Math: Formulas

There are several different formulas that rule the rolling of the dice, between the probability that dictates value chances, to finding out reasonable averages to expect during gameplay.

For starters, there are a few terms that need to be defined:

Event: A certain condition that is, or is to be, made true.
Probability: The odds of an event occuring.

For starters, we can begin at how the dice system works, here. Power Advantage runs off of a system of 'x'd'y'+'z', or . For example:

Players start with a 1d4 Kick:
x = 1
y = 4
z = 0

Most attacks that deal damage will follow this formula, which can be converted, into the AIM chats, as:

//roll-dicex-sidesy

And then you add z to the final roll.

Now, with these rolls, there must obviously be maximum possible rolls, and averages of the max and minimum values.

To figure out the max value, one could use the formula of xy+z, or . In this case, you multiply the x and y values together, then add z to the result, yielding the maximum possible value.
Let's give this a test:

A player has a 10d3+5 Flaming Rapid Cut, and wishes to find the maximum possible damage.
So, if the player wanted the maximum damage, he would need to roll perfectly on every d3, thus getting ten values of three. Using the formula, it can be much easier than counting these all up.
xy+z, so (10)(3) + (5) = 35, max.

Now, it would be great to always roll these perfect values, but, alas, it is rarely so. So rare, in fact, that it usually isn't worth hoping for, when you need it. What is more important, however, is the fact that every roll has an average value it can be, which can usually be attained, more often than not.
The formula for this is somewhat more complicated. It follows the standard ((Max)+(Min))/2 formula, but it is suited to fit the values we use, here.
The formulas is ((xy+x)/2)+z, or . For example, in the 10d3+5, we could replace the values with:

(((10)(3)+(10))/2)+5    =    ((30+10)/2)+5    =    (40/2)+5    =    20+5    =    25.

So, if you wanted to set a goal on how much damage you needed to do with your Flaming Rapid Cut, 25 would be a very fair guess.

The last formula this article will cover will be one of my own devising, based around the Accuracy Dice that Power Advantage uses.
Brandon King came up with a very unique way to roll accuracy, which would later become the model for most online-RPG accuracy systems. Brandon King allowed Power Advantage to use this system, and although we made some minor alterations to values, it is essentially identical to the original system.

The accuracy dice, as you may see in the Handbook, start at 2dy values, then raise to 3dy values, and continue to grow, generally keeping y as a multiple of 10. When rolled during a battle, if any roll is equal to half of the max roll (Ex. 2d50, max roll is a 50, so if any are 25 the condition has been met), the attack hits.
What is odd about this system, is that although the numbers are growing larger, the chances of hitting are actually decreasing, until you reach the next x value. 2d40, in fact, has a higher chance of connecting than a 2d100.

Now, it will be claimed that this is because "there are more numbers that can miss, at higher values". This, although true, is entirely irrelevant. Yes, at 2d40, 1-19 is a miss, so there are 19 'miss' numbers and 21 'hit' numbers, whereas 2d100 has 1-49 missing, so 49 'miss' numbers to 51 'hit' numbers. The number of values that make the condition true rise in the same manner as those that miss, so that cannot be the case.

One can puzzle this out, using a pencil and paper, but it works much better at smaller values. We will begin with the ever-popular 2d6, which is what board-games, such as Monopoly(tm) use.
If this 2d6 was being rolled, it is common knowledge that there are 36 possible outcomes: the first die yielding 1 to 6, while the second yields a 1 to 6 on every possible value of the first.
Now, if either of these dice is 3, or higher, it has matched 50% of the max, and would be a hit. Therefore, the only missing combinations would be those only containing ones and twos. In fact, the only combinations of these are 1,1   1,2   2,1   2,2.
So, out of 36 possible events, 4 of them would be misses, leaving us with 32/36, or 88.8888...% accuracy.

The formula for this is somewhat strange, as odd numbers, having no direct half-value that is a whole number, will be different than even numbers.
The formula for accuracy is as follows:
(((y-v)/2)/y)^x, or

Where x and y follow the previously discussed xdy, and v is a value of my own creation.
When y is odd, v = 1
When y is even, v = 2

This formula also follows mathematical rules, and will only work when:
y > 0 AND
x > 0

Now, let's take a test run with this little formula, using the 2d6 we previously used, as to allow us to compare the truth of it.

In 2d6:
x = 2
y = 6
y is even, so v = 2

(((6-2)/2)/6)^2    =    ((4/2)/6)^2    =    (2/6)^2    =    (4/36)
And since 4 values miss, 36-4 = 32 values hit, showing us that this formula works.
This can also be tried with 3-sided dice, such as 3d4, but that will not be discussed in detail, here.

Now, to use this for an accuracy roll: We shall begin with 2d40.
(((40-2)/2)/40)^2    =    ((38/2)/40)^2    =    (19/40)^2    =    (361/1600)
So, out of the 1600 possible events, there will be 361 misses (Only 19 and lower), and 1239 hits. 1239/1600 = 77.4375%
When done with 2d100, we see everything work, as before, only the end values are (2401/10000). The final percentage for accuracy with this is 75.99%
This rights itself, however, in 3d60, where the final values are (24389/216000)(Note: You would not want to do this one by hand...), with a final percentage of 88.709...% accuracy.

Now, why would we use a system where the accuracy gets worse as you grow?
Well, when someone uses Kaioken, or any move that makes -20 to your accuracy, you certainly wouldn't want a 2d40, or 2d30. However, even if the penalty to accuracy is -50, and you have a 2d40, if you roll a 40, it hits, irregardless of accuracy modifiers.

There are many, many formulas that help to define Power Advantage, and several key ones have been named here. I am ending the article, at this point, and hope to have another up, soon.


-Lander Zander
(12/08/02, December the 8th, 2002)