Percentiles and Player-Defined Skills?

[Sir Thomas]

I was wondering if anyone here can give any guidance:

I love percentiles, and I'm thinking of a system using Unknown Armies' "flip-flops" as narrativist tools, where the player has to narrate how they have improved so much.

However, I'm not at all interested in skill lists, and would like a HeroQuest-style system with definable skills, without being freeform. But I can't work out how you could implement it well with percentiles.

Does anyone have any ideas how you could do this?

[Lxndr]

I know nothing about heroquest other than the fact that it uses a twenty-sided die.  So take this with a grain of salt.

But really, if you want something hero-quest-like, but using percentiles, then just take the heroquest stuff and multiply it by five.  It works out to the exact same scale.

[Ron Edwards]

Hi Sir Thomas,

One of the key points about flip-flopping in Unknown Armies is that they are really just plain old "add 20% of your current skill" bonuses in disguise. So what you're looking for, I think, is simply a bonus based on the player's ability to justify or jazz up a particular stated action.

It could apply to any dice system, as such, so let's not get over-focused on d100 vs. d20 vs. anything else.

What really matters, it seems to me, is what sort of input you're looking to reward the player with. What might be some examples of player statements that would and would not get the bonus, as you see it?

Here's your other point, regarding defining the skills. Correct me if I'm wrong, but doesn't Unknown Armies (I have first edition) already permit the player to define the character's skills from scratch, with no official skill list to work from?

Again, I don't see how percentile vs. other flat-line dice systems makes any difference to the question. If you want the players to come up with their own skills rather than pick them from a list, then that's what you do. The only concern is how good the characters' skills are permitted to be for a starting character.

Best,
Ron

[mrteapot]

Hi Sir Thomas,

One of the key points about flip-flopping in Unknown Armies is that they are really just plain old "add 20% of your current skill" bonuses in disguise. So what you're looking for, I think, is simply a bonus based on the player's ability to justify or jazz up a particular stated action.

There's a little more to it than that, in that the bonus you receive is a function of your skill level, giving the most advantage around 50% in the skill and dropping off as you approach either end of the spectrum.  It can also be used to change the degree of success or failure in some situations.

I see how this is mostly the same as a flat bonus, in terms of how the system affects the narrative, except that it encourages using the character's passions to flip-flop their primary skills (usually in the 50% range) more than lesser skills in the lower ranges.

As for the original question, I don't know anything about how HeroQuest works, so I'm not sure what you mean about open skill lists without defined skills.  Therefore, I can't really help much.

[AnyaTheBlue]

Here's your other point, regarding defining the skills. Correct me if I'm wrong, but doesn't Unknown Armies (I have first edition) already permit the player to define the character's skills from scratch, with no official skill list to work from?

You're absolutely right, Ron.

Although, on the other hand, there are the 8 "default skills" which are sort of defined, and if you add up the various "special" skills sprinked about in NPC writeups in the various supplements, I suppose you could make an argument for a sort of meta-fixed skill system underlying at least some of UA.

I wouldn't do that, though.  It's free-form enough that the skill list is effectively unfixed, IMHO.

(By the way, the 2nd edition UA rules are worth the purchase -- the background is easily two or three times as deep, and there are some minor tweaks to the core rules as I recall)

[Daniel Solis]

Can't really say I've figured out how to best implement Do-It-Yourself skills either. It seems the simplest way to encourage the "right" kind of DIY skills from the players of your game is to link skills to a categorization scheme that helps define the skill's sphere of relevance.

For example, UA uses Mind, Body, Soul and Speed stats to govern all the DIY skills. Even an ambiguously named skill like "Call in the cavalry" placed under a certain stat can give the GM a quick idea of what it does. Say it's under Body, it probably means that the character gets a sudden burst of combat effectiveness if he's close to death. If under Soul, it could indicate an ally or allies ready to come to his rescue.

Another thing that helps a lot is to provide example skills. This especially true if you're allowing the DIY skills to have their own individual mechanical quirks beyond simple task resolution.

For example, some of the sample skills in UA2 affect initiative while others provide bonuses to other rolls if a roll of that skill is successful. I asked Greg Stolze about the rhyme or reason of these mini-powers' mechanical effects and he said he just used what felt right to him at the moment.

That's a perfectly valid way to go about it, but you may want to have a predetermined set of mechanical effects that "special" skills can and cannot do. From there, you can provide sample DIY skills that use mechanical funkiness in the way you intend for your setting. You could even explicitly state, right there in the game text, the mechanical effects prohibited from manipulation by special skills.

EDIT: One last bit... If you're very turned off by the idea of skill lists, then just describe the example DIY skills as part of character writeups or in the examples of character creation. This way, the skills have some sort of context in relation to the type of character might possess them.

[M. J. Young]

When I saw this two days ago, I decided not to go into Multiverser because someone would say it has skill lists--and rightly so, in one sense, because the bias system requires that everything fit somewhere. However, I think there's something to draw from it here.

In Multiverser, we talk about baseline for skills all the time. A particular kind of skill requires so much time, reaches a specific distance, has a defined effect. Users define skills, sometimes on the fly in play; they don't look up skills in the book and do that, but rather describe what they want to do and what they expect to have happen, and the dice are rolled to determine whether they are successful. In making that determination, the skill as described is compared to the baseline version of that skill. If the skill is at the baseline--it has the same factors in every category as described in the book--then the target number is determined in the ordinary way based on character numbers.

However, it's unlikely that a player will define a skill exactly as stated in the book; he doesn't usually have the book. Thus the factors he's described are compared to the baseline values, and the relative power of the skill is determined from that. So for example, if the skill is supposed to do damage to an opponent, we see how much damage it's supposed to do as compared with the book value; if it has a range, we compare that to the book; if there's a time factor or speed factor, we compare that to the book.

Any time a factor for a skill is doubled or halved in the player's favor (made faster or more powerful or easier, for example) there's a -10 on the chance of success; any time a factor is doubled or halved to the player's detriment (takes longer, uses more energy, does less, works in fewer situations) there's +10 on the chance of success. All the modifiers are totaled, and you wind up with a number to attach to the skill which in essence lets the skill do whatever the player wanted it to do at whatever cost he wanted to pay, with a probability of success adjusted to increase or decrease his chance of doing it proportionately to how powerful it is if it works.

You could do the same thing in a simpler context (given that you don't also have to cover bias) by creating baselines for abilities generally, and adjustments to probability of success based on degree of variance from the baseline. So you could say that all attack skills do X damage; but if they do 2X damage the chance of successfully using them goes down.

We used doubling and halving because it enables great differences in ability to be generated within a percentile system. Rather than using the baseline as a yardstick to be strictly measured linearly (3X, 4X, 5X providing linearly increasing penalties) we found that letting it be used geometrically allowed a lot more ability within the game (4X, 8X, 16X would be the levels at which the penalties accrued). We also found 10 point penalties useful in good part because it was easy to say that 6X was two and a half doublings, thus 25 point penalty (converting quickly to percent of the next value).

That might help. We provide a lot of detail in the book on what kinds of changes constitute what increase or decrease in difficulty, if you're interested.

--M. J. Young

[Ron Edwards]

Hi there,

Mr. Teapot, you have mis-read my post. I did not write "add 20% to your current skill." I wrote, "add 20% of your current skill," which is an accurate rendition of the rule.

Please read more carefully; this forum relies on people paying very close attention to what others are saying.

Best,
Ron

[Lxndr]

If you flip flop your roll, your skill of X% actually has a y% chance of succeeding, as per the table below:

10%  18%
20%  35%
30%  50%
40%  63%
50%  74%
60%  83%
70%  90%
80%  95%
90%  98%

20% of, say, 10, is 2.  so if flip-flops were "add 20% of your skill", then it should be a 12% chance, not an 18% chance, right?  Or am I now misinterpreting what you said?  Are you taking 20% of some other number?

[Mike Holmes]

Code Select Expand

          Flip         %
Skill Norm Flop Delt increase
   1   1%   2%   1%  100.0%
   2   2%   4%   2%  100.0%
   3   3%   6%   3%  100.0%
   4   4%   8%   4%  100.0%
   5   5%   10%  5%  100.0%
   6   6%   12%  6%  100.0%
   7   7%   14%  7%  100.0%
   8   8%   16%  8%  100.0%
   9   9%   18%  9%  100.0%
  10  10%   18%  8%   80.0%
  11  11%   19%  8%   72.7%
  12  12%   21%  9%   75.0%
  13  13%   23% 10%   76.9%
  14  14%   25% 11%   78.6%
  15  15%   27% 12%   80.0%
  16  16%   29% 13%   81.3%
  17  17%   31% 14%   82.4%
  18  18%   33% 15%   83.3%
  19  19%   35% 16%   84.2%
  20  20%   35% 15%   75.0%
  21  21%   35% 14%   66.7%
  22  22%   36% 14%   63.6%
  23  23%   38% 15%   65.2%
  24  24%   40% 16%   66.7%
  25  25%   42% 17%   68.0%
  26  26%   44% 18%   69.2%
  27  27%   46% 19%   70.4%
  28  28%   48% 20%   71.4%
  29  29%   50% 21%   72.4%
  30  30%   50% 20%   66.7%
  31  31%   50% 19%   61.3%
  32  32%   50% 18%   56.3%
  33  33%   51% 18%   54.5%
  34  34%   53% 19%   55.9%
  35  35%   55% 20%   57.1%
  36  36%   57% 21%   58.3%
  37  37%   59% 22%   59.5%
  38  38%   61% 23%   60.5%
  39  39%   63% 24%   61.5%
  40  40%   63% 23%   57.5%
  41  41%   63% 22%   53.7%
  42  42%   63% 21%   50.0%
  43  43%   63% 20%   46.5%
  44  44%   64% 20%   45.5%
  45  45%   66% 21%   46.7%
  46  46%   68% 22%   47.8%
  47  47%   70% 23%   48.9%
  48  48%   72% 24%   50.0%
  49  49%   74% 25%   51.0%
  50  50%   74% 24%   48.0%
  51  51%   74% 23%   45.1%
  52  52%   74% 22%   42.3%
  53  53%   74% 21%   39.6%
  54  54%   74% 20%   37.0%
  55  55%   75% 20%   36.4%
  56  56%   77% 21%   37.5%
  57  57%   79% 22%   38.6%
  58  58%   81% 23%   39.7%
  59  59%   83% 24%   40.7%
  60  60%   83% 23%   38.3%
  61  61%   83% 22%   36.1%
  62  62%   83% 21%   33.9%
  63  63%   83% 20%   31.7%
  64  64%   83% 19%   29.7%
  65  65%   83% 18%   27.7%
  66  66%   84% 18%   27.3%
  67  67%   86% 19%   28.4%
  68  68%   88% 20%   29.4%
  69  69%   90% 21%   30.4%
  70  70%   90% 20%   28.6%
  71  71%   90% 19%   26.8%
  72  72%   90% 18%   25.0%
  73  73%   90% 17%   23.3%
  74  74%   90% 16%   21.6%
  75  75%   90% 15%   20.0%
  76  76%   90% 14%   18.4%
  77  77%   91% 14%   18.2%
  78  78%   93% 15%   19.2%
  79  79%   95% 16%   20.3%
  80  80%   95% 15%   18.8%
  81  81%   95% 14%   17.3%
  82  82%   95% 13%   15.9%
  83  83%   95% 12%   14.5%
  84  84%   95% 11%   13.1%
  85  85%   95% 10%   11.8%
  86  86%   95%  9%   10.5%
  87  87%   95%  8%    9.2%
  88  88%   96%  8%    9.1%
  89  89%   98%  9%   10.1%
  90  90%   98%  8%    8.9%
  91  91%   98%  7%    7.7%
  92  92%   98%  6%    6.5%
  93  93%   98%  5%    5.4%
  94  94%   98%  4%    4.3%
  95  95%   98%  3%    3.2%
  96  96%   98%  2%    2.1%
  97  97%   98%  1%    1.0%
  98  98%   98%  0%    0.0%
  99  99%   99%  0%    0.0%
 100 100%  100%  0%    0.0%

Mike

[mrteapot]

Mr. Teapot, you have mis-read my post. I did not write "add 20% to your current skill." I wrote, "add 20% of your current skill," which is an accurate rendition of the rule.

Please read more carefully; this forum relies on people paying very close attention to what others are saying.
/quote]

I did read your entry correctly, and, as Mike's chart indicates, your reading of the original rule was off.  (A way of simpilfying the rule that is closer to, though not the same as, the original would be to say the flip-fliopping effectively allows another roll when you fail.)

However, I was attempting to say that the specifics of the mechanic, while different from what you had stated, were not terribly significant, and should not be a large concern for the original poster.

[Ron Edwards]

Sigh ...

... my point remains uncommunicated, but it really doesn't need to be.

Back to the original discussion.

Best,
Ron

[SumDood]

I use a flat skill list percentile system mixed with single digit attributes for damage modifiers. I believe this is what you are trying to avoid.

One example you might want to look at is Rob Lang's Icarian model. It's system uses percentile rolls, but also has a circular chart on the character sheet for determing the percentiles. I haven't spent a lot of time with it yet, but it might be something similar to what you're looking for.

Icar is at http://www.icar.co.uk

Hope this is some help.