I decided to test whether or not Retribution Aura did indeed have a +Hit component to it and the results were very interesting.

Some specifics about the testing:

This was on Live server
Cookie cutter 0/5/56 Ret Spec
Human
Tested on the lvl 70 Training Dummy in IF
All tests were auto attack only, no CS, DS or Judgments
No seals active
500 auto attack samples for each test

Test 1
Changed up my gear so I had 0 +Hit.
No buffs at all
No aura

500 attacks – 24 misses – 4.80% miss rate

Test 2
The only change in this test was to activate Ret Aura

500 attacks – 14 misses – 2.80% miss rate

Test 3
In addition to Ret Aura active, I changed my gear to get +34 Hit for 2.16% Hit Chance

500 attacks – 2 misses – 0.40% miss rate

This all led me to believe that Ret Aura, either base line or talented, did indeed add 2% Hit to those affected by it. So I added more +hit to get to +48 Hit, which is 3.04%, added to the theoretical 2% from Ret Aura should push misses off the table on this test dummy.

This proved not to be the case because 126 attacks into the test, I missed.

So, I thought maybe the +Hit % was actually 1% and not 2%, so I bumped by +Hit up to 65 for a 4.12% Hit.

Test 4
Ret Aura active and +65 Hit for 34.12% +Hit

500 attacks – 0 misses – 0% miss rate.

I am convinced that Ret Aura does indeed have a +1% Hit component. I wonder if it stacks with Draenai Heroic Aura? If this is true, and it does stack, would explain why a lot of us with low Hit are not seeing any misses in raids.

That helps explain my beta results! I had a Draeni premade as Ret with the aura up and had 6.5% miss. Adding 2% hit from both effects to get close 8.5% makes sense (I only did 1200 swings, so the missing 0.5 can be ignored).

Also I did a different test with Conc aura up and had 8% miss.

It doesn't explain my beta results though. No talents, no aura active and a miss reduction rate of ~3-4% on target dummies 3 levels above me. Curiously Unarmed autoattacks don't appear to be undergoing said miss rate reduction, but my control for that test particular is too shoddy to be very reliable.

You might want to try an Aura other than Ret to see if the results change, and consider blowing 50g in the name of science to try the test talentless. Annoying it appears that the EU character copy process is screwed; I had hoped to give this a more thorough going over at 80 with maxed weapon skills this evening but my plan has been foiled.

This is a pretty screwy bug, whatever is causing it.

Did your paladin have Precision (pre-beta)? Precision gave 3% hit to melee weapons, so maybe unarmed isn't a weapon?

The premades existed after Precision was removed, so it can't have any ghost hit.

500 swings isn't enough to be sure of the actual hit %.



Over 500 swings you'd see: (SD is Standard Deviation)

Swings	500	500
%miss	5%	2%
SD	4.87	3.13
+2 SD	34.7	16.3
Expected	25.0	10.0
-2 SD	15.3	3.7

95% of the time, you'll see anywhere from 15~35 misses in a 500 swing sample. (3%~7% miss)
95% of the time, you'll see anywhere from 4~16 misses in a 500 swing sample. (.8%~3.2% miss)


Note that the ranges come close to overlap at 500 swings, and one of your data points (the Ret aura trial) is right on the edge of both ranges.

EDIT: That said, the mechanics do seem to be acting strange, based on all the reports we've heard so far. What if we try 4% hit with no Ret aura?

So the newest version of Rawr (2.0.2) Is showing that Felspine is beating Torch again, is this correct?

It also shows SoB handily beating SoC, I think it's just not very accurate as of right now and I believe the author stated it will be 80 before it's correct once again.

I'll add my own data sample to what we've got so far. Decided to do a little hit trial on a training dummy myself after noticing my complete lack of misses in sunwell with 113 hit rating.

Test setup:
- live server
- level 70 bloodelf
- 1h sword (used over a 2h for the attack speed)
- 350 sword skill
- 0 hit rating
- 27 expertise (6.75% dodge/parry reduction, probably not a factor affecting the test)
- had 3/3 precision pre-patch, have respec'ed twice since patch
- attacked level 70 dummy from the front for 1000 swings
- only buffs were ret aura, vengeance stack, and the occasional shard of contempt AP proc
- no outside debuffs were applied to the training dummy at any point
- talents: Talent Calculator - World of Warcraft

Result:
Hits: 680
Crits: 260
Glancing: 45
Miss: 15

For a 1.5% miss chance over 1000 swings. A lot lower than the previously expected 5%, it certainly feels like something has changed but I'll leave the actual statistical analysis to others.


Edit:

Re-ran the above trial without ret aura as a comparison for the ret aura investigation:

Result:
Hits: 670
Crits: 254
Glancing: 56
Miss: 20

This is because DM: C is not modeled correctly in the current version of Rawr, and its proc effects are ignored.

Collected my own data set on Live. Human, no hit rating, no Ret Aura, no expertise (used axe), had precision pre-3.0, on L70 target dummy.


SoB + auto only:
732 swings
12 misses (1.7% effective miss)
43 dodge (5.9%)
38 parry (5.2%)


SoB, auto, CS, DS:
766 swings
13 misses (1.7% eff. miss)
44 dodge (5.7%)
40 parry (5.2%)


edit: Note that those are the same trial run, looking at different sets of attacks.

EDIT:
I'm fully talented, with ret aura active and 0 hit rating vs lvl 70 dummy and after 2k attacks I'm at 4% miss rate on auto attacks. Suggesting 1% ghost hit for myself. Pre patch I had 1 pt in precision. I think this settles it.
God I wish I had 3/3 precision pre patch.

even if im not hit capped?

I thought hit was the best stat in the game if your not capped?

hit capping is no longer a priority above strength.

Because between the new Divine Strength and Kings, Strength got pushed above Hit Rating. It's not so much that hit rating got worse, it's that Strength got a lot better. As well, Crusader Strike doesn't refresh Judgements anymore, so it is no longer vital that it hits.

From my experience the spell power component of the trinket is not proccing on judgments since they are now considered ranged attacks so that would leave exorcism and consecrate to stack the spell power buff. In my opinion the trinket is sub par with the slow stacking of +spell power. I would much rather use the [Battlemaster's Resolve] for the static AP and there's no chance of the buff falling off in a movement fight.

Vindication also procs the trinket, which makes it a bit more acceptable when combined with Exorcism/Consecrate.

you have no draenei online? with a draenei 133 hit is overcapped...

Did you test or are you taking word of mouth? For me I couldn't tell a big difference between glyphed command and martyr infact martyrs 100% damage means its a bit less bursty and more steady.

I mean i'll jump on the max dps bandwagon any day of the weekbut from my experience (sure its down to human error and all that jazz) SoC was behind..

Tested, for hours on the target dummy over the last week since the patch. Your statements are contradictory but yes, from my testing they're so close I'd need to be a lot more scientific about my testing if I wanted to tell you which did better. Without the need for SA and the really high recoil in blood now I just much prefer command. Next week I may try blood in Sunwell assuming there's no more "hotfixes" and compare from this week but I'm not hearing of too many ret's doing over 3k in there so I think it's a safe bet results will be close to the target dummy's.

We never really raided with a retribution paladin, and after latest patch i am testing retri in sunwell. Despite my gear being really mediocre the dps was decent seeing the classes and gear i have to compete with. I did 2800dps on first brutallus kill and 2500 yesterday. However as the perfectionist i am, i want moarrrrr and feel it should be possible.

About the DPS Cycle. I've seen some contradictive posts but in general they seem to agree that it should be: Seal, Judge, CS, DS, Cons (HW/Exorcism squeeshed in on demon/undead).

However i have a question about micromanaging the cooldowns/GCDs and maintaining the priorities. There will be points in the cycle where you'll end up with Crusader Strike 1sec CD left, Judgement 1.5sec CD left. So in theory you could use CS before judgement, but you'll lose 1sec on casting your judgement because of GCD.

In this case is it worth waiting for Judgement to get of cooldown and prio it over CS. Or really keep to FCFS and smash CS then wait 1.5 then judge?

And as a final: below 35% it should be Judge > HoW > CS right?

edit: seems my armory hasnt updated, great.

We need to get people to stop thinking this. MECHANICS HAVE CHANGED, READ THE LAST 10 PAGES.

I set my model based on how stuff worked on beta when I made the module. I can't release a new version every time blizzard introduces a bug. But yea, theorycrafting at 70 at this time is fairly pointless. We are strewn with bugs, will be having significant changes soon, and are BALANCED around 80.

Given the current state of ret paladins (and raiding in general) you don't need to theorycraft until WOTLK (though its fine to theorycraft about level 80 stuff now). Every raid instance is a joke and any monkey could /faceroll on a keyboard and kill anything now.

No, you're better off just hitting the CS and then judgement right after. Use the macro a few pages back.

Hehe yeh thought so.

And yeh i saw the macro, but im not really to fond of those macro's. I been holy for a long time, so dont wanna make retri into a one button spamspec now Though i'll get it just incase football is on.

One thing im not 100% sure about though is. The macro has divine storm before consecration, However some ppl say you should prio consecration over divine storm. I kinda neglected consecration yesterday.

Is it now viable, due to no mana concerns, to twist Seal fo Blood, Seal of Vengeance in an attempt to keep up 5/5 debuff? From a fellow ret pally in my guild, he said he did not want to risk this ask it may mess up his gcd. Is it worth it to do this, or is he right, and i should preserve my gcd for my other abilities?