- I took the three herds (16, 32, 64) and to each of the 4 "test characters" I added the specific advantage (so FAST as in Fast Company is -4 DM vs. Impact, -8 DM vs. Pen). I ran the simulator
- I removed all the self-vs-self battles. I wish I'd done that from the start--I insisted they be there and it was probably a mistake--if the test character is overpowered he still gets a tilt towards 50%-POV (Percent-chance Of Victory) by fighting himself and his test-character peers.
- I got a POV-increase from the standard test character. So if I added +2 AGI I got a bump of like +18% POV for my four test-guys (and these guys are 50% weapon, 50% defense).
- I compare that increase to another chart I'd already made: the test characters with +1 Armor (+1, +2, +3, +4, etc.). We have tested and discovered armor to be pretty consistently worth 1 AP per point of armor. So if +5 Armor gave a 15% increase in POV and some TAP-power gave a 15% increase then we could say, by deduction that the TAP power was worth 15 AP.
- Problem: The problem was that a LOT of these powers fell between the armor values ... in some cases substantially. I made a bunch of judgment calls I'll have to revisit later about which way to go since we don't generally sell fractional AP. I then used that "eye-balled" value to determine the % of total AP. I need to go back and revisit this and we'll see if I can get a "more scientific" measure.
- I then rounded up for all the AP values to try to smooth out the curve and try to erase any serious mistakes that my testing, data quality, etc. might have introduced (i.e. rounding up will mean that if I've underestimated then I should be able to compensate somewhat).
So testing will continue. What this means, however, is that certain powers (like Stretching) which give -Damage Mods, certain kinds of armor, and so on, will have TWO (and in some cases 3) cost components. These are:
FIXED COST: If a power like stretching includes "flow through keyhole" that, in JAGS, has a fixed cost--it doesn't change no matter what the rest of your character is like.
TAP: Total-AP component. Abilities like the above (-DM's, extra AGI, and so on) are based on your Total AP. So if Stretching includes -4 DM vs. all physical attacks then it'll cost based on your Total AP and be different for different point-level characters.
Damage Level Based: Some powers (like extra REA) cost based on what your character's Damage Level is relative to AP (that's Very Low if 1/8th or less of your AP are spent on attacks, Low if 1/4th or less of your total AP are spent on attacks, or "normal" if more than 1/4th of your total AP are spent on attacks).
Some powers like Fast Company which include standard abilities, negative Damage Mods, and extra attacks could include all three. :-O.
Our plan to deal with these monsters is to have arranged costs as a table where we've done the math with a side-bar that explains how to work it out if you're either off the chart or doing something unusual (like adding "Fast Level 2 to a pre-existing character").