Thinking about an advancement system

Saelorn said:

So the adventure "starts", and then everyone spends a month in their basement to max out all of their skills.

Click to expand...

That image is based on flawed (or mis-represented) versions of the concept.

Specifically - in such a scenario, the player does not get to claim they may make die rolls sequestered in a basement. The GM says when a skill check is called for, and should only call for a skill check when failure would have a narrative consequence in the scenario. Sitting in your safe, isolated basement far from anything like meaningful action does not put you in a situation where failure will change the narrative, so you can't earn experience that way.



On the general method:

I, personally, prefer to not have to consider my character advancement during the action of play. I find it either distracts from the action, or gets forgotten in the midst of the action. A variant that allows me to deal with the advancement *after* the action scenes would be preferable.

Umbran said:

That image is based on flawed (or mis-represented) versions of the concept.

Specifically - in such a scenario, the player does not get to claim they may make die rolls sequestered in a basement. The GM says when a skill check is called for, and should only call for a skill check when failure would have a narrative consequence in the scenario. Sitting in your safe, isolated basement far from anything like meaningful action does not put you in a situation where failure will change the narrative, so you can't earn experience that way.

Click to expand...

That wasn't how the rule was presented, in the opening post. There was no caveat about "only the GM decides when you roll" or "only when failure would have a consequence". While adding such caveats can address the issue, I've found that it turns the game into one of manipulating the GM rather than manipulating the rules, which is not a good thing.

If you can only roll when the GM tells you to, then the method of play becomes convincing the GM that you should be allowed to roll. If you can only advance when there would be a consequence for failure, then your job is to convince the GM that the consequences would be meaningful. The rules put the GM in the position of playing the bad guy, who resists your attempts at having fun.

Saelorn said:

That wasn't how the rule was presented, in the opening post.

Click to expand...

Dude. Give some slack for it being an *idea*, not a polished work for publication, hm?

There was no caveat about "only the GM decides when you roll" or "only when failure would have a consequence".

Click to expand...

There needed to be? Doesn't common sense apply at this stage?

I mean, in your game, do the players say, "Okay, before we begin, I spend a month hunting down (checks calculations) 432 rats, which pose no physical threat to me. I slay them, and go up a level,"? If not, suggesting that would be allowed here seems... a bit like you think the author is kind of stupid.

If you can only roll when the GM tells you to, then the method of play becomes convincing the GM that you should be allowed to roll.

Click to expand...

Wow. Who do you play with? Do your sessions look like this:

Player: "I attack (*rolls dice*) and kill the orc!"
GM: What orc? I didn't say there was an orc here.
Player: Well, I don't have to wait for you to tell me to roll dice, so I just did it myself!

The player may be able to *suggest* a thing, but in the end, the GM is the one who arbitrates use of the mechanic - including when the mechanic is used at all. You cannot attack an orc that isn't there. And you don't use skill checks when the GM doesn't say they are necessary to resolve the scene.

Name RPGs in which this is *not* the case, please, to support your contention otherwise.

Umbran said:

There needed to be? Doesn't common sense apply at this stage?

Click to expand...

What you consider to be common sense, may be a radical idea to someone else. I am surely feeling the effects of that, on these boards, with some people seeming to go out of their way to resist obvious concepts. I try to give them the benefit of the doubt, that their experience is radically different from my own, but it's still exhausting.

Umbran said:

I mean, in your game, do the players say, "Okay, before we begin, I spend a month hunting down (checks calculations) 432 rats, which pose no physical threat to me. I slay them, and go up a level,"? If not, suggesting that would be allowed here seems... a bit like you think the author is kind of stupid.

Click to expand...

My game doesn't award XP for slaying creatures which pose no threat. You would have to play out the process of hunting down those rats and finding them, because there's a non-zero chance that they might kill you, and an even greater chance that you'll stumble across something worse than a rat.

I'm not saying that the author is stupid. I'm saying that this is a universal problem with any system that allows advancement through skill use or practice. Both GURPS and Traveller suffer significantly from this. It's not a trivial problem to solve.

I recall issues when I used to play similar Chaosium systems like CoC and Stormbringer. There was a lot of bookkeeping to update incremental changes multiple times during a game session. Not to mention wear and tear on paper sheets and erasers.

I would suggest something structured around the actual game session. During play, just keep a list of any failed skill used during the game. You only need to fail once so multiple failures don't count. Then at the end of the session, players can do their upkeep to improve characters to simulate their growth and experience.

At this point you can impose whatever rules and limitations for your particular system. For example, limit players to improve up to 3 skills from their list of failed checks per session. That prevents abuse by those who would otherwise roll and fail constantly just to rack up skill advancements.

I would also suggest a die roll or fixed amount for increase rather than 1 at a time. The die or number would be based on the character's current skill level. The less skilled have more to learn, thus a potentially larger gain. As he approaches mastery, there is little left or new to learn.

I have never found the edge cases to be relevant, and I have played a lot of CoC; it's tracking san loss before death. Even in playing Mythras/M-Space, I haven't found that to be an issue with a random few percent here and there, it's just something a little extra to take the sting out of losing.

Morrus said:

Inspired by CoC and similar systems. I haven’t thought it all the way through yet. It’s very similar to the way CoC does it.

Percentile system. Everything is a skill. You improve with failure.

Every time you fail a skill check, the skill improves by 1%. Technically they can go over 100% (modifiers might bring it down).

Hmmm. Definitely easy to abuse though. Find lots of easy tasks and repeat. Though I guess you’d stop failing them, so your score would remain static.

Click to expand...

I've always liked the CoC system in that it "made sense" that what the player would get better in would be what they practiced doing, and it had built in balance that the better you were the harder it was to advance.

Mouse Guard does something similar where to advance you must accumulate a certain number of successes and failures.

And that probably makes even more sense.

But over the years I've gotten away from always caring about what made sense. Right now I mostly care whether the math works. And for an advancement system, that's mostly questioning whether characters advance at predictable and useful rates so that after X sessions of gaming you have some sense of what sort of characters you'd have to work with and can plan and prepare accordingly. That tends to mean a system that can't be abused and isn't overly random, which CoC system for all it's elegance can be. It's also worth noting that CoC's system in particular puts high emotional pressure on players to cheat, in that you can complete a session and get a lot or get nothing, so that players who are tempted by that sort of behavior are prone to do so.

Likewise, from a game play perspective, I really only care whether the rewards are spaced far enough apart that they feel both earned and useful (in that your advance matters at least once before you advance again), and yet spaced close enough together that players don't get depressed waiting for them.

How a system achieves that and whether it is realistic is not something I put at a high priority any more.

A related question about advancement systems - how fast for skills you're not so good at?

For example, say we have a character that is advanced via whatever method the system uses. Level in a D&D-type game, etc.

They want to start using a new skill. Or just do - the hacker finds herself in more and more social forums trying to drum up support for their cause and while she wasn't designed with social skills she's now using them.

Do you design advancement to be rapid at first, so they can interact meaningfully against challenges at their advanced level? (Though not up to characters who have been primarily focusing it) At the same speed as characters who have been doing it a while so they will always lag the same distance behind and can never catch up? Even slower because skills take a lot of time to get the basics that is represented in the background?

I think how advancement is going to handle non front-edge skills will determine at lot about your outputs mechanically as much as how it handles your front-edge. By front-edge I mean skills/etc focusing on what that character primarily does. If a rogue concept is great at sleight of hand and picking locks and disarming traps, they are front-edge at that - in front of the other PCs. And probably not front-edge at ancient theology. And how you want to handle both of those cases will really impact what you want your mechanics to do.

I am generally in favor of a "the more you do things at the edge of your ability, the more you learn" system. That said, I disagree on whether failure is essential. If I try a skateboard trick many times, the failures might get me closer to the moment when I can pull off that trick, but the first time I actually DO pull off the trick, is IMO the turning point in whether I can reliably do it again.

There are also skills for which this model is inappropriate, such as knowledge or lore skills. If, in the Cave of Caerbannog, I try to name the capital of Assyria, and I can't because I never learned that particular fact... and then later on, someone else asks me to name the capital of Assyria... my failure the first time *didn't teach me the capital*. If, between those sessions, I read a book on Assyrian history, so that "next time I'll be ready", THEN I should advance my score (or modifier) in Lore: Ancient Politics. But that should be a "did I train during downtime" mechanic, not an "I developed my skills by using them under pressure" mechanic.

By the way, the capital of Assyria changed a few times; at one point it was Ninevah. Just in case anyone ever asks you.

Umbran said:

There needed to be? Doesn't common sense apply at this stage?

Click to expand...

Nope. This is the Internet.

Saelorn said:

If you can only roll when the GM tells you to, then the method of play becomes convincing the GM that you should be allowed to roll. If you can only advance when there would be a consequence for failure, then your job is to convince the GM that the consequences would be meaningful. The rules put the GM in the position of playing the bad guy, who resists your attempts at having fun.

Click to expand...

There is a good point here, although I'm about as happy with the packaging as Umbran is. Advancement by (conflict resolution) failure means there's no advancement if the GM doesn't ask for conflict rolls. In such a situation, yes, the player has an incentive to try and roll. Which is bass ackwards to me, since it drives me nuts when players roll before they role.

Morrus said:

Percentile system. Everything is a skill. You improve with failure.

Every time you fail a skill check, the skill improves by 1%. Technically they can go over 100% (modifiers might bring it down).

Hmmm. Definitely easy to abuse though. Find lots of easy tasks and repeat. Though I guess you’d stop failing them, so your score would remain static.

Click to expand...

Better tie intelligence into this. The skill improvement percentage depends on how well the character's grey matter works. And if skills can reach superhuman levels, you're going to need level caps too.