Actually it depends on your hp pool.

If your hps are at about 7.5k or higher mitigation starts to do more for you than hps do per gear point invested.  (obviously for tanks dealing with strike through this isn't always the case).

So if you're getting raid buffed to 8k + its a safe bet that mitigation on your gear will do more for you than hps at that stage.  This is completely ignoring the point that mitigation provides better healing efficiency (makes it easier for healers to keep you topped off with AOE heals etc.)

To look at it simply. 
5% mitigation on 7500hps is about a "free" 375 extra hps.  Pretty close to the raw hps you'd sacrifice on gear to get it.
5% mitigation on 9000hps is about a "free" 450 extra hps.  Well more than you get from the gear pieces so its a hands down better deal.

I've got about an extra 10% migitagion coming from gear and raid buff to between 8k to 9k depending on who's present.  There is no way I'd get 800 to 900 raw hps from the gear I'm getting that mitigation from.


That said if you're raid buffed below 7500hps...you definitely need to focus on getting your hps up first.  You'll get better returns. 

And yeah simon, you're right on the money in retrospect.  I had never looked at the psi buffs closely.  Oh well.   
On the bright side thestran boosts damage to fire and ice...so that's always nice. 

It really is as simple as I laid it out.

To go by your approach however, consider that in scenario 2, your monk lost more %wise in the first hit but that's merely a matter of your choice of how many hps you feel the 8% mit would cost in gear.  The part your missing is that the remaining 32%hps on the monk with mit gear is more valuable than it would be on the monk with 33%hps left.  AKA he's also going to mitigate the next attack incoming as well.

In a nutshell the glitch in your math is that it goes with the view that only the first attack is mitigated.
Consider your setup with 10khp and 8%mit. 

Scenario 1, the monk with 10khps and no mitigation dies in two hits.  (10k -5k -5k = dead)
Scenario 2, the monk with 9300hps and 8% mitigation lives (10k -700 - 4600 -4600 = alive (barely))

So its CLEARLY more beneficial at that point to go with mit.

Additionally mit doesn't "seem" to cost 170hp per 2% I'll have to double check but I felt it tends to be more around 130-150ish?  Some peices of quested dropped jewlry seem to have mit at a good value, RI jewlrey has over 2% on a single piece.


But to put it into a more practical situation, most apw attacks are of two varieties.  Punishing attacks for very high damage, and sissy but frequent attacks.  With decent hps and/or mit you're going to typically be able to take about 2 big hits.  If your mitigating more of that it will be easier for a healer to keep you safe without peeling aggro.  AKA most healer's "big heal" with reasonable gear heals about 5kish.  If you're taking notably more than that it becomes very hard for them to keep up.  If you're taking less than that its much easier and sustainable.

If you are getting nailed with damage sheild hits, constant small aoe hits etc. more mitigated once again makes it more likely for a healer to keep you up and running easier.

But this is assuming you've got all the important stuff nailed out already.  Like having enough hps to make mit worth it.  That and, once your survival is "good enough" ... unless you tank a lot in groups you'd be better off going for more %damage/stats/crit etc.

Either way going for a lot of hps or a good amount of hps+mit will get the job done. 

You are correct most AOE are spell based, that said mitigation adds to both melee and spell thus it works on ALL AOE.

You're thinking about physical absorbtion crystals (melee mit only)

Protection adds general mitigation = mit vs all.  Hence it works peachy on AOE.

Yeah I understand the dynamics of using protection crystals (hp/energy trade off for mit, hps also range further than 230) etc.  One of my alts is L50 leatherworker. 
I just didn't do the math to see how many points of hps gets converted to how much %mit.  Its likely to be a static value regardless of what piece you use.  I just didn't take the time to peg it exactly, might do that this weekend just for purposes of accuracy.  /shrug

That said I've often considered the merrit of making a few peices of spell absorbtion (head peice etc.) to use in conjunction with protection just for situations where you know in advance you'll take heavy spell damage.  (spell or melee only mitigation gives a fair amount more %mit per gear point than general mitigation does, and as you pointed out most AOE are spell based so it may be preferable to skip general mit in favor of spell mit in APW).

Of course...its damn annoying that its essentially impossible to get med crafted gear without some worthless stats on it other than pariah's for monk.  Which of course loads us up primarily on dex...there isn't a good way to get a large amount of str/int without also taking on vit/wis which does us precious little good.

Edit took a little time.  1point of general mit = roughly 2.7ish hps  And 1point of mit = roughly .027ish% mit.
So roughly 100hps converts to 1%mitigation.  So my initial guesstimate was pretty off.

So yeah takes A LOT of hps to break even. 

Still I like it. 


Incidentally crit bonusgive slightly less bang per hp spent trading at about 1 point per 3ish hps or roughly 111hps per 1%crit.


Specific mitigation (melee or spell) gives a much better return per hps sacced, but also (obviously are not quite as versitile).

Outside of APW I'd say melee mit is vastly more useful, but inside I woud lean towards spell mit being more useful.

Actually it does and doesnt.  Items do have set points to distrbute but i think the reason you see such variance is its probably item dependant.  As in to put crit at all on an item costs X points, then the amount of crit is based on item specific multipliers.

Im sure the code is so convoluted that its not a simple 20 points = 100 hp or 50 mitigation.