As everyone knows, the 3 piece aggro reduction ability for mages is a fixed, -100 threat reduction on Scorch, Frostbolt, fireball and Arcane Missiles.

This has various strange effects, given how it scales (i.e. it's better for Scorch than 8 piece arcanist, but very poor for Fireball etc). However, I've been doing some testing on spells recently, and found some very wierd things.
In particular, if you take off enough damage gear to drop your rank 1 frostbolt to below 100 damage, it appears that it deals absolutely 0 aggro no matter how much damage you do.

As a test, I had a hunter friend shoot a level 56 mob in Winterpsring, taking it to 94% health. I then proceeded to nuke it from 94% health to 0% health, without getting aggro once. We repeated this several times, with the same results.
What interests me though is - does the threat reduction actually allow a mage to reduce his threat level on a mob? i.e. if your rank 1 frostbolt only hits for 80 damage, are you reducing your threat level by 20 every time you cast it?

I know it would only be a tiny amount, but it's better than doing nothing when your close to the aggro ceiling on some some mobs, and if it does actually lower your aggro level rather than just count as 0 aggro, I can see the occasional use of it for guilds like mine where BWL isn't on complete "auto farm" status.

Problem is, I can't think of an easy test to see whether or not it does actually lower your aggro. The best bet would be to get aggro on a mob off someone else, and then cast rank 1 frostbolt and see if the mob returned to the other player - but that would require very precise amounts of damage being cast.
Anyone think of an easier way to test this? Or indeed, does anyone know the answer of whether or not the 3 piece bonus will reduce your threat per cast to below 0?

(Naturally, like all mages, I'm praying for a proper aggro tool in 1.11. I can but dream!)

That's a very interesting hypothesis.


I guess one way to test whether or not sub-100 damage Frostbolts actually reduce threat is:

Have a friend aggro a mob and do x damage.
Stand in melee range and pull aggro by doing y damage.
Note y - x = C.
I'd assume that in order to revert aggro back, you'd need to do C + 0.1 * x hate reduction (the assumption is that aggro will not exchange until you have done not only C hate reduction, but an additional 10% further threat reduction).

Since you are working with a very limited Frostbolt damage range, you can probably estimate a fairly accurate threat reducing number per cast.

Two things I am going to point in what you said.

Its not scorch, only AM/Fireball/Frostbolt.

And its better if you test this on Elite mobs.

To what your "test" seems like, your hunter friend pulled with an Aim Shot, taking off 6%. Then you hit with a normal spell (which was not enough to pull agro) and the second crit and it fell over. So... you tested nothing.

Its just like me going off and shooting the level 7 Murlocs in Goldshire. Hey look! I didnt get any agro from the mob either!

3-piece NW was amended to include Scorch.

Sorry, but that's not right.

Netherwind was altered to affect Scorch a couple of patches ago - it's the 8 piece bonus that doesn't work with it.

And re. the mob - my friend hit it with a normal shot, not any special.

It took a *substantial* number of casts to kill the mob, as each frostbolt was only doing 80 damage or so a hit.

Roughly each frostbolt took about 5% to 8% off the mobs health. He just stood there getting hit by the mob as it slowly died.

I'm assuming he used a very low rank Frostbolt. How much damage were your Frostbolts doing from 94%-0%, OP?

Edit- Beaten.

Hmm... my bad on that one. Shows you how much I know about scorch (its not even on my bars :P)

Well, this test doesnt prove anything if you generating "negative" agro. It simply shows that you where hitting the mobs for zero agro (or a negative amount, but I think its more of the first) and thats why the mob would not go off your friend.

Also, you have not mentioned if your hunter friend first shot was a crit, or how many crits you had.

When I tested this months ago, I could punch a 5k hp mob once for 30 damage and then stand there, and a mage with 3/8 NW could chain-cast frostbolts for 70 damage to bring the mob from 99% to dead and it would never turn away from me. If he put on more gear and his frostbolts did 110 each instead, after ~4 bolts it'd go for him. Seems quite clear-cut.

Hmm, seems I missing a key in his post.

Disregard what I said in my last post, but I dont think you can generate negative agro.

I can confirm that doing less than 100 damage on a frostbolt generates no additional aggro (same applies for scorch/fireball). I've never done any test where I've been satisfied enough to say that it does (or does not) reduce overall aggro. But it's certainly a possibility.

You cannot, however, go below "0" aggro overall (ie, just enough aggro to be on the mobs' threat list).

If you want to test whether it creates negative threat you could have someone hit a mob, then hit it yourself til you got aggro. Then use frostbolt rank 1. If it switches back to whoever hit it first it does, if it doesn't it is just zero threat.

It would be useful because for guilds like mine, where BWL isn't a complete MC farm fest, we still have the occasional aggro ceiling incident for mages, where they end up doing nothing at all for the rest of the fight. Particulary on Broodlord - there's always just ONE who goes too high and ends up with aggro.

Being able to continue dealing damage and actually lowering your threat at the same time would give them something to do, until they actually learn how to control aggro a bit better... :)

Obviously, this is not going to be something that's used often, or that helpful, but as a curio it is interesting to try and work out if you can lower aggro using R1 frostbolt.

---

Of course, the latest announcement about the mage changes sounds interesting -

Aggro management for mages?

I'm vaguely interested in this. What tests were run to discover this?

I have been wondering for some time if it was possible to drop below zero threat but still be considered in combat, particularly with the subtle and delicate Boss "in combat" mechanics.

Pretty simple to test, I think. Get one person to bodypull a mob (i.e zero aggro but on the list) then have a rogue open with a feint, then a white damage. If I remember correctly, the rogue will still get aggro even though feint should reduce aggro by about a thousand. Please correct me on this if I'm wrong.

In my personal opinion on Broolord it comes down to tanks rather than the DPS. Parrying/dodging knock aways and not dying to mortal strike generally keep you nice and high on aggro. Comes in time with gear I guess :D

Our solution to this was to just kill him faster. Ever since we told people to just burn him down as fast as possible, we haven't had a caster pull aggro before 5% (we did have a hunter pull aggro at 50%, which resulted in a very emberassed hunter and a reminder to wait for FD to actually clear aggro). With everyone going all-out, he dies before the tanks hit the aggro limit, and the threat is spread out more.

Hmm... "Wait for FD to clear aggro" It doesn't do it immediatly?
To the OP, im assuming that no spell in the game can do both, that is damage and negative amount of aggro, just reduction to the aggro done on the skill... since we are alone with that type of bonus, wish it was 10% or something instead...

Has anyone tested whether this is applied before or after threat multiplicators?

My impression is that it is applied after, which would make it significantly more useful to players with high threat reduction. Arcane missiles for 1600 damage with 58% threat reduction (BoS + subtlety) would hit for 572 threat, whereas it would be 630 threat if it was applied after.