hurts a little, getting hit with a big sword should hurt a lot. It all makes sense!

Unfortunately, this kind of scaling often has the effect of making it nearly impossible to injure someone with smaller weapons. Knives are the most common victim, but slings and staffs suffer as well. In systems like Burning Wheel and Legend of the Five Rings, even master knife-fighters have difficulty inflicting a hit of any significance.

Mechanical Explanation of Knives

Attacks in Burning Wheel can inflict three levels of damage: Incidental, Marked, and Superb. Someone with 4 Power who’s wielding a knife inflicts a strength 7 Superb hit. That’s a medium wound for most characters, and it’s the most a knife wielder can ever get, and they need two successes over the obstacle to get it. Most of the time, they’ll have to make do with a Marked hit at best, which is barely a scratch.

In Legend of the Five rings, knives do 1K1 damage. That means 1d10 is all you get. A character’s strength, usually 2-4 is added in unkept dice, so you’re looking at between 3K1 and 5K1 for damage, but you still only get to keep the highest die. PCs can make their attack roll harder in order to roll more dice on damage, but they only get to keep more if they get over 10 dice, which is practically impossible. So no matter how good and/or strong the knife fighter is, they’re only keeping one damage die. That’ll barely slow a bandit down!

At best, characters need to hit their opponent many times over in order to cause any harm. At worse, even that is ineffective. It gives the impression that the character is wielding one of those trick knives that slides back into the handle. To say this breaks suspension of disbelief is an understatement. Most players know that knives in the real world are dangerous, that they can easily maim or kill. A game that treats them as toys will destroy immersion fast.

But a loss of immersion is nothing compared to what happens if a player decides their character will be a knife fighter. The player imagines their character darting in and perforating the enemy’s kidney, but in reality they inflict little more than a paper cut. That’s enough to make even a patient player walk away from the table.

Beyond the mechanical limitations, this kind of mechanic negatively impacts a game’s setting as well. A gang of dangerous cut throats and their switchblades will suddenly seem quaint and nonthreatening, as will the monastery of kung fu monks armed with quarterstaves. It’s even possible that players without a foundational knowledge of weapons will leave the table thinking that such weapons are harmless. Then those players will write fantasy novels where everyone needs a sword the size of a car to do any damage.

4. Useless Immunities

To samurai from Legend of the Five Rings looking at a sword. Unbreakable you say? Could we make buildings out of it?

Not all special abilities let a character do something new. Instead, some of them grant immunity to one or more kinds of harm. This might be something as simple as immunity to poison or as abstract as protection against harmful insults. So long as these threats are actually present in the game, then immunity is a valid ability. But too often, systems give players the option to protect themselves from something that was never a danger in the first place.

The best example of this is the Kaiu Blade from Legend of the Five Rings (L5R). This is a special sword PCs can buy at character creation. It does more damage than a standard katana, but more importantly for our purposes, it is unbreakable. Based on that description, you might think having your sword broken is a serious problem in L5R, but it isn’t. If there are any mechanics for breaking swords, a decade and a half of playing hasn’t been enough for me to find them. Even if they’re buried somewhere in an obscure supplement book, it’s clearly not a common problem.

The Kaiu Blade is just one way this mechanic can manifest. It also crops up in abilities that grant immunity from a single type of poison or a D&D feat that allows characters to cast spells without their material components. Chances are very low that someone will target a PC with the specific poison the PC is immune to, and spell components are practically ignored by D&D’s rules anyway.

These useless immunities leave the GM in an awkward position, similar to what happens with “special” abilities. If the GM runs the game normally, the player who purchased the immunity will feel like their points were wasted, because they were. Alternatively, if the GM goes out of their way to make the immunity useful, it will punish the other players.

Consider L5R again. There’s a reason the game isn’t full of monsters that break swords. Katanas are hard to replace, and without them a samurai isn’t much use. It wouldn’t be very fun for the group if they were suddenly attacked by the Oni of Broken Swords, just so that the points one player spent on a Kaiu Blade wouldn’t be wasted.

5. Penalties as Difficulty

Cover art from Chronicles of Darkness, showing a shadow on an empty street.

Most systems make a task easier or harder by increasing or decreasing how much a PC has to roll in order to succeed. A few use a static difficulty, as we talked about earlier. But there’s a third method that’s somewhere in the middle: systems that change the difficulty of a task by giving the PC penalties.

Technically, Call of Cthulhu works this way, but the penalties are so vestigial that most GMs forget to use them. For a system where they are more baked in, we must look to the Chronicles of Darkness.* In that system, characters roll a number of dice equal to their attribute plus their skill. Any die that comes up eight or higher is a success, and PCs only need one success to pass a test. At first it seems impossible to fail, but the GM is supposed to subtract dice from the PC’s pool based on how difficult something is.

Mathematically, this system isn’t any different from changing the number of successes required for a roll, but from the players’ perspective it makes all the difference. When players are told they must roll against a high difficulty, they accept it unless the number is unreasonable. But when players are given penalties to their rolls, even when the penalty is reasonable, they’ll contest every lost die.

Players aren’t doing this to be difficult, they’re doing because getting penalized doesn’t feel good. In the moment, it feels like the GM is picking on them. Broadly speaking, players have been trained to understand that a higher difficulty means the task is harder while penalties to their rolls mean there’s something wrong with their characters. So when their rolls are penalized, they don’t see it as an attempt to do something difficult but as a sign that their characters have become incompetent.

Penalties are fine so long as they come from a problem connected to the character. Most players will accept that their character’s broken leg gives them a penalty to their attack rolls. But they’ll have more of a problem with the idea that they should be penalized because the target is far away.

Finding one of these mechanics in your favorite system doesn’t mean you have to throw it out. Many systems are good despite such flaws, but they are unquestionably flaws. If you’re looking to design your own system, leave them out. Your playtesters will thank you.

Treat your friends to an evening of dark ritual murder. In a fictional game scenario, of course. Uncover your lost memories and save the day in our stand-alone game, The Voyage.

Read more about



  1. Titto Paolo

    On the matter of penalty as difficulties, i never encountered player who contest penalties as long as the GM also remembers that it means you get bonuses for easier tasks.
    I had the opposite problem with rolls to target number, where i kept finding difficulties arbitrarily high (or at least it felt a lot that way) making my character’s actual scores inconsequential, meanwhile with the same GMs in other games with penalties as difficulty i could take the penalty and still keep a reasonable amount of my bonuses to make the difference.

    • Carolina

      Exactly what I think when I read.

  2. Alexander Guerrero-Randall

    Okay, once and for all, the ‘Stealth’ move reads:

    “When you conceal yourself from view, roll +Fluid. On a 7 or greater, enemies can’t see you and are tagged as _Unwitting Prey_.

    On a 7, 8, or 9, also select one of the following:

    -Someone is on alert now
    -You can’t keep track of everyone from your hiding place
    -You can’t stay undetected here for long.”

    Notably, giving people Tags is a big deal in this system, often requiring a successful roll (e.g. Commit Open Violence) to do.

    Also, the Hunter’s ‘Ambush’ move reads “when a foe is marked by a Tag indicating that they are not aware of your presence, you can spend one Chi to immediately defeat them and remove them from the action.”

    This obvious combination, which can be taken on a starting character, is, if it’s not obvious, Very Special and gives the Hunter a cool thematic thing that only they can do.

  3. JackbeThimble

    That’s actually a really good point about the knives but it seems like a very difficult one to solve. In reality the reasons why swords would be preferable to knives mostly have to do with reach and the reasons why a knife would be preferable mostly have to do with close quarters, concealability, or finding a gap in enemy armor. Unfortunately all of these things are very difficult to handle mechanically without adding an onerous level of complexity that will almost certainly be ignored. The only system I can remember that even came close to getting around this was D&D 3.x where the sneak attack ability meant that a rogue could be an absolute maniac with daggers or shortswords by taking weapon finesse and two-weapon fighting so that it was at least possible for a competent individual to be a threat with a knife if they were built for it. Of course they’ve nerfed sneak attacks and made it so finesse doesn’t require a feat in 5th edition so now there’s no reason to do that anymore and any smart rogue will go with a rapier in the mainhand and a shortsword in the off at best.

    • C. R. Rowenson

      That’s a good point about the difficulty of knife mechanics and is very similar to a problem I have with a lot of systems… GUNS!

      Most of the games I’ve played, at least the ones allowing guns, seem to have odd rules obviously put in place to limit the damage and effectiveness of the weapon. I understand this is needed for balance but it always feels… off to me.

      I don’t know. What do you guys think? Do you know any systems out there that handle guns really well?

      • Michael Campbell

        You know I recognise that this anecdotal evidence but when my brother was court officer down at the Darlinghurst Courthouse; there were six cases involving knives and six involving guns.

        Of the six involving knives there was one fatality:- The attacker was victim of Saddam Hussain’s torture regime and was probably given (if the victim’s previous behaviour is anything to go by) the “if you don’t pay your rent then I’ll make sure you get deported back to whatever godforsaken country you came from” speech.

        Of the six cases involving guns, there was only one survivor:- An off-duty cop who put his hand over the mussel of the firearm so that the bullet would have to pass through his hand and his forearm before getting anywhere important.

        Maybe guns should be very deadly and knives merely “bloody dangerous”.

    • Oren Ashkenazi

      I think Dogs in the Vineyard actually did a really good job making guns feel dangerous without making other weapons feel like toys. I think the solution is to lean harder into narrative, abstract territory.

      • VoidCaller

        In my system (Void System RPG) good dagger isn’t strongest weapon in terms of damage, but it still can perform one hit kill. My system has brutal rules for damage

  4. Erebus

    Agree with all.
    I checked the article to see if my own Shadow Lords ( ) falls in any of those traps and it happily doesn’t

    Anyway giving “damage rating” and damage rolls to weapons is the worst way of handling cinematic and thematic combat in an rpg (in my humble opinion).

    A Hero should be able to kill any foe with any weapon he is lethal with, as it happens in any fantasy story, as long as it plausible what he does.
    If he is attacking a full armored knight with a knife it should be more difficult than killing the same enemy fully naked (obviously), but not impossible, and neither take hours of game because “this weapon does only 1 HP of damage”.

    In Shadow Lords there is no “damage roll”, everything contributes to your ability to do what you intended to your enemy. And if you fail, your enemy does what he intended to you.
    You attack the knight with the knife, he gets to use his armor against you and if you fail you scratch his plate and he rams his armored fist in your face. If you have success you successfully find the opening in the armor where you can strike.
    No extra rolls needed.

  5. Lex W

    It seems like relatively few systems, and none of the big, successful ones actually manage to avoid these mechanics. Though perhaps Dungeon World does (not really “big and successful”, but certainly popular), but I’m fairly sure it has at least a couple of non-special special abilities.

    That said, it’s a good list, and yes, really all of these should go.

  6. Jason

    Re “Penalties as Difficulty”, I think there’s a situation where this makes sense.

    Let’s imagine the game’s resolution mechanism has a spectrum of results (e.g. 9=success), rather than binary pass/fail. Applying difficulty as a negative modifier to the roll makes things a lot easier and faster in this case. We can’t just change a target number/DC, as we would need to shift the whole set of result bands accordingly, which is just very unwieldy. Modelling difficulty as a modifier to the roll (and probably avoiding calling it a “penalty”) seems to be the most practical and elegant way to handle this

    • Jason

      Looks like the form mangled my post… I was trying to give an example where there are a set of possible outcomes or degrees of success, like “5- = crit fail, 6-8 = fail, 9-10 = success, 10+ = crit success”

      • Michael Campbell

        Agreed, a spectrum of outcome results that is fixed is much better suited to difficulties as penalties rather than having to print a table listing each range as adjusted for it’s relative difficulty.

        Mind you, once you had such a table, you could then have non-linear alterations. So critical successes could become disproportionately rarer in high difficulty activities and critical failures could become disproportionately rare if more mundane activities are attempted.

        It’s a classic debate for R.P.G. designers.
        Which is better?
        Simple formula Vs complex table.
        Some of us remember the explosion of rainbows that were the tables of the 80s.

  7. Michael Campbell

    On penalties as difficulties

    What expectations do players have about game mechanics and why are they disappointed when their expectations aren’t met?
    (I know my first thought would be “entitlement issues” but I’ll be the first to admit that I’m not actually a mind-reader.)

    So lets say the game has two different colours of die; White and Black.
    The pips of white die are added and the pips of black die are subtracted and a success is to have a total more than zero at the end of any die roll.

    Now our gunslinger is about to have a shootout at high-noon when he suddenly realises that some buckeroo; with a lever-action carbine at the window of the whorehouse, is taking a bead.

    So the player asks the question of his GM.
    “What am I looking at with the two targets?”

    The GM says; “Well, that makes the range `Medium’ not short and the window frame provides some cover, so, instead of asking you to roll against 2 Black die for the `duel’ against the lead-rustler (Mad Dog McKhrushchev) who is at `short’ range; I’ll ask you to roll `medium’ range with `slight cover’ for 5 black die.
    Not that I’m supposed to tell you what the target numbers are: just explain what is perceivable to your character.”

    And the player says; “How dare you! You know black die are only for the belly wound Mad Dog is about to give me.”

    Won’t the GM say; “It’s for both circumstances. Look it up in the player’s manual!”?

    Players who think that the rules of game X must apply to game T should be reminded that they are indeed playing game T. Or more accurately the variation of game T that exists within the GM’s mind.

  8. Leon

    Are there any combat systems that factor in reach in a realistic way? I take it everybody here knows how swords really work. I never liked the idea of rogues some how becoming invisible in the middle of a fight, especially since in real life a warrior isn’t muscle and hp, but keen senses amd sharp wits. It would be cool having swift knife wielding characters slinking around, trying to get in side the guard of larger warriors who could cut them in half.

    • Michael Campbell

      The newest iteration of Hack Master has a reach rule.
      He that has the weapon with reach gets the first attack outside of the conventional initiative clock.
      I believe the trial rules are free to download.
      I also rather like the initiative clock. The short-sword gets to attack slightly more frequently.

    • Oren Ashkenazi

      I guess it depends on what you mean by “realistic,” but I think the closest to what you’re looking for would be systems like Burning Wheel and Riddle of Steel, where getting closer to or farther away from your opponent is something you have to roll for, and will be different based on your weapon length.

    • Michael Campbell

      “I take it everybody here knows how swords really work.”
      I’m not so sure. I suspect the crystalline structure of metals and the advantages of different rates of dendrite growth, would slide past most of the lurkers around here.

      Basically, do you quench your sword in iced water or tepid oil?

      • I didn't feel like writing my name

        Tepid oil..?

        • Michael Campbell

          I’m not really sure what you’re saying.
          Are you asking about where chaos theory and metallurgy meet with regards to the laws of physics that mandate that dendrites initiate?
          Or that a bucket of oil in a blacksmith’s shop will be quite a bit warmer than standard room temperature once the blacksmith has brought the iron specimen up to the re-crystallisation temperature (A.K.A. red hot)?

          Since heat transfer rates are dependent on the difference between the two involved temperatures*, making the two temperatures involved; closer to each other, will result in a reduced heat transfer rate.
          A lower heat transfer rate (A.K.A. slower cooling) yields fewer & larger crystals (A.K.A. grains) while a faster cooling rate (quenching) yields more but smaller crystals.

          *Theoretically a heated object will never entirely cool-down because as the temperature difference approaches zero so too the heat transfer rate approaches zero.
          If you plot it out, you find the relationship gives you an asymptote.
          * Also, your air-conditioner will be most effective when the temperature inside is as close as possible to the temperature outside, because that’s when the insulation in your walls has the lowest heat transfer rate. So enjoy your July.

  9. ToodlePeep

    Can’t say I really agree about the static difficulty, having played a lot of these games I can’t say it’s ever come up as a problem once.

    The first and most important thing is that PBTA and other narrative games aren’t generally games where the players are to overcome a challenge the GM created, and they definitely aren’t trying to simulate a fantasy scenario. It’s improvisational story telling and failures are opportunities to spin the story off in a new direction, not setbacks on the path to completing the adventure that has been written. The mudslide and the avalanche might have the same odds of occurance, but it’s going to take your story in different directions when you fuck it up.

    In the example of how to handle something easy, don’t roll! if there isn’t a chance of a narratively dramatic outcome there’s no need to make the roll.

    Finally, it’s really common if you really need to make something particularly difficult in a PBTA to gate an outcome behind checks. “I’m going to take a shot at this robot/stab the master ninja” could lead to “ok, but first you’ll have to deal with his crackling energy shield/get past his hail of lethal shuriken” – it’s OK for a GM to nope a situation or add another challenge in front of something to make it more dramatic and tense because it’s again not trying to be any kind of simulation.

Leave a Comment

Please see our comments policy (updated 03/28/20) and our privacy policy for details on how we moderate comments and who receives your information.