---

Q u o t e:
Right now, ArP gives lower-than-expected results on test. I know this has been posted before, but I didn't find a recent post on it, and it's not listed as a "known bug/issue" in the appropriate thread, though the problem persists and multiple EJ testers have confirmed it independently.

On a mob with no Armor debuffs (no sunder/ff):
Armor Pen from Gear (Actual effect) gives 101.25% (NOT 125%) of live, 81% of PTR tooltip
Armor Pen from static abilities (battle stance, mace spec) gives 81% of live, 81% of PTR tooltip

All the credit goes to the multiple EJ testers who actually figured this out - I just want to be sure it's here on PTR forum for visibility as well, because I didn't see a recent post on the issue when I went searching.

Sources: http://elitistjerks.com/f31/t29453-combat_ratings_level_80_a/p16/#post1170842 and below.
Hellord's test data - http://www.hellord.com/wow/PTR_Arp_stack.zip

[MATH - Maths using Hellord's test data vs boss dummy, reprinted for those who don't want to figure out his excel sheet:

Here's what he saw hitting a dummy:
BT damage (to 0 armor): 1804
BT damage (observed @ 674 ArP rating vs 10645 boss dummy armor ): 1299

Here's what we calculate given his gear, based on tooltip:
Expected ArP for 674 ArP rating @ 12.31640 ArP=1%: 54.7238%
Expected final armor from tooltip ArP: 10645 * (1-.547238) = 4819.65 = 24.0356% reduction
Expected final damage from tooltip ArP: 1804*(1-.240356) = 1370.34 = 1370 dmg

Here's what we calculate assuming a global 19% penalty on ArP vs tooltip:
Expected ArP for 674 rating w/ 19% penalty: 54.7238% * 81% = 44.3263%
Expected final armor w/ 19% penalty: 10645 * (1-.443263) = 5926.4692 = 28.0093% reduction
Expected final damage w/ 19% ArP penalty: 1804*(1-.280093) = 1298.712 = 1299 dmg

Observed damage exactly matches calculated damage using assumption of 19% ArP penalty, and does not match expectation based on tooltip. I recognize this is a single test case, but there have been many others like it with same results. The math checks out in similar fashion for battle stance being only -8.1% armor instead of -10% armor, but the two values are additive.
Also note that BT does exact damage - there is no randomness, which is why it was used. Despite this being a single case, infinite repetitions of same ability in same circumstance will produce same noncrit value every time. This is not an isolated case.]

---