New GC post about pally haste.

[Nooska]

RPPM would be broken with that sort of proc in my opinion. The more mobs you have hitting you the more guaranteed you will have full uptime (even more than now) of ShoR.

[Sagara]

RPPM with ICD then?

[Nooska]

RPPM (which gets modified by haste) and a ICD of ~1 second (modified by haste) I would think, that way you can get the proc between 2 uses of GC and benefit fully if you are lucky, but without being guaranteed a proc between each.

(Nota bene; those numbers are just thrown out there as I should have my shoe son and be going out the front door to go to a meeting right now)

[Fetzie]

There isn't a problem with warriors using revenge when it procs (i.e. all the time when AoE tanking), why do we think doing the same with AS would be any different?

[daishan]

There isn't a problem with warriors using revenge when it procs (i.e. all the time when AoE tanking), why do we think doing the same with AS would be any different?

Revenge is only extra dps right?
I'm thinking about the high SotR up time we'd be getting, we already seem to take less dmg than druids, dks, and monks when aoe tanking, with the new GC model we'd end up been even stronger.

[Fetzie]

Revenge generates 15 rage when in defensive stance. 4 of them and you can cast Shield Block, which means you are guaranteed to block for 6 seconds and those blocks can be critical (60%). And Revenge resets its cooldown on EVERY dodge and parry.

Revenge: http://www.wowhead.com/spell=6572
Shield Block: http://www.wowhead.com/spell=2565

[daishan]

Ah ok I stand corrected then.
It just seems that Warriors aren't a long way ahead of us in aoe dps or mitigation (I could be wrong about that not got a warrior in the guild never mind a prot one), so us getting a sizable buff to tanking large groups of mobs might lead to us getting nerfed elsewhere to maintain "balance".
Not that I'm in anyway against been op

[Flex]

My overall concern with this train of thought from Blizzard, is how we're going to suffer DPS wise.

Top 100 parses on Raidbots for 25H has paladins as the top DPS of a tank. All parses have them second by a wide margin behind monks. There doesn't seem to be a huge DPS difference at a base class level that this should effect in any major manner.

[Flex]

Revenge generates 15 rage when in defensive stance. 4 of them and you can cast Shield Block, which means you are guaranteed to block for 6 seconds and those blocks can be critical (60%). And Revenge resets its cooldown on EVERY dodge and parry.

Revenge: http://www.wowhead.com/spell=6572
Shield Block: http://www.wowhead.com/spell=2565

Revenge hits for a third of what AS hits for. Would you be willing to gut its damage and adjust ShotR so it is a 100% proc chance?

[Fetzie]

No, but I'm just saying that just because we get holy power from AS isn't a reason why GC should get an internal cooldown just because of multi-target tanking.

[Koatanga]

The maths I did were slanted to minimise the influence. I don't have stats on boss swing speeds, so I used 1.5 seconds as convenient maths because we do our rotation at 1.5 seconds without any haste. I also computed our HP gen based on 0 haste, which is unrealistic since we like haste and tend to stack it up a bit.

So the nerf to our DPS on a single-target boss fight should be noticeable. 25% fewer AS means fewer SotR because we're not getting the HP from using the AS in the Grand Crusader window. Likely the nerf will be greater.

When we're not tanking, we're losing 3 x AS per minute (using very conservative maths), and therefore 1 SotR per minute. That's a significant DPS loss.

Yes, we get a DPS boost when we face multiple targets, but then it becomes a selection issue for progression raiders - sit the pally tanks on tank-swap single-boss fights, and use them only as AoE tanks. Is this BC again?

[theckhd]

Well, your math is also incorrect, so there's that. But you're right it's a fairly sizable nerf.

[theckhd]

Correct version:

Previous GC: 20% chance on CS/HotR gives an average proc rate of 0.2*(1+haste)/4.5 procs per second. At 0 haste, this is 0.044 procs per second, which is also the amount of HP generated. This assumes that you can take advantage of that proc before the next CS, which is universally true in CS-J-X-CS-X-J-CS-X-X.

New GC: 30% chance on avoid. Assume a 1.5-second swing timer and 20% avoidance after the -9% boss penalty (reasonable, it's about what I have on Theck with around 10% haste rating). Average proc rate of 0.2*0.3/1.5=0.04, about a 10% nerf to the generation rates. In general it will be a little lower because of bunching (i.e. 2 avoids in a row trigger GC before you can use the first proc).

In practice, the bigger issue is that we lose about 50% of that benefit because of tank swaps. In my opinion that's a problem - it should slant a little on the heavy side in order to offset that issue as well as our reliance on a fast boss swing timer. A ~50% proc rate (0.2*0.5/1.5=0.0667) would be a little more reasonable.

That said, the survivability simulations are quite interesting now. Mastery and avoidance are giving haste a solid run for its money.

[DisRuptive1]

So unless I miss my guess, our AS procs just got an 18% nerf.

Your math was right but you subtracted 50 from 68 when you should have divided 68 by 50. It's actually a 36% nerf if the rest of your logic is correct.

[theckhd]

Yeah, but it's not. He's calculating the chance of having 1 or more AS procs in ~9 attacks, but only giving that probability 1 holy power worth of weight. When in reality, you'll average more than one proc in 9 attacks. My calculation just looks at the steady-state time average; in theory if he does his calculation properly by counting up all the binomial statistics, he'll get my answer.