Die Mechanic Flavors

Yesterday on the bus I started mentally comparing Dark Heresy's roll-under d100 die mechanic to that of Zludge's 2dF. In particular, I was curious how much such a conversion would affect the percentage likelihood of success in different situations. But I came to realize that this is too low-level of a comparison. Instead, it's more important to consider the assumptions behind a mechanic and the resulting "flavor" that it imparts.

Here's what I mean in terms of 4 different die mechanics I've been working with lately:

d20: linear skill variation. Here, the die roll just provides a random bonus to the skill. The average d20 roll is about 10 (10.5, actually), so against a DC of 10+skill, there's about a 50% chance of success or failure. If using critical success (natural 20) or critical failure (natural 1) rules, there's a 5% chance (for each one) each roll, regardless of the character's skill level. Because the roll distribution is linear, you are just as likely to roll at the extremes of the range (+10 or -10) as you are at the character's actual skill (+0).

Zludge: curved skill variation. As for d20, this just provides a bit of randomness centered on the character's skill level. However, a curved roll distribution is weighted towards the skill level: there's a 33% chance of rolling +0 on 2dF, but only an 11% chance of rolling +2. If +0 is sufficient to succeed, this gives a 66% success rate.

I feel this curve limits some of the "gambling" flavor of d20, especially since the range of possible roll values is so small. It's almost like playing diceless--it's unlikely you'll get very "lucky" with a roll. But this goes the other way too: it's more unlikely you'll fail miserably just due to a bad roll. Instead, you have to intelligently play based on your character's skill level.

Dark Heresy: linear roll-under. Here, you roll d100 and roll under your skill value. The degree of success is determined by just how far under you roll. Thus, as your skill level increases, so does both the likelihood of success and the possible degree of success.

In Dark Heresy, the average skill starts at about 30, so this gives only a 30% success rate (though this can be modified based on the circumstances; it seems DH's default curve center-point is for a pretty challenging task). Since the distribution is linear, each skill improvement gives the same return: +5 skill increases your chance of success by 5%, regardless of whether you purchase the increase at a low or high skill level.

Assuming few characters ever achieve the max skill level of 100, there is always a chance of failure on a roll. It feels that your fate is controlled much more by chance with this mechanic: your skill only sways the likelihood of success, but there are no guarantees here. Every roll is a gamble. (I actually like QAGS's "the Price is Right" spin on the linear roll-under mechanic a little better, but it produces the same results.)

GURPS: curved roll-under. As a roll-under system, this too seems to have a bit more of a gambling feels. For example, you're just as likely to roll an 18 on 3d6 when you have a skill of 8 as you do with a skill of 16. However, since the distribution is curved, at least most rolls will be centered around 10.5. The chance of rolling an 18 is only 1/216. Skill increases at higher levels provide increasingly limited returns.


After considering these differences, I think I am correct in going with the Zludge die mechanic, since that is the flavor I want in my games. As applied to Dark Heresy, this may reduce the "grittiness" a bit. However, I do still plan to try QAGS a bit, to see if my expectations for a linear roll-under system really do bear true.