Jun 30

Fewer skill checks, please.


Note to readers, this article was inspired by Table Topics # 68  – Setting DC’s in D&D  You can listen to that episode here: www.therpgacademy.com/table-topics-episode-68-dcs-in-dnd


Fewer skill checks, please.


Role Playing Games, by their very nature, are an odd mixture of free-form cooperative storytelling and rigid mechanics. Everyone approaches this mix according to their own preferred style of gaming. It’s not hard to see how different players and groups could play through the same adventure and come away with vastly different experiences. One group may enjoy convincing a powerful NPC to ally with them and give them aid and resources.  Another group might instead enjoy maximizing their in-game effectiveness through tactical decision-making and utilizing synergy between the race/class/ability combinations of the characters they play. Yet another may simply enjoy the fun and laughter they have despite their characters failing at every turn of the adventure and accomplishing nothing of note.


My point is that there is no wrong way to play.  Play your game the way you and your group will have the most fun. For most groups, things will change from game to game. One session might be all role play, the next mostly combat, and the following all skill checks. A good mixture of these elements should make for a good session most of the time. But there is a specific situation where, in my opinion, no-matter the style of play you prefer, the mechanics get in the way: skill checks.  They are an easily-abused method to create a challenge or generate false drama. And they often create barriers to the game that players want to play.


I want to advise that you follow the often-quoted mantra: don’t roll the dice unless failure will be interesting. Not doing so can cause the game to needlessly lose momentum and, even worse, create a bottleneck in the story for no reason. By their nature, skill checks often come right as things are about to get exciting or in the middle of narrative/role play scenes. I believe that both of these situations would be better served (most of the time) by skipping the skill check unless absolutely necessary.


Consider this example:

Player (to an NPC) “You said you didn’t know anything about the queen’s disappearance but yet you are now trying to sell a ring she was known to wear!” (To the DM) “I want to pull the whimpering guard off his feet by his neck and bare my fangs at him.”

DM “roll an Intimidate Check, but use Strength instead of Charisma since you’re holding him off the ground.”

Player “19”

DM “The guard is brought to tears and visibly slumps in your arms. He gasps out, ‘she said she’d kill me . . .’”


Probably seems pretty standard, but do we really NEED the step in the middle?


Player (to an NPC) “You said you didn’t know anything about the queen’s disappearance but yet you are now trying to sell a ring she was known to wear!” (To the DM) “I want to pull the whimpering guard off his feet by his neck and bare my fangs at him.”

DM “The guard is brought to tears and visibly slumps in your arms. He gasps out, ‘she said she’d kill me . . .’”


The same result is achieved either way. However, by skipping the skill check roll, which the player was most likely going to pass anyway, the momentum of the scene continues to build rather than hitting an odd and useless speed bump.  And thinking of the in-game narrative, does a lowly corrupt guard have the internal fortitude to resist a hardy adventurer? Especially one that has bared fangs and has him lifted off the ground by the neck with one hand?  In my games, no.


I don’t ask for a check in this example for a few reasons. I want to reward the role playing. The players are most likely going to pass the check  anyway.  The humor that might come from the failure is not equal to the lost momentum of the scene. I want the player to feel successful whenever possible. And lastly, I don’t want to roll too often, so that when I do call for a roll it will always seem important.


Beyond the possibility of breaking the group out of an immersive role play scene, unnecessary skill checks can also create a bottleneck in your story and create more work for yourself as the DM. Think of a successful skill check as a key.  These skill check keys open doors leading further into the story that you as the DM are trying to weave around the players and their characters. What’s the best way to ensure the players open the doors you want them to? Give them the key.


A player wants to intimidate a guard. Why? Because the story is what the guard tells them after his will has been broken. A player wants to find the secret cache in the noble’s room. Why? Because the story is what the noble woman has hidden there. A player wants to pick the lock of a door in a dungeon. Why? Because the story is about what’s on the other side (or it could be about the charging orc on THIS side of the door, but that’s a different matter).


Now there are, of course, lots of examples of when failure on a skill check has created moments that are famous (or perhaps infamous) in gaming groups. So I’m not saying that you should do away with them completely. I’m saying that you should to use them sparingly so as to not put barriers between the players and the fun of the story.  Sometimes they ARE needed, but less so than many DMs think.


Let’s return to our examples.


We have guard that needs to be intimidated.  Do we ask for a roll? It depends on what’s more important to the story. Do we want to have the possibility of drama or a new complication that failure of the roll provides? Or would we rather make sure that the players learn that the sweet tavern owner, who’s been so kind to them, is running the city’s drug trade?

We have a secret room hidden in the nobleman’s chamber. Do we ask for a roll? It depends on what’s more important to the story. Do we want the tension of possible failure and/or being caught? Or do we want the players to find the decomposing body of the noblemen that they supposedly were talking to all night and who had invited them back after the party for a secret meeting?

We have a door that needs to be opened. Do we ask for a roll? It depends on what’s more important to the story. Do we want the possibility of failure so the barbarian gets to use her lock picks (i.e. the heel of her boot) instead? Or do we want the players to get access into the depths of the labyrinth where the minotaur is said to roam?


Related to this topic is the very popular concept of failing forward. If you happen to not be familiar, it’s the idea of success at a cost when the dice dictate failure as opposed to using a typical binary pass/fail result. Instead of simply failing to pick a lock, the door opens loudly and draws the attention of some nasty monsters.  Instead of simply failing to bribe the guard, the guard takes the cash and also alerts other corrupt guards who want to want to get paid. Instead of simply failing a climb check, the rope breaks and the party must now split up to find another way around. This concept offers a solution to the bottleneck issue as well. The adventure needs to continue even after a failed check, so it now adapts and takes a new path. My method of eliminating unnecessary checks accomplishes the same task of keeping the story moving. But instead of forcing you to react to failure to do so, it removes the chance of failure. Of course, when the potential of failure is interesting or important, a skill check can and should happen. And for the record, I make use of the fail forward method, when I do ask for rolls.


Consider two final examples to wrap up this idea.


A party of adventurers needs to scale a wall to enter an abandoned castle. You have lots of cool encounters planned for them to find inside. In your mind, it doesn’t make sense for the castle to be too easily accessible, so they have to find a way to get in.  You plan a secret passage beneath the calm waters of the turgid moat, but the players decide to climb over the walls instead. Do you ask them to make a check here?


You certainly could. Perhaps if one of them falls, it would provide a moment of levity as you (or they) creatively narrate how it looks when they fall. You could even have them fall into the moat to possibly discover the way in that you had envisioned all along.  But what if they all pass the climb check? Then nothing really happens, except everyone gets to roll a die or two.


What if instead you just said, “Thorn the rogue is proficient at climbing, It’s easy enough for him to climb the walls and then lower a rope to aid the rest of you. Soon enough you are all on top of the front battlements.”


You reward the party for having a rogue as well as rewarding the player of the rogue for being proficient in climbing.  This is kind of an extended version of a passive climbing score, or older editions’ mechanics of taking 10 or 20. Since there is no external threat (trying to climb quietly, avoiding a guard, getting up before sunrise, or avoiding a barrage of arrows) the players are able to take their time so ultimately, no roll is necessary. The momentum of the story is not interrupted by slow dice rolling and nothing unexpected pops up that derails the story. A minor obstacle stays minor and the players get to the good part of the story.


Last example.


A group of thugs crash into a tavern. The stout fighter wants to flip a table to create cover against the thugs who are firing crossbows. Does this require a check? It could work either way. If I ask for a check, there’s a chance of failure. At the very least, the player wastes a turn. At the very worst, a complication arises and the fight becomes more deadly or more time consuming than you originally planned. No-matter the specifics, the player doesn’t get to do something cool like we see in the movies. If I don’t ask for a check, I’m giving the players an advantage in combat. I’m also letting them be the big heroes that the game wants them to be. Either way can be entertaining and fun. The choice depends on what your group wants and expects.


The concept of automatic successes is not new. Several games use them. I’m not sure where they originated, but I compare it to my (limited) experiences with Dungeon World. In that game, characters can always perform the actions that are most stereotypical of their class. A ranger feels like a ranger because he can always do rangery things. In my games, I’ve developed a simple way to adapt this idea to the established mechanics of D&D. If the PC’s key stat is 15 or higher, I’m going to allow auto success on checks that make sense. Rogues can pick locks. Rangers can find tracks. Wizards can research. Bards can recall lore. Clerics can perform rituals. By allowing characters to have auto-successes on skill checks like this, it reinforces within the narrative that these characters are the best representatives of their classes. Plus, in my opinion, it lets everyone have more fun and keep the game moving smoothly.


In conclusion, I don’t think there is an easy always yes or always no answer here.  Sometimes failing is fun, and it makes the story more interesting, realistic, and dynamic. But often, unnecessary skill checks slow the game down for no real reason. Players can get bored while they wait their turn or might not even get to participate if another player gets the necessary roll first. And if no-one gets lucky with the dice, you might end up simply handing them the info they need to move the story forward. The benefit of the auto success is that you allow the players to feel like their characters are competent. It keeps the story moving. Whether it’s based on player narration and role play, PC stats, or just what makes sense in the moment, letting things happen with out rolls creates a better, smoother game. And when you DO ask for skill checks, they will be more important because they are not happening frequently.


It’s your game world. Give the players the keys they need to explore it and keep them role playing as often as possible instead of just rolling. And of course, make each die roll matter by eliminating the needless ones.


Use this as a guide:


The DC Do I call for a check?
0 – 12 almost never
13 – 15 Some of the time. Failing forward as necessary
16 – 18 Most of the time, Failing forward as necessary
19+ Almost always

I hope this has been helpful and remember: if you’re having fun, you’re doing it right.

~ Michael

Comments and Feedback are always welcome.



E-mail us at Podcast@TheRpgAcademy
Follow us on twitter @TheRpgAcademy
Visit our Facebook Page
Join our new Google+ Community page: The Rpg Academy
Become a backer: Patreon.com/TheRpgAcademy and get episodes early and other great rewards


Skip to comment form

    • Jen / Pixelscapes on June 30, 2016 at 10:19 am
    • Reply

    Totally agreed. The most deflating scenario in a game is when someone does a great job of explaining how they’ll do something… you know, good RP?… and it’s sensible, and then the DM calls for a roll and it comes out abysmally low. Then, oh well, guess it didn’t matter you had that great plan after all, why did you bother?

    Passive scores make sense, too. Of course if it’s a particularly challenging thing, then a roll may make sense — and even then, a low roll may represent it taking longer than expected, or being noisier than expected, or some other obstacle. Not “Nope, it just doesn’t work.”

    Of course, this all ties back to your key point of only rolling when the failure is interesting.

    1. Jen,

      Thanks for reading. It’s a skill of knowing (guessing) when it’s right to ask for a roll or to let the scene move along. Hopefully a few people will move in that direction. I also prefer failing forward vs. nope it just doesn’t happen.


    • Melissa on June 30, 2016 at 5:55 pm
    • Reply

    Hey, found this on Reddit and I really appreciate the hard work and thought you put into this post.

    As someone making my own RPG this has been one of those nagging questions in the back of my mind. Do I make a complicated skill system and make them check for everything, or do I keep it simple and only ask when necessary? I think your post has really helped put the thoughts I’ve had into words on the page.

    Thank you for posting.

    1. Melissa,
      Thank you for reading and commenting. It means a lot to us to interact with people. Keeps us from feeling like we are shouting into the void. If you are into podcasts please consider giving us a listen and if you’re not, try us anyway. Maybe we’ll change that 🙂


    2. Melissa,

      Excellent. Happy to help in any way. When I was younger i often called for skill checks more than I do now. I’m not exactly sure what the genesis of it is, but I find the game more fun with less skill checks and more ‘you can just do that thing ’cause you’re supposed to be good at it’ it’s a style choice for sure, so decide how you want your game to feel and why and then try to mirror that with the mechanics. Caleb is our Crunch guy, if you wanna discuss your game and how it works he might be able to help.


  1. Love the idea of ‘failing forward’, I honestly hadn’t thought of that, and will likely be incorporating that in my campaigns now.

    I think another good way of knowing when to force a roll on something a character can do is pressure. A thief can pick a lock easily most of the time, but can he do it when there’s a monster charging at the party? It’d make more sense (and have more dire consequences) if he fails while under pressure.

    Interestingly enough, my players are quick to explain abysmal rolls in-character. A bluff check to not react when they see a friend tied up while speaking to guards rolls low… despite a +15 to bluff, because they are worried and/or angry at seeing them mistreated. A thief rolls a 2 in stealth, giving themselves away out of shock as they see a dreaded enemy while sneaking up on a camp.

    Amusingly, my PCs tend to crit-fail when things fall really ‘in-character’ for their backgrounds. For example a character who is simply so bad at magic and handling magic that she crit-fails any arcana and religion check she tries, and has thus handicapped her character of her own free-will to take much longer to attune to magical items, and came up with rolling a 1d4 to see how many days it takes when it’d usually take a long rest.

Leave a Reply

Your email address will not be published.