The previous Design Diary introduced you to our playtest process and design goals. Consequently, there wasn't much room to talk about the details of the system. This time I want to talk about a personal expertise of mine: dice mechanics! While there are plenty of innovations scattered throughout the system and setting of Asylum, we're not trying to reinvent or complicate what already works.
Lots of games out there use dice pools or combinations of different dice to determine outcomes but as a statistician and system designer I've noticed those sorts of mechanics come with more than their share of problems. Quite often these game mechanics have complicated probabilities that neither the players nor the designers fully understand and anticipate. If you remember from our design goals, we want the Asylum rules to be clear and reliable. That's why most actions in Asylum are handled by rolling a single d10 and comparing it to a number. As a result, we know that every +1 modifier is +10% probability for success, regardless of how skilled the character or difficult the action is. However, that d10 roll doesn't just determine if the characters succeed or fail. It also determines the degree of success, called Impact. This "one-roll" mechanic is the core of our entire resolution system. It's proven quite intuitive for most of our playtesters because it's very simple and higher numbers on the dice are always better.
In our first few playtests, Impact was generated through a separate roll much like how many combat systems will have you roll separately to hit and then damage. To speed up gameplay, players were encouraged to roll two d10's at the same time and read them left to right but this had a habit of creating unclear or confusing situations. Many players reverted to the traditional method of rolling to see if they succeeded first and then rolling for Impact. What we noticed was that players would sometimes roll a 9 or a 10 when rolling to succeed, and then generally less for their Impact. Of course, there was nothing special about that first high roll but, accross the spectrum of age and prior RPG experience, our playtesters subconsciously expected that first high roll to mean a better result. By adapting the system to rely on only one roll, we've managed to play into that expectation, while also narrowing the variance in player participation.
Another dice mechanic that underwent a change is something we call the Specialty die or Expertise. The basic concept was that when a character was attempting an action they were highly specialized in or had magical assistance for, they got a second die to roll and could use either result. This mechanic was actually a lot fun for our playtesters, but we've had to be very careful about where and when we employ it. As I pointed out at the beginning of this article, rolling multiple dice can significantly alter the probabilities of each result. Consequently, determining the difficulty of tasks becomes far less reliable when Expertise comes into play, as the effective bonus from this mechanic actually changes from character to character and situation to situation. After we combined Impact and success into one roll, Expertise was just too powerful in most situations. We could have just stripped it out of the game, but we didn't want to completely remove what our playtesters were enjoying. Instead, we're only introducing Expertise for powers that have immediate costs and consequences, and as an incentive controlled by the Narrator to encourage players to take narrative risks.