We’ve been playtesting the Aurora RPG Engine during our regular Sunday game and it’s been going very well! (Much smoother than I figured it would, actually, which has been a most pleasant surprise.) As we play on, I’ve noticed a couple of really neat things.

## Postscript 2 – Further Probabilities

First, because Margin of Success is determined by the number of “remaining” dice, there’s an easy way to figure out what the probabilities are for achieving a certain MoS for a certain number of dice. This is because asking “What’s the chance of generating an MoS of 1 on 5 dice?” is the same as asking “What’s the chance of generating success on 4 dice, so that I have one die left over?” To which we already have the answer from the previously generated probabilities chart: roughly 50% (though again it’s really 45%).

This also makes it nice an easy to generate a matrix to quickly reference the probabilities for each MoS:

Or, as a chart showing the chance of getting at least MoS X:

## Postscript 3 – Hidden Rolls & Discoveries

Second, the above leads to another very cool and interesting thing: the system is *bidirectional*.

Typically, the difficulty of a task is handled by adjusting the base dice pool. This allows the player to viscerally feel their chances and properly size up the situation, thus letting them make appropriate choices (and fully RP it out). However, there are certain instances where it may be more appropriate to *not* “give away” the difficulty of a task to the player and have them roll blind. In certain campaign and genre styles – such as one where the characters are expected to be over their heads or one where it is a grim and failure-heavy milieu – this may apply to most of the tests/rolls. For others, this may be saved for more uncommon circumstances where the level of tension is heightened by explicitly being uncertain and not in the know.

The sweet thing here is that, as evidenced above, the probability of success when removing dice is the same as requiring an MoS equal to that number of removed dice. Therefore, if a player needs to succeed on a test where the difficulty is pegged at minus 2d, the test can instead easily be run ‘blind/hidden’ by having the player roll and checking if they get an MoS of at least 2. If not, then the test is failed.

This can also be useful for areas like perception, discerning realities, investigations, and similar, where the GM may not want to tip their hand that something is there (by specifying there is a penalty) and where the overall margin of success can be used to determine the amount or exactness of the information gained or discovered. For example, the players are searching a room and there is a particular item that is difficult to find (the GM has determined it’s well hidden indeed at a -3d difficulty). They roll; on an MoS of 0 they find a few mundane items, on an MoS of 1 they find some important documents, and on the MoS of 3 they find the secret compartment containing the important item. To flip it around, if the characters already knew the object was in the room but just not where, and they were actively searching for it, the GM could let them test normally with the up-front 3d penalty.

(As an aside, Star Trek Adventures does a version of this for many types of searching, sensor, and etc tests, giving a basic amount of information on a success and allowing the player to spend Momentum (their version of MoS) to give additional and more exacting details and information.)

With this bidirectionality, the engine gains even more flexibility and adaptability, all in service of running the game in a way that supports the campaign genre, tone, and style.