You are working under the presumption that you are going to get a DS off before you pick up the adds. Granted I am working under the inverse presumption, but still. There is less chance of having a blood shield up during that time, than there is of not having it up. If you have been hanging around, or if you have travel time to grab all the adds.
Like I mentioned though, it's not a huge thing really, I am simply pointing out obvious situations where mastery is reduced in value, as most people are finding the places where it increases. ie Magical dmg contributing to larger heals. Using it intelligently etc..... Those factors benefit mastery only slightly over an encounters duration. My negatives also only effect it slightly, but they do effect it nonetheless.
I actually love mastery, I simply want the negative to be advocated so that when the final numbers are eventually calculated, all the possible pro's and cons have been voiced and concidered.

Just wanted to post this and see if anyone had similar results and/or logs that reinforce this...

Stoneskin Gargoyle
Damage Taken	5240192.16
Damage Absorbed	1508115.21
Healed	1521940.67

Swordshattering
Damage Taken	5061867.94
Damage Absorbed	1450282.76
Healed	1488766.13

The stats used in Bloodsim were my current stats (note - unbuffed) in game. For Stoneskin Gargoyle it was 1390 dodge (11.35%), 1400 parry (12.35), 1750 mastery (17.75), 170,000 health, 8500 attack power and 30300 armor. For Swordshattering Health changed to 168,000 and 29140 armor. The aboves are average values for 500 iterations of 8 minute fights.

Another thing to note is that when raid buffed my armor is roughly 34500 (with Stoneskin), which would further reduce the effectiveness of the additional 4% armor due to returns on armor stacking, causing the gap to get larger. Once I get into a raid group tomorrow night I'll record where my stats sit and rerun the simulations just to verify.

I've noticed a lot of tanks swapping runeforges in and out, does anyone have parses to compare Stoneskin vs Swordshattering to see how accurate these results are?

To progress further and use real world values (rather than static swaps as Lichloathe did above with his hit/exp example) I went ahead and reforged my gear from the stats mentioned above to be avoidance heavy and reran the simulation to review. My new stats were 1565 dodge (12.15), 1540 parry (13%), 1435 (16) mastery, 170,000 health, 8500 attack power, and 30300 armor (Stoneskin) and 17% parry, 168,00 health, and 29140 armor (Swordshattering).

Stoneskin Gargoyle
Damage Taken	5250955.83
Damage Absorbed	1345675.85
Healed	1521324.75

Swordshattering
Damage Taken	5080392.75
Damage Absorbed	1286315.64
Healed	1487150.18

Obviously reflecting that the itemization points for mastery to avoidance is quite poor (even at low hit/exp, as I have .67% hit and 12 expertise), and one should gear towards heavy mastery rather than avoidance (which was already assumed and further reinforced by Lichloathe's posts.

It would appear that the optimal runeforge is Swordshattering as it provides a static 4% avoidance that is not subject to DR and the best way to reforge is to maximize mastery. As far as optimally reforging after maximizing mastery, you can take 2 routes. You can opt into using melee hit and soft exp caps to further reinforce your mastery stacking (which, per Lichloathe's posts above advise that this is a 1% DR reduction, however he did not take into account the damage avoided from losing avoidance stats to maximize hit and expertise which would make this obviously suboptimal for survival purposes, but well worth while in the event threat becomes an issue) or to aim for maximizing avoidance, which would optimize your survivability on fights.

This is a minor point, but the armor increase in Gargoyle lowers the damage you take per hit. This will generally make your blood shields smaller as you will take less damage over the five second period before you chose to use DS. There is a lack of synergy there. so when trying to maximize shield size...might be good to trade out armor in favor of avoidance even if the two starts are otherwise equal--so your not working against the Bloodshield size?

Minor point, but could make a difference when picking trinkets, rings, enchants, etc, that focus on armor vs avoidance. You'd never want to OPT out of armor with no tradeoff...just saying that when choosing between one or the other, maybe we would prefer trinkets, rings, runeforge etc that focus on things other than armor (so long as we are getting stamina or avoidance or some other tanking stat).

Note: I'm assuming most people use bloodshield like I do--you maximize it's usage and time it after big bursts of damage. If you are just using it every time it's up my above point is moot--your avoidance in short spurts wrecks the size of your bloodshield. When I made my above comment I was thinking about large strikes from bosses that take off a third or more of your health--that sort of thing. I use my DS's immediately following taking several hits or one really large hit--armor reduces all hits...avoidance avoids them. So if I get hit then parry, I just would hold off using DS until I get hit a few times in a row (usually a short wait lol).

Wrygryn, the problem with that theory is that over time avoidance will also lower your overall damage taken, which will lower your theoretical blood shield average. I don't think that is a compelling reason unless armor reduction per point is significantly higher than avoidance per point, not to mention that the ideal situation is to minimize total damage taken. If you are reducing your constant, passive damage reduction in order to increase your active, occasional reduction number it strikes me as robbing from the left hand to give to the right.

Also relating to the post above, can we please stop claiming that armor has diminishing returns in relation to effectiveness? This was shown years ago to be false: the same amount of armor lowers damage taken equivalently at high values as at low values. The maths has already been referenced in this thread.

What do most DK tanks uses when forced to dps? Like on Council when its phase 3 and you are no longer the main tank and its pretty much a DPS race? Does everyone just swap to unholy stance apply diseases and spam HS?

If you've happened to look at your DPS breakdown on Recount, HS does not do nearly the amount that RS / DS do. While it is beneficial to use, it is not our top damaging attack. Go into Frost Presence, Diseases > RS > DS / HS (Rune dependent).

Most tanks shouldn't have to be relied on to DPS during this part anyways.

You can't use rune strike in frost presence. Between that, the 20% faster rune regen from improved blood presence, and extra AP from bladed armor, blood presence will give you more DPS than frost presence. Just make sure you don't pull threat (as long as the other tank has time to get vengeance going you shouldn't be able to catch up).

Given the new values, not necessarily set in stone, napkin math is heavily in support of avoidance. AvDam is going to go up quite a bit in a long fight with heavy physical damage, making avoidance even more favorable.

So just to clarify, with the latest DK changes on the horizon, Death Strike has again been lowered, from 7%/25% to 7%/15%, correct? I don't see how avoidance won't become superior to mastery after this round of changes.

That would be true Ganashal if you hit Deathstrike every time it was up resulting in people using deathstrike when they get 3 parry's in a row...but people pick and choose when to use it--they wait till they need it. So if you get 3 parry's in a row, that's not when you use death strike--you wait till you are down some health and have been hit. You cannot do that with armor, it just flat does less damage to you all the time. In identicle (albiet best cast for my theory) situations 3 hits in a row with 15% more armor means smaller shields--3 hits in a row with less armor means bigger shields. And if you stack avoidance, it has nice synergy with this because on the times where it kicks in, you take 0 damage, and you hold off on Deathstrike and work your threat until you need it. I also find DS is up more often when I need it because of the little avoidance streaks where I don't want to use it.

As I said, it's a minor point, but if we see rings with 1k armor on them, or trinkets that have clicky effects to increase armor by 10k for 20 seconds...we may find it hurts our shields enough that it's really inefficient compaired to instead just picking up a trinket that grants avoidance or master for the same time period. To be statistically relevant we'd need to be able to shift our total armor by a significant amount.

In short, the spikes in avoidance tanking actually maximize our shield size potential and the fact that DS can't be used non-stop (rune CD's). Armor lowers damage all the time, but it will contribute towards lower shield size as well...I'm just not sure if it's a statistically relevant amount =).

This is a question regarding the change

'# Death Strike now heals for 15% of the damage sustained, down from 25%'

Is it known for sure that this is an actual change and not just a change in the tooltip? Looking of logs for a few nights, it seems like a change to the ability; can anyone confirm this?