Recently, Green Ronin began publishing design diaries about their upcoming Dragon Age tabletop RPG. The latest design diary explains the core dice system for the game. In essence, it’s 3d6 plus a modifier of -1 to +7 (attributes range from -1 to +5, and skills can add an additional +2) vs. a difficulty. So far, it’s not too different from GURPS: it should have a strongly center-weighted distribution where each +1 bonus is exactly equivalent to -1 difficulty. The meaning of a +5 vs. +0 should be exactly 5 steps of difficulty (i.e., the +5 will hit DC 15 exactly as often as the +0 hits DC 10).
The interesting change is the introduction of the Dragon Die: one of the d6s, designated before rolling, indicates the success amount for successful rolls. This is fairly counter to the Margin of Success (MoS) systems that are common, and makes the die system into a strange hybrid of GURPS and In Nomine. And, as with In Nomine, players already worry that the Dragon Die concept will produce strange results at the table: because the Dragon Die also adds to the chance of success, successful rolls at higher DC/lower skill will succeed grandly more often than successful rolls at lower DC/higher skill.
The math behind this is quite simple: once DCs hit the point that the player must roll well on all three dice, it’s going to be very hard to get a success without getting a good roll on the Dragon Die. Meanwhile, a character that succeeds most of the time has a far wider spread of possible results on the Dragon Die that still result in success. Against DC 10, a character with a -1 modifier will expect an average of 4.3 on the Dragon Die for successes, while a character with a +7 modifier can only expect a 3.5. Meanwhile, the MoS for the -1 is 1.9, while it’s 7.5 for the +7. This obviously seems counterintuitive: The higher bonus is going to beat the DC by an average of 5 more points than the lower one, but actually get one point less success based on the Dragon Die.
If this is an issue, there are a couple of ways to deal with it without completely reinventing the wheel or adding lots more math: direct MoS and Governed MoS.
In direct MoS, you ignore the Dragon Die completely and just take the Margin of Success (Result – DC). This is how most games seem to do it, but it might cause problems with the expected results; with the Dragon Die concept, the rest of the game engine will probably be based around only successful results of 1-6, and a straight MoS result could get as much as a 15, which may drastically skew the results of high-bonus play.
In Governed MoS, you take the MoS as long as it’s not higher than the Dragon Die. This system removes the advantage of the Dragon Die for low bonuses: because your Dragon Die result can’t be higher than your MoS, the greater bonus retains the advantage. Because your MoS can’t be higher than the Dragon Die, it preserves the maximum of 6 for results (though it does add a 0 that might need to be dealt with). The real flaw in the system is that it’s pretty punishing for lower bonuses: in order to even out the results, it lowers the average for most of the range by 1-2 points off the base system, and, at the higher end of the bonus scale, is exactly the same as the Dragon Die system. It’s essentially just penalizing the lower bonuses with no benefit for the higher ones.
And, overall, the high Dragon Die on successes for high DCs might just be an acceptable wart if the rest of the system manages to be good. Because, taken in with the actual average of total results (where clear misses count as a 0 on the die), the average success for a straight Dragon Die system still follows a curve that favors having a higher bonus.