Armor in StarCluster affects Chance. It always has, and it always will. I get flack every so often about this, and it has reared its ugly head again. Michael Scott and Randolph Allen are writing a StarCluster based Vampire game called Society of Night using the percentile variant, and in a phone conversation with Scotty yesterday, I found out he had been applying the percentage penalty which was supposed to be applied to Chance of Success (to hit in combat) to Quality of Success (damage in combat) - removing a percentage of the damage inflicted. When I informed him of the actual rules, he chuckled and said something about doing it the D&D way.
That is guaranteed to set me off on the Song and Dance! I never - ever - do anything just because that's the way something else did it, especially D&D. In fact, given my personal utterly burned out disinterest for anything D&D after running it for 20 years without a break, this decision was reached despite it being used in D&D.
I also did this - using armor this way - knowing it was intuitively wrong. Intuition says armor makes you take less damage. Thing is, the division into chance - AKA to hit in combat - and quality - AKA damage in combat - is not really there. it's just two aspects of a single thing. This division is a gaming distinction, a statistical abstraction. If your enemy's armor absorbs damage, you will be doing less damage over time. If you miss more often, you will be doing less damage over time. It's just a choice of presentation. A preference.
The design decision to put that diminution of damage into the chance roll is so the modifier can be pre-loaded, resulting in less calculation and handling, and quicker combat resolution. It is the choice I made, it works, and I'm happy with it. It irritates some people because it is intuitively wrong, but that doesn't mean it actually is wrong.
StarCluster is a heavily abstracted system. This is just one more example of that abstraction. When you are in a firefight in StarCluster, you are not just firing one or two bullets per minute. In every firefight, the vast majority of shots do not hit. In actuality, you are pumping off a hell of a lot of bullets which go flying around, not hitting anything important. In the system, I abstracted out the majority which aren't going to hit, and focused only on those few which have a decent chance of hitting. Rather than tracking every maneuver and establishing a precise modifier for it, I dumped all the maneuvering into an abstraction of trading points. This is a design decision to enhance other aspects of play.
I have made this point what seems like a million times since I released StarCluster 1E eight years ago. Some people are fine with it, some people don't like it but accept it, and some people refuse to accept it. After 8 years I just shrug and move on. You can't please all the people.