Armour and Spell Penetration is not so simple to calculate as it is going to be subject to a bit of fuzzy numbers in the background. As noted already there are invulnerabilities which the bosses may begin with and are unknown to us, and there are other uncertainties like what diminishing returns a raid boss has to higher armour/prot scores. If they have very high scores and are subject to diminishing returns then debuffs will accomplish very throughout the diminishing return curve.
It can also be a little tricky putting in flat % increases due to penetration buffs and I'll try to illustrate with the following example with imaginary numbers:
(1) No penetration or armour debuffs:
Base DPS: 1000
Bosses Mitigation = 5000 = 50%
Actual Output: 500 DPS
(2)1000 armour penetration
Base DPS: 1000
Bosses Mitigation = 4000 = 40%
Actual Output = 600 DPS
This is a 20% increase from (1).
(3)4000 raid-wide armour debuffs (is: 5x stacks of torments)
Base DPS: 1000
Bosses Mitigation = 1000 = 10%
Actual Output = 900 DPS
This is an 80% increase from (1)
(4)4000 raid-wide armour debuffs AND 1000 armour penetration
Base DPS: 1000
Bosses Mitigation = 0
Actual DPS = 1000
This is a 100% increase from (1)
However comparing (3) with (4) you gained 11% damage, while comparing (1) with (2) you gained 20% damage. In both instances you only added 1000 penetration but your relative output was significantly different. For this reason a strawman only test is not going to be an accurate reflection of the gains you might find in a raid.