Aggro model -- critique requested

[Yakk]

I posted an aggro model over at Graffe:
http://www.graffe.com/forums/showthread.php?t=31242

My current aggro model:

1> Hate from spells is the same resisted or landing or criting
DD Spell Hate = N * Spell_Damage

Foci, crits, and resists do not change the aggro from a spell.

N is some constant, probably 1.

2> Hate from a weapon swing is:
Weapon Swing Hate = K * (WD + DB)
and is the same if the swing hits or misses
(where K is some constant, WD is the weapon's damage rating (not the actual damage done on the swing), and DB is the damage bonus).

It seems likely (based off defensive having lower aggro) that spells like Lion's Strength increase the "Weapon Damage" portion of this hate.

K is probably around 1.

3> Hate from aggro modifying spells is the same whether or not the spell is resisted. (a hate-over-time that is resisted will, naturally, not generate hate-over-time)

4> Aggro = Hate + AggroBonus
AggroBonus includes things such as:
a> Sitting
b> Being the puller
c> Being in melee range
d> Low HP aggro
e> Player level relative to mob
f> Etc

With a few exceptions (pets, charm, root, etc) the player with the most Aggro is attacked by a mob.

5> A successful "taunt" gives you
max(all players Hate for the mob) + T
where T is some relatively small constant.

Note that "Taunt" does not take into account AggroBonus, so it will give you the most total Hate but not always the most total Aggro.

6> One's first action against a mob acts differently. For nukes, the hate generated is boosted by roughly 30%. For pure +hate effects, there are reports they don't work that well as initial aggro.

I think most of this is accurate.

Also in that thread are a few totally-theoretical case studies, to see if the numbers generated add up right.

Anyone see any errors or caveats I should add?

Things I don't understand include: how much roughly/exactly is stun/snare/ATK debuff hate, and kick/bash/flying kick/other "skill attack" hate.

(I do know that "debuff effects" like slow/snare/ATK generate an amount of hate that grows with the creature's power/level/HPs to a certain extent.)

[Phantron]

Being puller isn't a temporary aggro modifier like the others are and is subject to taunt. Whoever is the puller simply has a fixed head start compare to everyone else that is overcome the same way as everything else.

[Grayhelm]

Wasn't there some proof of 'initial hate bonus', so whoever gets on the hate list first has a bonus? I could be wrong, this is a question not a statement

[Elladon]

It seems likely (based off defensive having lower aggro) that spells like Lion's Strength increase the "Weapon Damage" portion of this hate.

Didn't we conclude that discs and spells that affect damage have no bearing on aggro at all? Leaving purely the ratio and direct mods to be the only thing that can change melee hate.

[bardeil]

Originally Posted by Grayhelm

Wasn't there some proof of 'initial hate bonus', so whoever gets on the hate list first has a bonus? I could be wrong, this is a question not a statement

They do, but it's not temporary. It's just 'free hate', as if they threw an extra nuke.

[bardeil]

Originally Posted by Elladon

Didn't we conclude that discs and spells that affect damage have no bearing on aggro at all? Leaving purely the ratio and direct mods to be the only thing that can change melee hate.

I believe you're correct. I don't think you'll find many tanks arguing they lose aggro with defensive on.

[Yakk]

Originally Posted by bardeil

I believe you're correct. I don't think you'll find many tanks arguing they lose aggro with defensive on.

Originally Posted by Elladon

Didn't we conclude that discs and spells that affect damage have no bearing on aggro at all? Leaving purely the ratio and direct mods to be the only thing that can change melee hate.

Interesting -- last time I looked into it, the consensus was that defensive cost you some, but "not as much as you would think", hate.

Anyone have a link to the thread in which it was proven/disproven?

Originally Posted by bardeil

They do, but it's not temporary. It's just 'free hate', as if they threw an extra nuke.

I thought this was the same as the wizard - discovered +30% "first action gives you this much more hate" effect.

For example, if you pull with dispell, an arrow will turn the mob away from you.

But if you nuke with Strike of Solusek, it takes 30% more concussions to turn the mob than if you first dispelled then nuked with Strike of Solusek.

This consistent with what you have experienced/read about puller hate?

[Phantron]

If any damage related mods (discs, attack, whatever) have anything to do with aggro it's not in any degree you can actually tell, so they might as well not. Frankly since you get the full weapon swing damage aggro on a mob that's completely immune (any mob that requires bane to hit), it's hard to imagine a mob would care about anything else related to damage.

[Frodlin7th]

The following factors also work into the hate formula:

Mob Max HP

Mob Max Resists

Whether a spell is a "class spell" or not (i.e. a stun proc produces a different level of hate than a cast stun, even if it's the identical spell).

[Yakk]

If any damage related mods (discs, attack, whatever) have anything to do with aggro it's not in any degree you can actually tell, so they might as well not. Frankly since you get the full weapon swing damage aggro on a mob that's completely immune (any mob that requires bane to hit), it's hard to imagine a mob would care about anything else related to damage.

Other than the raw weapon damage of the weapon?

The thing is, a few abilities (champion, lion's strength, defensive) change the effective weapon damage interval of your weapons.

Defensive does this the most: 55% weapon damage interval reduction. A 20/20 weapon acts just like a 11/20 weapon if you have defensive up, for most purposes.

If it acts like a 11/20 weapon for aggro purposes (main hand), that's a 25% "pure swing" hate reduction compared to a 20/20 weapon.

edit: Remember, I'm saying if. Such a reduction in aggro isn't huge, and might take effort to notice, especially with most of your hate coming from other sources (procs and incite).

The following factors also work into the hate formula:

Mob Max HP

Mob Max Resists

Whether a spell is a "class spell" or not (i.e. a stun proc produces a different level of hate than a cast stun, even if it's the identical spell).

I remember the "proc aggro reduction" thing a while back (so tash-proc weapons wheren't the best aggro in EQ, heh). Your claim is that this is a 'class spell' vs 'non class spell' difference, not a matter of 'proc'/'item sourced spell' vs 'cast spell'?

Is there an aggro-proc weapon that shows this to be true?

Back in the SoV/SoL days, they changed the "aggro is a linear % of hp" aggro on spells like tash, and made it a more complicated formula.

I never saw any evidence personally for your last comment -- Mob Max Resists impacts "status effect" hate amounts?

[Yoda]

there isn't anymore any item that procs/clicks a class-specific spell. Since the last patch many things have changed that make hate generation even more different from classes to classes than before.

[Yakk]

Originally Posted by Yoda

there isn't anymore any item that procs/clicks a class-specific spell. Since the last patch many things have changed that make hate generation even more different from classes to classes than before.

Well, there are items that click class-specific spells still. My wizard has one.

Aha! Enchanter colour stun wand procs an enchanter-specific PBAOE stun. Velk has some clicky enchanter stuns as well.

Using the two one could work out if it is "proc" "proc/cast" or "your class spells".

The Warrior DoN clicky (anger) is another data point. If it generates full-scale stun aggro rather than proc-stun aggro...

[Torrinx]

http://lucy.allakhazam.com/item.html?id=7822

same spell/proc

[Frodlin7th]

I remember the "proc aggro reduction" thing a while back (so tash-proc weapons wheren't the best aggro in EQ, heh). Your claim is that this is a 'class spell' vs 'non class spell' difference, not a matter of 'proc'/'item sourced spell' vs 'cast spell'?

Is there an aggro-proc weapon that shows this to be true?

Back in the SoV/SoL days, they changed the "aggro is a linear % of hp" aggro on spells like tash, and made it a more complicated formula.

I never saw any evidence personally for your last comment -- Mob Max Resists impacts "status effect" hate amounts?

The check is whether it's the class' inherent spell or not

The other two are factors, I have no idea what the magnitude of those factors is, just that they are indeed factors.

[Yakk]

Originally Posted by Frodlin7th

The check is whether it's the class' inherent spell or not

The other two are factors, I have no idea what the magnitude of those factors is, just that they are indeed factors.

Thank you very much for the information. And I do appreciate it -- being aware there are problems/errors/potential issues in the model is important.

However, I would really really appreciate a source. Did you hear this from developers? Did people work out this is what was done? That is why I'm asking for more information. =)

And Torrinx, we need a weapon that procs a spell that the class can use, to have a test case that distinguishes between "proc aggro is different than cast aggro" and "if you can't cast the spell, the aggro is less".