Characteristic x N rolls are an abomination (?)

[Stephen L]

17 hours ago, ffilz said:

My point is that multipliers and additive/subtractive modifiers do different things to the probability curve

So, if I summarise what I’ve learned from all the discussion here.

Tests that are a more difficult application of a characteristic or skill should, *in general*, be a penalty subtracted from the skill or characteristic x5, where they are simply testing the deeper levels of the skill or a stronger level of the characteristic (i.e. separating the men from the boys).

However for tests that are more difficult because they are introducing a level of luck (it’s easier to think of skills examples; firing an arrow in a fluky wind, or hitting someone in the dark, but perhaps lifting a slippery weight) then a multiplier modifier (e.g. characteristic x3 or skill/2) is appropriate.  These kinds of tests are more levellers, poorer skills have less penalty because they rely more on luck in the first place.

Edited January 20, 2021 by Stephen L

[davecake]

On 1/19/2021 at 8:35 PM, NickMiddleton said:

I have long maintained that a flaw in "modern" (post circa-1990) game design is this weird obsession with fitting all resolution on to a single schema or mechanism at any cost...

With HW/HeroQuest/QuestWorlds as the epitome of that approach in one way (same rules for no matter what it is about), even though it recognises the crucial point that not all resolutions in the game should have the same level of detail. Which kind of works for the 'Story and narrative is everything' stance of that game, but is hard to do well for RQ, which has a simulationist core that would not sit well with two different resolution systems for the same thing that can give quite varying probabilities. Of course to some extent it is play style (some people hate the idea of simplified mechanics for 'mooks' as against RQ principles, some people love it). 

RQ has a crazy large number, as a game with a very long history, and RQG seems to use almost all of them at various points. Some of them seem redundant. 

I personally like systems that have a few different core mechanics for different purposes - Gumshoes investigative vs general abilities, for example, is a really nice distinction for those games. But it's not really clear if there is anything consistent reasoning to why we use particular systems for particular things in RQ, as so many of the mechanics date from the era when everyone was sort of making it up as they went along without many examples to guide them. We should be lucky we got as many things right as they did. 

And some changes in resolution just seem to be GMin style, while some change not just the game but implicitly the world (switching from POW vs POW to skill vs skill for Spirit Combat, for example). 

I use attribute x N rolls fairly sparingly, mostly as a way of doing something that is too general to be a skill roll  but otherwise seems to fit that mechanic. One reflection I agree with the argument that a higher difficulty roll should probably be a penalty rather than a reduced multiplies, but that for something that requires both talent and significant luck a reduced multiplier is a quick and easy option - and honestly, this kind of roll is often for relatively unimportant things, like giving characters a chance to work out something their player missed, or if a player wants to do something that sounds not part of a particular skill, and also difficult and the GM does not want to say an outright no, but also does not want to just do it. 

 

 

[davecake]

On 1/19/2021 at 9:08 PM, Akhôrahil said:

And Pendragon's "roll under, but highest" is just awesome, of course.

When HQ moved to definitely using 'roll under, but highest' one of my players just hated it, and has never shut up about how wrong it is. Of course the two systems aren't quite the same in other ways, though very similar. 

On 1/20/2021 at 3:09 AM, ffilz said:

People can be wrong about what a given probability actually is, but people can't be wrong about their preference for how a particular way of determining what the probability of success is.

Well, they can be wrong about how a given mechanic translates into probability. Sometimes very wrong. Even professional game designers*, even famous ones, can be very wrong, as I found during the Hero Wars development process - maths is really useful for game design but yet not a core skill. More game designers need to have at least one person with a good grasp of probability works in their playtest team, and to listen to them (there are some dreadful examples out there, such as early editions of Vampire and Shadowrun). Though RuneQuest did make it into the old Murphys Rules cartoon in the Space Gamer for the quite large percentage of people in any given battle who cut their own head off. 

*this is not sub-tweeting Robin. 

[PhilHibbs]

8 hours ago, davecake said:

When HQ moved to definitely using 'roll under, but highest' one of my players just hated it, and has never shut up about how wrong it is.

Yeah, "So I'm trying to roll low, but not too low, unless I can get super-low, how am I supposed to do that?"

Answer: "Try just getting a random number, and not using psychic powers or cheating, and having a higher skill so you can get those better successes."

[Akhôrahil]

2 hours ago, PhilHibbs said:

Yeah, "So I'm trying to roll low, but not too low, unless I can get super-low, how am I supposed to do that?"

This rule works absolutely gorgeously in Pendragon. RQ isn’t very good at opposed rolls, since Success vs. Success is a very common outcome that doesn’t really resolve. Success delta is better. Roll under but highest is sublime.

[French Desperate WindChild]

49 minutes ago, Akhôrahil said:

Roll under but highest is sublime.

it is clearly counter intuitive that rule is a very simple rule to determine the best result

a more intuitive would be take your skill, soustract your dice when succeed, then compare with the opponent skill minus its own dice, if it is greater you win.

with the "roll under but highest", you just have to compare roll vs skill, then roll vs opponent roll, without any calculation, you're right, it is definitly sublime

[Akhôrahil]

36 minutes ago, French Desperate WindChild said:

a more intuitive would be take your skill, soustract your dice when succeed, then compare with the opponent skill minus its own dice, if it is greater you win.

Local RPG legend Anders Blixt, who will be lead designer on the Swedish RQ edition, recently made a BRP game (using a D20 instead of D100, which seems strongly recommended when calculating a success delta) that’s based on this design pattern throughout. It’s a really smooth and fast BRP design, and feels very modern throughout. Definitely recommended as an example of what you can do with the basic idea.

https://www.drivethrurpg.com/m/product/303046

Edited February 3, 2021 by Akhôrahil

[PhilHibbs]

2 hours ago, French Desperate WindChild said:

it is clearly counter intuitive that rule is a very simple rule to determine the best result

a more intuitive would be take your skill, soustract your dice when succeed, then compare with the opponent skill minus its own dice, if it is greater you win.

Yes, it's avoiding subtraction that is the advantage. I'm not a fan of the over-100% rule for this reason.

Edited February 3, 2021 by PhilHibbs

[Beoferret]

With opposed rolls, when both parties succeed at something where only one can be the winner, the easiest thing may be to declare that the one with the widest margin of success (between dice roll and target number)has come out on top. That's how GURPS does it. Simple. RAW's position that equal levels of success (i.e., regular vs. special vs. critical) leave a contest unresolved and/or tied, are still fine (to my mind) if that would lead to some fun drama (Argrath and Harrek both scramble for a sword on the floor, but they each roll a special success; they both end up with their hands on the hilt and now they have to wrassle for it!)

As for the problem of characters with particularly low stats still having a chance at success, when they shouldn't have any at all, couldn't that be solved by assigning a minimum attribute score to the check? I don't know why the rules officially state that everyone has a 1-5 on a D100 chance of success, even if their attribute score is lower than 5.

In terms of using Attribute x 5 with +/- modifiers for attribute tests, does anyone have a rough estimate of what they might be for various degrees of difficulty? 

[Stephen L]

9 hours ago, Beoferret said:

In terms of using Attribute x 5 with +/- modifiers for attribute tests, does anyone have a rough estimate of what they might be for various degrees of difficulty? 

Statx5 is simply turning a stat (measured on a scale of 1-20), to a percentage (1-100).  So, it very intuitive: use the same modifiers for skills.

• -25% for a difficult task (turns an expert, 75%, or Stat 15, to average, 50%/ Stat 10)
• -50% for a masterfully difficult task (Turns a master, 100%/Stat 20 to average, 50%/ Stat 10)
• -75% for a heroically difficulty task (turns a hero, 125%/ Stat 25 to average, 50%/ Stat 10)

I tend not to use any finer modifiers, as I don’t think slight graduations of difficulty adds to the drama of the roll.

[PhilHibbs]

12 hours ago, Beoferret said:

With opposed rolls, when both parties succeed at something where only one can be the winner, the easiest thing may be to declare that the one with the widest margin of success (between dice roll and target number)has come out on top.

How is that easier than "highest roll wins"? They are statistically identical, and comparing two dice rolls directly is objectively easier than doing two subtractions and comparing the answers.

Edited February 4, 2021 by PhilHibbs