That's the point. It always affects the whole batch. Either all, or none.

Seems like that's the explanation. In Arkoth's raid 10 logs the effect is as we should expect it.
Here is a raid 25 log on Yor'sahj (no MS mechanics on him) with two Paladins with the normal version on the mace : World of Logs - Real Time Raid Analysis Add a query with spell id 107835 (normal version of the proc) and run it.

Most of their Cleansing Flames' procs are on 15-20 people and are at 50-60% of what they are supposed to give. But the rare times there are only 2 targets hit by the proc they got the expected results :
Seems like there is a cap to the maximum healing capacity of the proc, just like the old AoE's cap mechanics.

From first post:

Healing Rain ticks indeed proc Jaws. Earthliving and Riptide ticks however, do not. The (1 second?) internal cooldown is also still in effect, which makes 'vastly better' a bit of an overstatement in my opinion. It's still a struggle to reach 10 stacks in 20 seconds.

I'm running a Focused Insight build with reforging for Mastery, and I'm using 378 Necromantic Focus with DMC:Tsunami. So far for our Heroic progression it has worked much nicer for me because I've been able to keep from running oom with a a 1.5-3k HPS drop. Having more mana has let me use quite a few more emergency heals to keep people from dying that otherwise would have.

My question is if I get my hands on a 391 Necromantic Focus, is it worth it to keep over using 2 397+ trinkets for as long as I use the FI build?

No, the Heart of Unliving is better than the tsunami and replaces its function as a regen trinket. The SotSS would be better than the Necromantic Focus because the haste proc is nice when it lines up with heavy healing phases and if it does not line up in such a phase, you can use it to TC. As an added bonus you won't need to keep flame shock up and can just use it to FI a healing rain.

I don't see any reason to use older tier trinkets given what we've been given in those slots this tier.

Has anyone taken a look at crit vs haste recently? I've been running some numbers due to an argument started in the wow healer forums about it. Here is what I'm getting. If my math is wrong somewhere or I've made leaps of logic please let me know.

128.06 points gives 1% haste. 179.28 points gives 1% crit.
The formula for combat mana regen boils down to 1172 + .63*Spirit. Since the square root of our intellect isn't going to budge much at all I'm assuming a value of 75 for it to simplify some math.

At 5% crit and 7% haste the numbers come up to 896.42 for haste and 896.4 for crit, so I will be using that as my baseline of comparison.

100 casts of GHW(or HW) at 7% haste will net 107 heals worth.
100 casts at 5% crit will net 105 heals worth + 1.5 heals worth(from Ancestral Awakening) + 11465 mana(230mp5 over that time frame).
This boils down to 107 for haste and 106.5 + 230mp5 for crit. The numbers for Chain Heal are similar, 107 vs 106.5 + 300mp5. So while haste has a .467% lead in throughput, the mp5 gained from crit could potential be used to reforge spirit into another throughput stat, putting crit just slightly ahead. Riptide and Unleash Life crits also give back mana but are not as simple to show with their CDs, nor are they spells we can spam on a continuous basis.

Crit for Hots scales the exact same as for normal heals 1:1, but no mana is returned on crit hot ticks. Haste however is only useful at the breakpoints. For non-goblins, 916 haste will give you 1 extra tick each for Glyphed Riptide, Earthliving, and Healing Rain. The next breakpoint is the 9 tick Riptide at 2005. Haste provides no additional benefits between these two numbers, whose difference is 1089 stat points.

Since RT is on a 6 second CD you can only ever cast it 10 times a minute, so we are either looking at 80 ticks in a minute or 90 ticks, or just over 12% increase in RT's effectiveness. But since it is on a CD it is rarely more than 25% of our total healing even using perfect rotation, so that 12% increase pairs down to about 2-3% increase in total heals. Whereas that 1089 rating would get us about 6% crit, which is then useful across the board.

Is there anything wrong with this analysis? Other than the fact that haste will be more even healing whereas crit will obviously be spikier is there any other reason haste would be preferred. The RNG shouldn't really come into the argument since over such long fights it will move towards the center, and the more crit you have the more even the RNG will become.

These two statements are where I have to disagree with you. Although those breakpoints do make a difference and are useful to hit if you're within a reforge or 2 from either of them, I've personally gotten a lot more from haste than a healing rain or riptide tick. I use a TC build (particularly useful this tier) and the more haste the better. I'm only 2/8 HM atm so I can't speak for other fights, but there generally isn't always 2 seconds of free time here and there and I want as fast a cast as I can get away with so i can fit as many casts in as I can.

Morchok and his aoe phase is a great time to stand and cast (get behind 2 on his pillars right next to each other and cast between em) or Yorsaj when he's summoning a new set of oozes (assuming your tank isn't dying from a previous stack of dots)

Don't get me wrong, I don't think crit is bad, or not also useful for TC (I love when my lightning bolts crit) but I don't want people thinking that haste doesn't scale somewhat linear in throughput aside from the spikes from the breakpoints.

There are 4 ways that haste can help provide throughput to a resto shaman:

1. You get extra ticks for HoTs at the haste breakpoints.
   • According to the guide at the beginning of this thread, the hps gain from reaching a breakpoint is less than you might think even in ideal situations. And if you refresh a HoT, that tick doesn't matter at all, further reducing this benefit. Sometimes you refresh HoTs, sometimes you don't. So the true benefit to reaching a breakpoint will be more than 0, but less than the theoretical maximum -- which isn't that much anyway.
2. HoTs tick faster with more haste.
3. Cast times and global cooldowns are reduced, allowing more casts in the same time.
   • These second and third benefits scale linearly with haste; with 1% more haste you will get 1% more ticks or casts in the same amount of time. You will also spend 1% more mana, but this cost can be offset by telluric currents.
4. Any extra ticks or direct heals that haste allows will give more opportunities for helpful procs like earthliving and equipment bonuses.

In short, and without getting into the math behind things, reaching haste breakpoints does help a little bit -- but that doesn't mean that haste between breakpoints is useless. More haste is always better, whether it helps you reach a breakpoint or not. But to belabor the obvious, more crit is also always better, and so is more mastery.

Every time I crunch the numbers I find that haste, crit, and mastery all improve healing throughput by similar amounts. Unlike some dps and healing classes, there isn't one secondary throughput stat that's significantly better or worse than the others. For an indirect proof of this allegation, note how many posts there are promoting one or another stat. There's no consensus, and people advocate the stat that fits their own healing philosophy and style.

Yea I wasn't trying to make the case that everyone should reforge into crit, I was just trying to get a consensus if my math was kosher. I agree they all seem to be +/- less than a percent difference in throughput from each other when I run numbers. I had put together a list of BiS items on the wow healer forums and there was several other shammies being very negative about crit, one even saying crit gear had no place in a BiS discussion. It just seems that alot are just having a knee jerk reaction to crit back when it wasn't as desirable a stat prior to the 200% crit heal fix. Then when I provided math to back up my analysis it was generally met with more wishy-washy 'you should hit the 2005 breakpoint b/c it is good' sort of talk.

And I guess I should qualify my statement when I said 'Haste provides no additional benefits between these two numbers.' I meant in terms of your hots haste will provide no additional benefit. Sure you're hots will tick marginally faster, but at no real throughput increase unless you can perfectly clip your hots to take advantage of that. And I think my numbers indicate that while yes you will get more casts and lower GCD with the haste points between those breakpoints, the crit you could have had will be just as good, if not just a hair better.

Shamans being negative on crit is a sad aftermath of Jynus' crusade against crit in his original 4.0.3 guide on the US forums I'm afraid.

As for this statement, the source is still there, but was recently challenged on the german forums to have been misinterpreted in the first place. I'll try to get that information available.

Thanks for the clarification. I had sort of figured as much. There is likely some point where X haste would get you an extra lightning bolt per fight given Y downtime in Z chunks, but like you said playstyle is going to be a much larger factor in that than stats.