To make an Attribute roll you will roll a number of ten-sided dice equal to [your Attribute]. You are looking for pairs, triples, and so forth. You're looking for the largest set (e.g., 3 of the same beats 2 of the same), and within that the highest number.
The way you read the dice is this: if you have three of a kind, e.g., 3 8s, that's "3 8s" or "38." If you have two of a kind, e.g. 2 7s, that's "2 7s" or "27." If your best is a single die, e.g. a 4, that's "1 4" or "14." If you have no dice to roll, because your Attribute is 0, then your roll is "0 9s" or "9"—but in practice, if you are bothering to make a roll with an Attribute of 0, it means either that you're goofing off or planning to get extra dice from some mechanic such as Fairy Fortune.
Ten-sided dice have either "10" or "0" on their tenth side. Read this as 0, so if you've rolled 2 0s or 2 10s, that's "20."
Your goal will be to equal either the difficulty of a certain action or the target set by an opponent's dice roll.
Subscribe to:
Post Comments (Atom)
You know, I have to say, you've conjured the least intuitive die resolution mechanic I think I've ever seen. And I've seen a lot of resolution mechanics, so that's saying something. It's *so* unintuitive that I had to actually sit down and code up an exhaustive die roller to model it just to get some kind of idea what it looks like as die pools go from 1d to 10d. I had to feed *that* data into a 3d graphing system to even get a grasp of where things fell. It left me with just one question.
ReplyDeleteWhy? Why, in the name of Hades' black Hell, would you inflict such a thing on unsuspecting people? With a glaring discontinuity between 5d and 6d as the probabilities shift up a tier to the gradual extension of the long tail of the probability curve as pool sizes increase, it is in every sense a hideous, misshapen monstrosity. Why, Dear Hell, why?
Bravo. Bravissimo.
This comment has been removed by the author.
ReplyDeleteHow delightfully chaotic. It's like the number-matching "cherries" in Unknown Armies on crack.
ReplyDeleteSo would some mechanics (like the Fairy Fortune) add dice, and others would up your chances by, say, letting you shift x number of dice by a magnitude of x or something like that? There are so many roll-modding possibilities here it's starting to blow my mind.
Though even one shift would make a huge difference. A friend pointed out that even a +1 -1 shift for one dice for, say, a character's specialty, could be enough--since even one shift could result in another match and an effective +10 to the roll.
I actually did out the math for this system back when I heard how WotG worked. I only bothered calculating 1d-5d, as the math gets really irritating to do by hand after 5d. This is because, for 6d, you have to take into account the effects of two triples, two doubles, and three doubles. Unless my combinatorics is off:
ReplyDelete0d averages to 9.0
1d averages to 14.5
2d averages to 17.12
3d averages to 19.38
4d averages to 21.55
5d averages to 23.68
0-1 = 5.5
1-2 = 2.62
2-3 = 2.26
3-4 = 2.17
4-5 = 2.13
So, ignoring the 0d number as it is weird, it appears that each die gives about a +2.1 to the expected value of your roll.
The above takes into account the fact that, if you get only single numbers (no pairs), your expected value still increases with the number of dice. So, with four dice, chances of getting only solitary numbers is still 50.4%, but your expected value for the highest die is 7.8. When you roll only one die, it is 4.5. This is because only the highest matters. With ten dice, you are guaranteed at least a 19 (a 9 on the highest die if there are no pairs) due to the pigeonhole principle.
Alex,
ReplyDelete'Cause it's easier to read the dice this way than in a dice pool, and more fun than a percentile. I kind of treat it as percentile in the back end, though it's possible that I may go the extra mile this time and find a way to do modifiers proper. I suspect I just need a lightweight way to compress the individual dice. Maybe turn 0/0-1/0-2 into . . . not a wild card, since that forces extra computation, and definitely not anything additive, but an orthogonal bonus of some sort. ^_^
Jenna
I'm not quite sure what you meant by "compress the individual dice," but if you mean reducing the probability spread such that it's easier to roll doubles, hmm.
ReplyDeleteI wonder what it would do to the dice if you let players subtract one from any odd number. So, a roll of 5 6 7 8 could be taken as 26. This would be similar to having people roll d5s instead of d10s, but more comprehensible and retaining the possibility of rolling odd numbers. Though, that may be compressing the dice TOO much. It would certainly make each die add a lot more to the expected value of a roll, which would spread out the numbers without increasing the number of dice rolled.
If that's too much compression, perhaps you can only do it with 1 and 3, or maybe only once per roll. Dunno, just throwing suggestions out there ^^;
You're correct that what I'm looking for is a way to increase the chance an individual die will match, but I'm also trying to retain a certain visibility-of-outcome. I don't want computation to be a part of reading the dice.
ReplyDeleteI've thought some about mixing in d8s or d6s, but the leading contender right now is something that makes low dice "useful" in some fashion that is not directly related to the numeric resolution. If 0s and 1s have a separate purpose, then they can cannibalize part of the impact of rolling d10s instead of d8s. That's not the same *thing* as rolling d8s (unless those 0s and 1s carry forward directly into expected success on future rolls) but it's close, and it means that allowing somebody to do that becomes a kind of "modifier."
Really, the thing is, I *really* like what this system does well---so I'm just poking at how to fix what it sucks at. ^_^
The only thing with reducing the dice pool from d10s to d8s is that you no longer have the full range of numbers to roll. Not that this is really a bad thing per-se, just kind of weird. With d8s, you could never roll a natural 19 for instance.
ReplyDeleteHere are a few more random suggestions if you don't mind random suggestions by a complete stranger. If you do mind, then, err, oops, sorry ^^;
What if 0's and 1's could be used as +5s to another set of dice? This makes them into half of a wild die. So, a roll of 0 1 8 could yield 28, same as a roll of 4 8 8.
Or, maybe, for every 0 and 1 rolled, the player may roll an additional d10, adding the new die to the roll? Only the 0s and 1s that were rolled in the first roll generate extra dice. After that, they would just be 0s and 1s. These would be like anti-exploding dice, and hopefully increase the matching number without making it feel too much like arbitrary extra dice are being rolled.
Or, 0s and 1s can trigger some other sort of game mechanic. I'm sufficiently unfamiliar with the mechanics to suggest something specific, but you do already use what sounds like a lot of tokens. Rolling 0s and 1s could allow you to maniuplate them in some manner.
It looks weird on paper and was hard to wrap my head around at first in WotG but ... it was a lot of fun. Especially with the River.
ReplyDeleteMaybe if you had some way of "banking" dice (like WotG's river) or could have one or more virtual dice that were set to some specific number ("your glibxnytz is 3 so you may replace one of your rolled dice with a 3 if you so desire when your glibxnytz is relevant").
Or perhaps have some trait granting optional rerolls of one ore more dice.