Sometime she dies, yes. With 6 healers (4 of them priests with variable amounts of brain lag), that can always happen at growth 15 and higher, depending on cave-in, silence and kick luck. Actually, we've been up to growth 18, so it doesn't necessarily have to proc in 15 to be useful.

I agree with you that one should never rely on the Scarab to proc. Still, in that situation, I don't see which trinket would be better. I do have enough mana regeneration there, I do have enough +heal there for downranking in the earlier phases. In the high phases, I am chain-canceling, then chain-casting max-ranks. If every heal there is higher than 5000, I'm pretty sure an additional +100 heal would not keep my tank alive, though that's certainly statistically possible. Getting that 5000 HP there quicker will - sometimes - just save my tank.

What I actually considered as an alternative: using the Skyguard trinket at Gruul for more inspiration procs.

One more thing about the Scarab: it is one of only a few trinkets that scales with the rest of the equipment. And as such, I will be sure not to shard it anytime soon.

to Beans - there's no limit to the times it can proc. other than internal cool downs there is no such think as a limit to the number of times an item can proc. Even PPM enchants like crusader can sometimes proc 2,3, even 4 times in quick succession. They simply have varying percent chances to activate based on factors like weapon swing speed.

Eye of Gruul has a flatly stated proc rate, and a tested lack of an internal cooldown. and it also reacts oddly with multiple target heals liek Chain ehal and CoH. So yes it is good for CoH spammy fights.

to Sikora- well, its more useful than the haste proc on Scarab imo. HOW useful however depends on several factors that i can't really answer for you. Such as how much value do you place on the 70 healing ( or other effect you'd have in that trinekt slot). How quickly you can activate and apply the haste effect and still use it effectively, and of course how much you usually cast HT.

In an ideal case I believe the trinket and say a lucky string of crits could have you landing healing touches almost as quickly as a flash heal. Certainly sounds powerful, and the mana saved isn't totally worthless either, but assume you decided you need the haste, do you cancel the current cast and click a trinket and hope that the canceled cast + haste cast time is shorter than just letting the original cast finish or Cancel cast and just NS a heal. For healers haste just doesn't really seem that good in comparison to skillful cancel casting, pre casting and that sort of thing. Also, I don't know any dreamstate druids so I don't know your casting style. Do you still roll a stack of lifeblooms then HT in between? or use longer running HoTs + HT, etc, just straight HT+ cancel casting etc etc. Anyway I think HTs long cast time is benefited by haste on like your rings and bracers, but when you could put 40 mp/5 in that trinket slot over this trinket, its hard to argue for it assuming you use a significant number of non -HT spells during any normal 15 second period, or that you cancel cast Hts regularly.

all in all I would steer away from non-permanent haste unless you can think of some situations where spell haste like this would make a big differences and only lasts 15-20 seconds long =P For example if I was a DPS class I would say "Man I'd love 25% haste and a mana discount for those damn shades in phase 4 illidan".and think of a dozen other situations where 15 seconds of haste is very good but as a healer all i can say is ... I guess its good for 1-2 fel rages or used late in phase 3 RoS, and even then it'd be tough to argue it over Essence of Martyr's own constant and burst throughput boost ( shorter cooldown, constant + healing Vs larger mana consumption and higher short term HPS)

So thats my opinion, here are some numbers which I don't find particularly useful but if they help you so much the better.

Wuushoolay's trinekt if you chain casted rank 13 HT and had a crit every single time for 15 seconds ( and started with NG up) would at best save you around 405 mana ( 965(base mana cost)x .05( trinekt discount) x (15/1.7(hasted castime) and give you again in a wildly unrealistic best case 8.9 casts ( assuming no lag as well).

Follow up: Tested the trinket (eye of gruul) in a Gurtogg fight where I spammed CoH for 6 minutes. Proc'ed 9x, so about 50 mp5. Definitely will continue using it on coh heavy fights.

COCAINE IS A HELL OF A DRUG - RICK JAMES