[Captain Kwok]

This is more applicable to leaky armor systems and not solid (hit first) armor.

When an armor or component is the majority hitpoint contributor and has a structure lower than the incoming damage, there is an increase in the number of re-rolls. The additional rolls lead to an increase in the chance that internal components will suffer at least partial damage.

In the event the majority HP contributor has a structure equal to or greater than the incoming damage, there is a lower chance of re-rolls and the damage is distributed closer to HP of component/total HPs ratio.

SJ designed two ships. Ship A had 6000 total HPs with about 90% from Light Armor. Ship B had 4000 total HPs with about 85% from Heavy Armor. His observations had Ship B winning nearly 70% of the time. The reason? Ship A tended to lose it weapons before Ship B, despite the fact that the ratio of HPs from weapons was much lower than ship B. You would expect the weapons to survive longer on Ship A right? But it doesn't happen.

The reason for this observation is that the light armor is only 10HP and incoming damage is around 100HP. Each time an armor/internal is destroyed, you get a new re-roll. The more re-rolls leads to a greater chance that the internals will take at least partial damage. You might get an average of about 8-10 re-rolls with light armor in this case.

On the other hand, heavy armor reduces the chance of re-rolls. If it is 50kT - you might only see a few re-rolls (2-3) at most.

In a spreadsheet I made I came up with the following data:

Ship:
Light Armor 40x10HP = 400 HP ~ 0.76 of total HP
Internals 10x10HP = 100 HP ~ 0.20 of total HP
Weapons 2x10HP = 20HP ~ 0.04 of total HP
Weapon Damage = 30 Damage

The chance that ONLY ARMOR components would be destroyed in a single weapon hit was 0.45, the chance that at least one internal would be destroyed was 0.48, and at least one weapon was 0.11. In this case there was no partial damage since all HP structure was 10HP for simplicity.

If we were using 30HP heavy armor (= to weapon damage) we should observe damage distributed by the ratio of HPs/total Hps. That'd be a 0.76 chance for only armor to be destroyed, 0.20 chance for an internal, and 0.04 chance for a weapon. Real observations would probably show a bit less for armor, and a tad more for the other two in the event that armor was not hit first (hence a re-roll was needed).

Overall we see about a 2x increase in the chance that internals will be hit in this scenario with light armor. The in-game effect would not be so dramatic since I'd expect armor actually represents about half of HPs on a normal ship design - so the factor would be closer to 1.5x.