About the Author
Mass Effect
Final Fantasy X
Batman:Arkham City
Borderlands Series
Weekly Column
Champions Online
World of Warcraft
DM of the Rings
Good Robot
Project Frontier

GM Advice: A Learning Mechanic

By Shamus
on Wednesday Mar 4, 2009
Filed under:
Tabletop Games


I had a reader ask about the mini-game that appears in my D&D campaign. A few people have expressed interest in it, and I thought it might be worth a look. I don’t pretend this is clever or innovative. This is very much a system I cobbled together as I was groping around trying to simulate a character learning.

In our game, I had a situation where one player was working to translate a “book”, which was several pages of backstory they needed to know. Any time his character had enough downtime, the character could sit down and spend a few game hours attempting to translate the next section. If he was successful, I’d hand him the next couple of pages of the backstory. If he failed, his character still learned from the attempt and his future chances of success went up.

As an aside, if you need to feed your players a bunch of backstory, this is a pretty good way of going about it. They will naturally recoil if you hand them 20 pages of prose to read all at once. Even if they’re into it, they will tend to skim it quickly so that they can get back to playing the game. On the other hand, if you make the backstory into a reward and dole it out a couple of pages at a time, they will be a lot more excited about it. Players can pass the material around the table without any one of them being taken out of the game for too long. It builds a bit of mild tension as they look forward to the next reveal.

Oh boy! We found a couple more pages of the Lost Tome of Naughty Evilry. Hand it over, Shamus.

The Goal

D&D has mechanics for long-term learning. (Skill points and leveling up and whatnot.) It has mechanics for immediate goals. (Roll the dice to open the lock.) But it didn’t have anything for mid-range character projects. Most gameplay mechanics are set up so that characters learn and grow from success. The more success, the more XP. I wanted a mechanic that would simulate an activity that was inherently driven by trial-and-error, and where (this is the important part) the character got gradually better at the activity as time went on. Learning would be fast at first, but progress would be slow. Later on, learning would slower, but success would be more frequent. In general, the system is good for specific tasks like “find cure for the plague” and bad for more generalized tasks like “learn biology”. Situations where you might want to use something like this:

  1. Looking for a secret formula or cure for a disease.
  2. Solving an abstract puzzle (that is, the character must solve the puzzle, not the player) in-game.
  3. Working to translate or decipher some runes, text, a book, a map, some symbols, or what-have-you.
  4. Inventing or building something new.
  5. Cultivating or breeding rare plants or animals.
  6. Researching a special plot-specific magic spell.
  7. Making sense of the parts of some ancient technology they’ve unearthed. (What the parts are, how they go together, what they make, and how to use it.)
  8. Using clairvoyance to investigate a [whatever].

I also wanted a system that would work independently of their other skills, levels, and abilities. They could learn this activity organically, without waiting for the next level up. This is great for simulating things that anyone can learn to do with practice, but are usually self-contained skills that aren’t going to be useful later in the life of the character. If the character is working on a Rubik’s cube, this game could simulate their progress as they learn the puzzle. Once complete, they would be able to solve a cube at will from that point on, but that learning won’t enable them to instantly solve a completely unrelated puzzle.

The System

(I had an alphabet novelty die that went along with the translation game, but let’s just assume we’re using Ye Olde d20.) The game goes like this:

The player writes down all the numbers from 1 to 20 on a notecard. Every time they roll a number, that number will be crossed out on the card. If they roll a 15, then they cross out 15.

Each attempt needs to simulate a stretch of in-game time. Hours of labwork, meditation, tinkering, writing on the chalkboard, or whatever is required.

When they make an attempt, they roll the d20. If the resulting number is already crossed out, then the action was a success and they get their reward. If not, they still get to cross out the number they rolled, which will improve their chances next time around. Using a d20, they have no chance of success on their first attempt, and a 5% chance on their next attempt. Every failure improves their chances by 5%, and every success moves them closer to their goal. You decide ahead of time how many successes it will take to reach their overall goal. (For our game, I had the book broken into 13 sections. So the character finished the translation after 13 successes.)

That’s it. Pretty simple, really.

Drawbacks & Quirks

There needs to be some sort of cost associated with an attempt, or the player can just have their character binge their way through the problem. Oh? Mini-game? Whatever. I’m just going to start rolling the dice. Tell me when I win. Time should obviously be a cost, but if you can work in another cost (money, lab equipment, magical components, computer time, energy) it will make things more interesting.

The system as portrayed doesn’t use any character stats, which means Grogtor the Barbarian will be able to decipher the runes just as quickly as Wizbeard the Mage. Obviously you’ll need to establish some sort of prerequisites for doing the work. (i.e. you must have an INT of at least 14 to even begin the task.) Ideally they should get a benefit from any character stats. (Perhaps a high INT score will make attempts take less hours or use less resources, for example.)

Also note that total success is inevitable. This system is designed for something that can be done with trial-and-error over time. As presented, you can’t ever fail at the task, unable to proceed. Unlucky rolling will just make it take longer.

The Numbers

Just to give you a feel for how the game will play out. I ran the numbers, and on average a player will need about 24 attempts to complete a task which requires 10 successes. The breakdown goes like this:

Success Needed Average Required Attempts
1 6.27 avg. rolls
2 9.38 avg. rolls
3 11.79 avg. rolls
4 14.00 avg. rolls
5 15.90 avg. rolls
6 17.71 avg. rolls
7 19.39 avg. rolls
8 21.04 avg. rolls
9 22.42 avg. rolls
10 24.07 avg. rolls
11 25.39 avg. rolls
12 26.75 avg. rolls
13 28.02 avg. rolls
14 29.34 avg. rolls
15 30.73 avg. rolls
16 32.09 avg. rolls
17 33.24 avg. rolls
18 34.43 avg. rolls
19 35.65 avg. rolls

So if you decide the character needs 15 successful lab sessions in order to find the cure for the zombie plague, then the player is going to roll the die about 30 times. If a lab “session” is a full in-game day, then a player character will need to work for about a month solid to find it.

Other Notes

I’d give a description for each success:

  1. You managed to isolate a viable sample of the zombie virus. You can now replicate it in the lab for study.
  2. You discovered a way to observe the virus work on tissue samples without needing to infect actual people.
  3. You managed to isolate the proteins that mumbo-jumbo the victim’s cellular whatchado.
  4. You learned how the virus works. Now you know how it spreads. (Blood, saliva, topical contact, Facebook invites, etc.)
  5. You can now make a cure from a single tissue sample that will destroy the virus. Downside: It only works on the zombie from whom you took the sample, and their brain is now mush. So it’s useless, but one step closer.
  6. You’ve figured out how to make the cure general-use. If someone is bitten, you can dose them to keep them from turning as long as you can get them to the lab while their brain is still in good shape.
  7. You’ve streamlined the cure so you no longer need the lab. You can just carry the cure with you and dose someone if they get infected.
  8. You’ve synthesized a general use vaccine. You can now make people immune to the plague!

You get the idea. Hope you find it useful.

Noodling around with this sort of thing makes me want to run a game again.

Comments (59)

  1. Seb says:

    I love it !

    About using stats : you could allow the player to cross one number on his card per stat bonus point (like with INT 16(+3) they could cross out 3 numbers). This would give a good head start to characters with the right stat, but not too high.

  2. Uncle Festy says:

    *looks at the single comment line*
    In other news, this is awesome. Makes me wish I wasn’t an utter failure of a DM. v.v

  3. Jeremiah says:

    One of my favorite things about Burning Wheel is how skill/stat advancement works:

    1) Using skills/stats works towards increasing them. You need tests of differing difficulties to advance (success/failure doesn’t matter except in a couple of special cases).

    2) You can also have a trainer train you.

    3) Or if you have downtime you can practice to become better.

    Now, Mouse Guard has something similar, but it’s a lot more simplistic. You just need a certain number of passed and failed tests to increase something. For instance, to go from a 3 to a 4, you need 3 successful tests, and 2 failures.

    Also note that total success is inevitable. This system is designed for something that can be done with trial-and-error over time. As presented, you can't ever fail at the task, unable to proceed. Unlucky rolling will just make it take longer.

    I don’t know that I like the idea of success being inevitable (more on that in a second), but on the other hand the fact that failing a roll doesn’t stop things dead in their tracks is very important to me. A lot of role-playing games just don’t handle failing well. Oh, you failed to pick the lock, sorry. This is another reason I enjoy games like Burning Wheel & Mouse Guard so much, failure isn’t a wall, it’s a speedbump, usually introducing some complication or twist: okay, you pick the lock, but you get caught in the act by a guard.

    This leads back into me not liking success being inevitable. Sure, in some instances, just the time investment can be a huge issue. But, I think I’d be more interested in it if, with a failed roll, some false information was slipped in with all the real stuff (this probably wouldn’t work too well with your system as it stands, because there’s likely going to be a lot of failed individual rolls). But when I think of someone deciphering some huge text, I think of time, and I think of mistranslating portions of it.

  4. El Quia says:

    Interesting method. I loved the idea of giving out the background as part of a minigame when I read your transcription and now that I know the mechanics, I found them really interesting. Both the book and the phylactery where interesting minigames and I’m itching of trying the same approach in my games.

    I love your RPG posts. Although I play videogames and I love them, my passion for tabletop RPGs knows no bounds. I got into your blog mostly because your RPG posts, although the rest of the content is an interesting read.

    If you run a new campaign, I would love to read a transcription of it. It’s just the expresion of a wish, as I know that the transcription of the last one was more of a pain than anything else.

    @Seb: nice idea! I think that that bonus could make it easier to an intelligent person start having succeses earlier than other characters. But maybe I would give them half their bonus in numbers to cross.

  5. Patrick J McGraw says:

    This reminds me a bit of the skill system in Call of Cthulhu.

  6. Tommi says:

    I’d say that without a price for failing every check, or every n checks, this is but a pacing mechanic with little connection to anything. In time-sensitive situations time spent is by definition a price, so they are a good fit for this mechanic.

  7. Mari says:

    I love this idea. Now I need a group of suckers – erm, players – on whom to use it. Shamus, if you ever get ’round to running an online campaign sign me up since that’s pretty much the only way I get my “tabletop” fix in anymore.

  8. Flying Dutchman says:

    The Facebook thing is pure genius. I salute you!

  9. Mike says:

    Oh, I like this. A lot. Now I just have to figure out how to adapt it into the d100 system in the game I’m running.

    Although, since all the players are normally D&D types, I could just have them use a spare d20 for it, but that feels like a cop-out.

    And with a few moments more thought, the answer becomes clear – split the table up into sets of 5 numbers at a time, and boom, compatible. Thank you, Shamus. This will be…quite handy to have. Muahaha and all that.

    Also, Seb, I like the concept there; especially for players who have invested much of their XP into intelligence traits, that’s a nice bonus for them in a system like this.

  10. Mordiceius says:

    Absolutely fantastic game mechanic. Talk like this makes me want to stop being lazy and and finally run the online campaign I have been wanting to run for a while.

  11. Kleedrac says:

    Wow … this is brilliant!! Might I be so bold as to suggest that perhaps a character’s stat could determine how many failures to cross off a number? ie the Wizard with 18 INT crosses off 1 through 18 after a single failure on it but 19 & 20 (being above his INT) will take 2 failures to cross off? (This is all assuming an INT based task) Whereas the Barbarian with his lowly 8 INT will take multiple failures at most of the scorecard to solve the puzzle? Still possible and inevitable but it’s going to take the Barbarian a lot longer. In this way it could also be possible to apply this mechanic to a DEX based task such as learning to throw a baseball. Either way this is getting added to my notes file for future DMing :)

  12. Sydney says:

    You could have, say, four “marks” on each number card. With high enough INT, you get to eliminate all four marks at once, making the next roll of, say, 13, mean a success for sure.

    Otherwise, roll a D4 (or flip a coin, or roll a D6, depending how many marks you chose) and cross out the mark to which THAT corresponds. Next time you roll 13, your chance of succeeding are higher – but not as high as if you’d had enough INT to just cross out all four marks at once.

    Bottom-of-the-barrel, can-just-barely-do-this-task characters might only eliminate one of four marks per “partial success”. Higher-int characters might get two rolls per partial success, and get to eliminate two marks, or maybe even succeed the second time. And so on.

    This symbolizes You can close to the next step on the way to the solution, but screwed it up and had to start over. You learn not to make that same mistake again. Back in high school chemistry, this happened to me approximately three metric fucktons of a lot.

  13. gi says:

    I’m running a modified campaign world based on what you did for this site Shamus. My guys are researching the book and the orb at about the same pace as your people did. I came up with similar system since this wasn’t posted yet. Enjoying seeing the tools in your tool box for being a game master. Thanks!

  14. John says:

    Great idea, but I tend to play with groups of game-breakers (sadly), or, to be a bit more generous, “out of the box thinkers”. In a fantasy campaign, they’d stop any serious plot stuff to find some way to magically decipher a book. In a future setting, they’d use the Internet, or some other Deus ex Machinica.

    On that note, maybe if you want a “support ability” for a particular challenge in a d20 context, think about allowing a number of dice rolls up to the value of the relevant attribute (minimum 1)? So translation (int-related) would give Bob the Barbarian 1 dice roll per time period, but William the Wizard would get 3 or 4. As your stats-chart shows, several time periods will still be required.

  15. droid says:


    If you are using d100 then you have the advantage of being able to tweak more easily: a low int character could cross off fewer items, and a high int character will cross off more items, say six at a time (add the int modifier(or your systems equivalent) to 5).

    Though it seems weird to me that the player marks which dice rolls will lead to success, it would make more sense to just keep track of the number of failures rather than what numbers they actually are.

  16. Dan Beck says:

    I like John’s variation: The amount of progress in a given time period is reflected by having one or more rolls (based on the relevant attribute bonus), so the PC with the best bonus is the most logical choice to attempt the learning task, yet progress isn’t totally closed even to groups who’ve min-maxed themselves away from the relevant attribute. Simple to use, easily remembered, logical, and rewards role-playing. That’s a keeper!


  17. John says:

    (yet another John)

    A thought for simplicity – isn’t this the same as having a target number for success which rises? That would be simpler than having a list to cross out.

    Example: (using d20 and DnDish approach)
    * Foobaria the wizard starts with a success number 16 (20 – 4 (her int bonus))
    * If she rolls her success number or higher, she gets a success.
    * If she rolls lower than her success number, the success number goes down by one.

    Feeding off on previous comments, perhaps a bad roll (‘1’ to be traditional) causes a reset of the success number up by three. It would be theoretically possible to push your sucess number up past 20, thus requiring some additional effort (spell, assistance, plot clue). The higher the “bad roll” level, the more possible it is to get backed into a corner.

  18. Bryan says:

    Great stuff Shamus! And I have an immediate use! In my current campaign, the group has finally found the villanous mage who has been poisoning the surrounding areas. They don’t yet know of the orcs who are tunneling between the poisoned lake and the town. They also don’t know how to neutralize the poison. They have decided to search the keep for additional clues. I can write up some “backstory” documents to find in the keep. For the fighter, I can have him decipher the coded messages between the orcs and the mage to discover the tunneling, and therefore prevent a disaster in town. The mage and the cleric can work together on the mage’s notebook to discover a cure for the halfling village who have already been disfigured by the poisons. All using your system.

    No offense, but I am totally ripping this off!

  19. Luke Maciak says:

    I just wanted to say that the idea of spreading Zombie plague via Facebook invites made me chuckle. Not to mention I found it totally plausible.

    @John – well, you could still use it if your players decide to really go out of their way not to have their characters involved in the process. Let’s say they go an hire a wizard NPC to do magically translate the book for them. It will take that wizard time to do it, no?

    So make them roll for it – use the same technique Shamus described here. Wizard tells them to come back in a day, and when they show up they get to roll a dice to see how far he got in the translation process.

    If the players are getting annoyed that it takes so long, have the wizard suggest that one of them could tackle the simpler parts of the translation. This way they get 2 dice per attempt. Eventually they might figure out that it’s just better to cut the guy out of the loop, and translate the rest of the thing themselves for free. :)

  20. For people running skills-based systems, the mechanic is basically something the rules already take care of in one way or another. For some games, what Shamus is coming up with an ingenious way to bolt on to D&D is the basic mechanic of the system! Not just newer ones, either–Bloodbath 2000 . . . uh, I mean Twilight 2000 springs to mind.
    But the basic plot-oriented notion of arranging these things in steps, so players can get a series of partial successes leading to the main payoff, or divide unmanageable expository lumps into smaller doses of information that seem like a reward rather than a burden, is something that could apply to any campaign.

  21. Trianglehead says:

    @mari Good call! Put my name on the list for an online game as well, Mr. Shamus! :)

  22. You know, handing out a few pages of backstory at a time is actually pretty smart.

    I think I learned a new trick today.

  23. Josh says:

    I’ll definitely be using this mechanic in my Savage Worlds games. In fact, the Savage Worlds rules give me an idea for making the difficulty of these tasks adjustable: just use different dice for different levels of difficulty.

    For something relatively simple (translating a Latin book with the help of a dictionary) use a d4 or d6. For something very difficult (translating the Necronomicon from its original infernal language using only Wikipedia references) use a d20.

  24. Dev Null says:

    Learning would be fast at first, but progress would be slow. Later on, learning would slower, but success would be more frequent.

    RuneQuest used to have a nice mechanic for this style of learning. You had a percentage skill for doing something, which might start based on a relevant stat, but would generally be pretty low – say 10%. The first time you try to use it, you have a 10% chance of succeeding, BUT you have a 90% chance of improving your skill. If you skill gets up to 30%, you have a 30% chance to succeed, but only a 70% chance of getting better at it. You control the pace at which they progress by how often you let them roll to improve (maybe every time they try the skill; maybe once per game session) and how much their skill goes up when it improves (1%? d4%? d6%? Based on a stat?) Associate a cost (in gold, xp, time, need for trainers or materials…) to keep them from just sitting there saying “I try again” over-and-over til it works.

    As an aside; did you just propose adding a Flash mini-game to tabletop roleplaying Shamus? *evil grin*

  25. Dev Null says:

    Interesting. Your new system for moderating comments doesn’t give me the opportunity to go back and fix my typos. I liked the fact that I could edit comments in the old version.

    Edit: Even more interesting; I can edit this one. The use of HTML in the previous comment?

  26. Decius says:

    An interesting twist to the “Zombie cure” plot would be to have the research reach a point where further samples/field work was required… and until you got to that point in the research, nobody knew what field work needed to be done.

    That allows for a setting where both the egghead scientist and crazy mercenary can shine.

    Yeah, I’m totally going to steal this. If I ever publish anything, I’ll try to remember to credit you for the idea.

  27. Sam says:

    I really like this system. I also now wish I was running a game, because I’d use it.

    I’d also be interested in an online campaign if you’d be okay with letting total strangers like myself take part in said campaign.

  28. John says:

    I’d like to throw in an endorsement for Burning Wheel, too. It’s an indie system with strong mechanics, but also a strong sense of player driven narrative (Unfortunately, you only get to roll d6s — I understand, that’s sometimes viewed as a negative).

    It’s not perfect but it’s dirt cheap. The core books are about $30. Yes, there are extra books, but they’re optional.

    I realize that this post isn’t completely topical, but seeing mention of BW, I wanted to chime in.

  29. bryce says:

    This is a great mechanic but there is no drawback, or penalty, only lost time.

    As a proper GM I believe that there should ALWAYS be a penalty for failure. I wouldn’t do any in game to the characters, as the goal is to make the players feel the frustration that their charactors are feeling at each failure.

    The way it would work is with every missed roll, I would have it that they get to cross off the rolled number on the list, and then I get to put one item that is on the game table itself inside of them. No fair rummaging about for anything not physically on of the table proper.

    This gives the players the proper anxiety about undertaking the task and has the added benefit of having keeping the play space nice and tidy. Suddenly you don’t have to nag the players to keep the flaming hot cheetos off of the battlemap.

  30. Brandon says:

    This is a very interesting trick, but I don’t understand why you’ve stepped so far outside of existing skill structures. I think there could be just as elegant a way to do this in which you do skill checks instead of just rolling a D20. I’d elaborate more but I’ve been sick for 14 days, so my proposed alternative will have to wait.

  31. Bryan says:

    @ Bryce:

    “This is a great mechanic but there is no drawback, or penalty, only lost time.

    As a proper GM I believe that there should ALWAYS be a penalty for failure. I wouldn't do any in game to the characters, as the goal is to make the players feel the frustration that their charactors are feeling at each failure.”

    In the example I cited above, (#18) lost time in itself can be a failure. If the fighter does not decipher the military notes in time, a whole town wil be flooded with toxins, disfiguring ~1500 residents. If a cure cannot be found in time, the disfigurement could begin to mutate, requiring even more time to find a cure. I know that these would be considered penalties in my group.

  32. bryce says:

    Bah! That’s feel-good hippy DMing. The only way a player can truly feel the agony of a town full of NPCs being disfigured is to have them insert a Grenadier Miniatures hireling pack (including the original verson of the torchbearer) into an orifice chosen by d12.

  33. Jabor says:

    If “taking too much in-game time” isn’t a significant penalty, you don’t have enough of a handle on the pacing of your game.

    That’s the stock response I give to anyone who claims Psionics are ridiculously overpowered, too.

  34. SiliconScout says:

    Glad my simple email was so well received there Shamus, and indecently you explained it far better than I had hoped.

    Such a simple idea really and one I will adapt to our home brew game. It’s a % skills based system for the most part so adding this kind of thing could work well.

    Make your skill check once an interval level of successes on the check = # of rolls you can make (probably 1 per 10’s die). Take has 40ish successes to make, etc.

    Simple, elegant and awesome. Thanks for that.

  35. vdgmprgrmr says:

    You could make it so that the character’s ability modifier is changed to a fraction, and use that as a mark-to-attempt ratio. So Wizbeard would (having an intelligence modifier of +4) have a 4:1 ratio, which means for every one attempt he makes, he would mark off four numbers. And Grogtor would (having a -4 INT modifier) have a 1:4 ratio, meaning he would only mark off a number every four attempts he makes. This way the average characters (-1 to +1) still have a 1:1 ratio, but the characters who lie further away from the average would have a noticeable handicap/affinity for the activity.

    I really like the system, too.

  36. Jonathan says:

    If you want to run a PBP (play by post) game, I’m up for it. Giant In The Playground or Myth-Weavers…

  37. Hal says:

    Shamus, I wish I had known the details of your game mechanic when I ran your campaign. I kept the book translation (and the Orb deciphering, incidentally) in the game, but the mechanics were quite a bit different.

    The translator was a bard, so I had him make a Bardic Knowledge check everytime he wanted to do a translation. The catch was that, whatever his check result was, he subtracted that number from 24, and that was the number of hours it would take to translate the next chapter of the book. I gave him a +1 bonus to the check for each chapter he’d translated already. By the last chapter, he had something like a +17 bonus to the check, so he’d be able to translate quickly.

    The upside of this was that it meant a lot of in-game effort in the beginning. When translating one chapter took 12 hours or more, they spent considerable time on it in character. I liked this.

    The bad part was that it wasn’t much different from any other skill check. “I rolled a 15.” “Okay, here’s your chapter.”

    On a side note, having 3 players in my campaign was almost perfect for giving the players distinctive roles. The bard was the translator of the book. The cleric was the bearer of the orb, and the one who deciphered it. The dwarven fighter, not really suited to either of those, got to be at the front of the party when dealing with the intrigue for the coming dwarven invasion.

  38. Sam says:

    Aah, roleplay post, the reason I started reading this blog. I like the idea of a diceroll leading directly to a small tidbit of exposition as a tasty tasty treat, especially coupled with a gradual advance to success. Especially as you could easily tie any number of quests to the system.

  39. Saint Rising says:

    Hey, Shamus, I was reading this, and I found something incredibly silly. So silly, in fact, that I wanted to make a book cover for the Tome of Naughty Evilry because I’m looking for new creative and silly things to occupy my time with.

    Full credit would go to you, of course.

  40. Yar Kramer says:

    You know, Shamus, the concept of “failure as speedbump instead of roadblock” fits very well with your philosophies and opinions on the way game difficulty should work, like you described in Reset Button.

    One problem I have with this is that success the first time is impossible, and while this is fine for the trial-by-error type of “learning,” it wouldn’t really work in a situation where there’s at least some chance, no matter how infinitesimal, of getting something other than “learning from your mistakes” right the first time.

  41. Jabor says:

    Perhaps you always get a success on the next section with a natural 20 (i.e. that starts out crossed off), and to counteract that, always fail on a natural 1 (i.e. that never gets crossed off)?

    That would fit a little bit more with D&D norms, I think.

  42. Nico says:

    Assuming you’re using standard stat arrays I’d be tempted to have failures work like this – roll a 1d20. If you roll equal to or lower then your relevant stat you can make an additional roll to unlock progress in what you’re working on (per Shamus). Therefore the 16 INT Wizard only fails on a 17,18,19 or 20 (an 80% chance of success), while an 8 INT Barbarian only has a 40% chance to succeed. If you don’t like the concept of automatically succeeding on 20’s or higher you can also make the 1 an auto-fail.

    I would handle failures 1 of 3 ways.

    1. Failures could slow down progress, erasing portions that have already been unlocked.

    2. Failures could reduce the quality of the final piece. A zombie vaccine might need to be reapplied often, or use rare ingredients, or not work in 10% of the population.

    3. Enough failures could also, obviously, halt work from proceeding. Fail X or more times and you are stuck with whatever progress you made so far, but no more.

  43. I was iffy about whether I’d use this in my own games, Shamus, but it just occurred to me that *this is the way I like to play video games*. I like them to start out difficult, so I’m having to challenge myself and pay attention to every little thing, then gradually become EASIER as I become more adept/acquire stuff/gain levels. I don’t mind if there’s still the occasional boss that smacks me down the first time I try it–that’s because the boss’s are novel compared to what I’ve encountered elsewhere in the game and I have to learn how to apply what I’ve learned in a completely new situation. But I usually get frustrated and quit games that just steadily get harder and harder and never give you a break and a chance to say “hey, I know this, watch: BOOM!!!” It’s WAY too tedious for me.

    I’m a bit like that in real life, too. I don’t want to be the fastest, most accurate typist on the planet, and I’m not. I learned how to type well enough for my needs and that’s it. If my computer had some sort of inbuilt system that kept pushing me to go faster and faster even though I don’t really have any use for greater speed, I’d get really frustrated and angry and probably quit using the computer altogether.

    I think that the way you figure out what you really want to do with your life is by deciding or finding the thing that you want to be *good* at, not just “as good as I need to be”.

  44. inflatablefish says:

    If we were incorporating a failure mechanic into this system I’d be very tempted to merge it back into the rest of the game, rather than have this learning system be so isolated from what the party are actually doing.

    So rolling a 1 might mean that an inopportune gust of wind has blown away several pages of your notes into the night. Or maybe the next bit of plot happens while you’re in the middle of your alchemy research, and you don’t have time to take your potion off the boil. Or maybe you’ve just spilled your only sample of zombie virus, possibly on yourself. Or maybe you just used what you thought was a perfectly safe divination technique on the Mysterious McGuffin and it suddenly started to tick…

    Essentially, just a way to involve the non-research characters in this minigame a little.

  45. Felblood says:

    In a couple of weeks, I’m starting a campaign, that picks up 300 years after the previous one ended.

    A major feature of this campaign will be a quest to find out what happened to one of the old PCs, and finding new and interesting ways to leave clues is something I’m scrambling to do.

    In short, I am totally stealing this idea.

    I may require skill checks at a modest DC, in order to roll against the card for the day, so as to integrate this into the existing skill system, and make reasonable limitations on who can participate.

    This also opens up the possibility of an indefinite lack of progress, if you keep botching your checks. –Particularly if they get harder, every time you complete a stage.

    The fragments of a degenerating madman’s journal might get progressively harder to translate as you go along, as the later entries will be written more hastily and be … well .. crazier.

    Think Al Hazred, Doctor Jekyll, and anyone who writes something down in a Cthulhu mythos story.

  46. theonlymegumegu says:

    Hmm, interesting. Reminds me a bit of the Extended Test mechanic in Shadowrun 4E, which I thought was a great addition in that edition. Extended tests are listed with a number and a time period (ex. Extended Test Mechanic (4) (30 minutes)), where the number is the total number of successes needed and the time period is how much time passes every time you make a check. Your method is interesting because after the first check, you’ll either succeed or make progress. Though, the Extended Test does allow for success on the first pass. Cool stuff.

  47. Decius says:

    What is up with so many requests to make failure have a ‘cost’? The cost of failure is the same as the cost of success, but without the benefits of success. Be that cost time, expensive components, or the only chance you’ve got- If there’s no price for success, failure is meaningless.

  48. Tizzy says:

    Interesting! How did you generate your table of values, though? I only tried the case of 1 success required, and I came up with


    which evaluates to a slightly different value: 6.29358.

    On another note: How did it work in practice? I would assume that if you require many successes, they would tend to come in as clusters as the end, so maybe for the zombie cure it wouldn’t be ideal (given that partial successes can have practical consequences in-game.)

    But then again, humans are notoriously bad at understanding probabilities, so who knows?

  49. Vulpis says:

    Hmm. Nice way to adapt the ‘learn skills during downtime’ mechanic that’s been in GURPS and the like for years to D20–the alphabet dice mechanic is a nice twist, and makes it more interesting than ‘You spend 8 hours in the library, and earn 3/4 of a point towards your Gnomish Trivia skill’.

  50. Seansie says:

    Very Cool… I have a player who wants to research how to become a lich in my group… this may be the best way for me to track the progression of his research (which should take him a few levels worth of game time). Thanks for the awesome post!

  51. Schnabatdern says:

    Hey, this is perfect! I’ve been running a campaign with a wizard trying to create an arcane elemental, and I’ve been wondering how to pace out his research beyond “You read some books, try some rituals, blah blah blah, but you’ve still got a ways to go.”

    This idea definitely seems like it would work well. Thank you! :D

  52. John says:

    Old D&D had a system for learning weapon mastery that took the learner’s and teacher’s skill level into account and took time to complete. Basically there are five levels of mastery. If you are taught by someone much higher than you you have a good chance to increase. If you don’t increase in a period of training then you still increase your chance by 10%.

    Training time and Costs
    Level sought Time req(wks) Cost/Week
    Basic 1 100
    Skilled 2 250
    Expert 4 500
    Master 8 750
    Grand Master 12 1000

    Chance of training success
    Student’s Trainer’s knowledge
    knowledge Basic Skilled Expert Master G/Master
    None 60% 80% 95% 99% 99%
    Basic 1% 50% 70% 90% 95%
    Skilled – 1% 40% 60% 80%
    Expert – – 1% 30% 50%
    Master – – – 1% 20%
    Grand Master – – – – 1%

    1. The student finds a teacher who will teach him.
    2. The students pays the costs and starts training.
    3. Half way through the student rolls for success.
    Pass – go up to the next level.
    Fail –
    a – stop training.
    The student now has half the time to spend doing other stuff. The student may get half their money back. They can study with a new teacher with the same chance of success.
    b – keep training.
    The student spends all their money and time. Next time they try they have to find a new teacher but their chance of success rises by 10%.

    This system makes it harder to rise a you get to higher levels. The student also has to keep looking for new teachers (sources of knowledge) each time they fail.

    You can modify the costs and time it takes. If something is easy to do then they may only need to get to Master or Expert level to finish it. If they have some knowledge of the area then you can start them off at Basic or Skilled level instead of none.

    The best idea is that they need to keep finding new places to learn each time they fail. So the person deciphering the book may start in the local library. When he fails (has read all its books on the subject) he needs to go somewhere else (a sage, a private collection, an old elf who was there, the library in the big city). You can incorporate that into you adventures!

    I never thought about using this another way until I read your post above. Your way is very clean and simple and I like it a lot. Your way also lets you decide in advance how many pieces you want to give out. I present this for your interest as a system with a similar intent.


  53. Keenath says:

    Clever idea. I’m a little split on the implementation.

    Technically, this is identical to “Roll a straight d20 against DC 21, and each failure permanently lowers the DC by 1”. If you use the suggestion of marking off X entries to start with (where X is equal to your stat bonus), it’s identical to “make an intelligence check DC 21…”

    Not a complaint at all, but it might be an easier way to state the same concept while maintaining the D20 system concept “higher is better”. As written, a 19 isn’t a good roll and a 1 isn’t a bad roll… it only matters what you’ve rolled before.

    But on the other hand, the marking-off-numbers may be more viscerally satisfying to a player, so there’s something to be said for that.

  54. A few thoughts:

    (1) The cost of failure is equal to the cost of the attempt. If nothing else, each attempt will probably cost time.

    (2) The fact that success is guaranteed is more troublesome. However, if you want it to a problem which might theoretically prove completely intractable for the character then you simply need to set a maximum number of attempts. If you haven’t achieved success after X attempts, then you’ve exhausted your insight into the problem. (Having multiple people working on the same problem is useful because it both (a) speeds up the resolution and (b) gives greater insight (as represented by more potential checks) into the problem.)

    (3) One way, therefore, to get the character’s skill to factor into the problem would be to limit their number of attempts based on their skill modifier. In D&D, I’d probably recommend something like 10 + skill modifier.

    (4) The other way to factor skill into it would be to set a DC for the task. Each d20 roll becomes an actual skill check and, if you succeed on the check, then you get to cross off an additional number of your choice.

    (5) Note that, if you don’t want to keep track of a note card, you can just keep track of the number of attempts you’ve made: If you roll a number lower than the number of attempts you’ve made, then you score a success.

  55. bearfoot says:

    You mean a skill challange in 4.0?

    Our DM just did a cool bit where the baddie was invunerable so we had to do skill challanges in combat.. failure she summoned more minions to deal with us. It was a quite memorable fight because of this.

    The other nice part was that the skill challange HAD to end with a thievery check, as we had to steal her holy symbol from her.

    The other successes were to find out HOW to steal the symbol. Good times.

  56. Will says:

    This is a neat idea with the card, but it’s mathematically equivalent of rolling 1d20 vs DC 21, and decreasing the DC by 1 on each failure.

    I mention this because some people might be more comfortable with a more traditional DC-based check, and it’s easier to integrate character stats into the system. For example, in 4e, I’d make it a skill check, with the initial DC equal to 25 + 1/2 the “level” of the task, and I’d say that the DC drops by 1 on a failure only if you are trained in the skill. Or something like that.

    I do really like the idea of parceling out the successes a little bit at a time like that, though.

  57. davwalp says:

    GREAT idea Shamus.

    Here’s how I am going to incorporate it into my campaign.

    Alchemy does not currently exist in my campaign world, however, the characters have recently discovered some coded notebooks that have the formula for alchemical items.

    To make life easier, I am considering that the Crafting DC of an item is the same as the research DC.

    For example, consider a wizard trying to discover the formula for Alchemical Fire.

    Alchemical Fire Craft DC = 15. So, the base DC for research is also 15.

    Our wizard has a +3 bonus to INT and 4 base skill ranks in decipher script. No ranks yet in Craft Alchemy because the skill doesn’t actually yet exist!

    I use a Time Unit system when there are major breaks in activity, such as characters are in town for awhile. Basically, I determine how much time they will have before we move on to the next adventure.

    The TU concept is a DM control because I determine when these will be given out based on pacing of the campaign. And it forces characters to make choices. For example, time spent in diligent research on alchemical items means that our faithful wizard is not therefore research new spells, crafting items, etc…

    1 TU = 1 week.

    d20 rolls = number of rolls per TU equal to decipher script ranks

    # of successes = Craft DC – INT mod

    So therefore:

    Research DC 12 = 15 (base) – 3 (int)
    Can roll 4d20 per TU (4 base ranks in decipher script)

    12 successes requires an average of 27 rolls
    28/4 rolls = about 7 TU required

    Also, once he discovers his first formula, he will get 1 rank in Craft Alchemy. Additional crafting ranks must be purchased normally.

    Once a formula is discovered, need to perform normal Crafting checks to create the item. So even though our wizard has uncovered the ancient secrets of alchemical fire, he still has limited success at first in actually creating it in the lab.

    Additionally, 5 base ranks in Craft Alchemy will give a synergy bonus of +2 to the research TN.

    I feel like that sounds more complicated that it actually is, but I like being able to incorporate both ability mods and relevant skill ranks.

    Of course, it’s all relative. This based on a very campaign specific research question.

  58. SiliconScout says:

    What kind of formula did you use to figure out the table there Shamus?

    I am looking to incorporate this system in a different gaming system and would like to play with some numbers as I refine the system.

  59. […] Young has posted his proposal for an interesting Learning Mechanic at Twenty Sided. Here’s how he describes the goal of the […]

One Trackback

  1. By The Alexandrian » Blog Archive » Untested: Research Checks on Thursday Mar 31, 2011 at 12:17 pm

    […] Young has posted his proposal for an interesting Learning Mechanic at Twenty Sided. Here’s how he describes the goal of the […]

Leave a Reply

Comments are moderated and may not be posted immediately. Required fields are marked *


Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun.

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>