Pretty sure people were just advocating that at low gear levels during the start of the expansion. SP starts scaling up well the more gear you have. Its totally useless to compare what stats provide qualitatively you need to compare them quantitatively, or its the numbers that matter.

Ups, sorry for that heavy metall issue...


Back to overheal. I see your point and I will have to put some more thinking into this, but here are some first thoughts anyways.

Obviously any overhealing modell will have to operate with estimates and rules of thumb, while this is a problem, not accounting for a known effect at all is too.

The way i play my healadin i end up with around 50% overheal, if not more. I will use the logs of our last ulduar 25 Razorscale kill as an example.

My crit HLs averaged about 4,9 k healing, and the normal hits about 4,7 k. Droping the 0.2 diffrence for the argument all i got of my 60% crit rate was a nice amount of mana, but no healing at all!
Considering that a HL heals about 4,9 to 5,4 without any SP, even the SP wasn´t worth much of its i lvl point.
Now, I am not quit sure if mp5 gives more mana than crit rate per i lvl point with my gear/raidbuffs in a 10 minute fight, in which i didn´t run into much mana problems anyways, but soft caping haste rating would have been worth it for sure.

For me, haste seems to be immun to any overheal argument, as is mp5. Crit. and SP effectivness could be modelled by average healing. Let us say my gear averages 10 k non crit HLs while in the considered boss fight i only average 6 k.
The question would now be, how to evaluate the SP accounting for HLs over 6k. This could be done by any distribution, i would probably use a "normal" distribution (this is a direct translation from german, since we call it the "normal Verteilung", it has a bell form, graphically.).
This could be done analog for crit. rating, but since a crit. is usually not the live safer in fights where it is often overheal anyways, one may use a more agressive formula. I would advise using simply the healing set by the average crit. heal relativly to the average non crit.

I haven't seen a mention of this: Pandora's Plea's proc is actually 751 spellpower instead of 850. Source: char sheet with proc up and down, tooltip of the buff.

I've healed in 25-man Ulduar in fights where I've done nothing but FoL HS and overhealed by 65%. You can never assume that FoL HS won't overheal. Someone with less latency than you or an instant heal vs your casted one or who has more haste than you could heal someone immediately before you do. The hit that requires the heal might hit right after your heal lands. You could be FoL spamming the tank, HS random ppl in raid, and tossing a HL only when you see the tank has dmg and still OH 50-60%. There are any variety of healing assignments or raid situations where you can overheal using any spell. Usually, there is less OH from Flash spam than HL spam, but not always. It just depends.

I'm not sure that OH settings would reduce the effectiveness of crit rating that much, since the crit is more of a mana conservation thing than a healing effectiveness thing in many cases. I CAN cast more Flash now because I have a high crit % and spell power.

That said, I think little gauges for HL/FoL mix + average OH % (since I know given X people in a raid I'm going to OH probably by Y amount - it's really consistent) is a good idea. However, I've personally found that RAWR gets within about 200k of my actual healing output on an 8min fight - which is pretty amazing. Also, most of that 200k is the difference between RAWR calculating the amount of damage absorbed by Sacred Shield as a healing amount - which clearly does not appear on meters/web stats.

The fights where it is really far off are the ones where we really have too many healers and I'd be just as useful, if not more useful, by not being there.