foxglove

Well, I'm getting the impression that the Emberseer, at least, won't do the trick. He's flagged as a boss, but after over an hour of taunting him (admittedly a small sample) I had 7 resists in 494 taunts with no +hit or +spell hit equipped (feral druid). That's about 1.4%, which looks to me like the inevitable 1% spell resist. Certainly a 500-ish sample is too small to be confident on the number, but it's enough to make me feel I'm wasting my time.

Since Drakk is a big clear, perhaps Azuregos or a green dragon (are they tauntable?) would be good for testing. Pity Onyxia is taunt-immune.

 

Roywyn

Many things that are marked as skull are not actually classified as real bosses.
If you don't glance against them and they don't crush you, it's most likely not a proper boss.

 

◊ Penguin

Going to jump on the 'odd problems with Taunt hits' bandwagon here. I had a taunt resisted on Hyjal trash while wearing 152 +hit rating. It was a Gargoyle mob, according to my chatlog.

 

nfw

You don't need to find a skull mob, any +3 would do. So a level 67 can test it by taunting a level 70 mob repeatedly.

 

foxglove

Well, I gues s this depends what you call a "real" boss. Emberseer is listed as a "boss" in wowhead, for example; I don't know how exactly wowhead gets this data (but it's probably the same condition that results in the display of the little skull, regardless of level, as you mention).

The presence or absence of glances and crushes should depend on the mob's defense and weapon skill versus the character's, right? So, it's conceivable that the level 60-tuned bosses have a mere 315 weapon skill and defense, and consequently a level 70 player will not see crushes or glances. (Can anyone confirm whether a raid boss like Ony crushes a level 70?)

My goal is to find a boss mob that can be tested by an individual or small group in a controlled situation, since it's going to be hard to get more than the anecdotal "I got a resist" from raid logs. Even in a situation where a tank could get away with sitting around spamming taunt on a boss (say: Al'ar), the data set is going to be very small. Furthermore, I don't think my guild would appreciate me getting 2-shot because I wore a funny set of gear to have exactly 0% / 9% / 17% plus hit.

Edit re. bosses versus +3: Or I could read the first post of this thread. Duh. Thanks Emeraude.

 

Emeraude

Think you're mistaken. The base miss rate against a 72 mob is 6%. No way it's 5.6% against a 73, EJ threads in the class discussion forums confirm this.

 

foxglove

Based on this suggestion, I used a level 57 druid versus a normal level 60 mob to try to get a baseline resist rate in the current patch.

Character: Pianissimo on Uldum, level 57 (has leveled since testing this)
+physical hit: 0%
+spell hit: 0%
Mob: level 60 Twilight Geolord in Silithus
Total growls cast: 587
Resists: 82

That's roughly a 14% resist rate with (I think) a standard error of 1.4%. (I would have gotten a larger sample if I could, but ran out of time on "borrowing" the character.)

So, assuming the norm is 9% physical miss and 17% binary spell resist versus a +3 as the class forum threads state... in my sample, taunt still has a resist rate more characteristic of a spell than of a melee attack. It may be subject to a 1% base miss rate that cannot be overcome, like other spells, even though it is affected by physical +hit.

The 2.3 working theories of theorycrafting thread says the following:
However, I don't see a citation to any post with data demonstrating this.

I will try to get some +hit gear for a pre-70 (like the character above) and test with different amounts of +hit stacked.

Edit: I can post the combat log for this testing if people want to look at it; I just need to figure out some place to host it. WWS doesn't seem to list resist rates for "auras" like taunts so that's not an option. I don't think it's a very exciting log, though.

 

• Ren

Are you sure you weren't afflicted with Banshee Curse (Banshee Curse - Spells - World of Warcraft) at the time?

 

◊ Penguin

According to my logs, it was the 2nd wave of Azgalor trash. There's no Banshees until wave 6.

11/30 22:06:42.796 [Party] Ehandel: Taunt **resist** by Gargoyle.
11/30 22:06:42.843 [Raid] Ehandel: Taunt **resist** by Gargoyle.
11/30 22:06:58.421 [Raid Warning] Vandemar: *** Wave 3 in ~60 sec! ***
11/30 22:07:05.484 [5. aftank] Ehandel: 158 hit rating
11/30 22:07:07.703 [5. aftank] Ehandel: WTF

 

Nite_Moogle

You need 142 hit rating to no longer miss with normal attacks. Taunt's resist rate is the same as a normal attack's chance to miss. Connect the dots. Your sample is inconclusive at best, your attempt number is low and Geolords are known to have modified stats from the standard NPC template as evidenced by glancing blows against them by a level 60 character. However I feel the need to point this out: at no point does that post state that your chance to have taunt resisted will be removed by 142 hit rating since no conclusive proof exists one way or another.

Honestly until somebody goes in to Kara with 142+ hit rating and a pack of healers and taunts Midnight about 5000 times out of range of Attumen's aura there isn't going to be a good test for this, and I really doubt anyone wants to waste that much time.

Additionally, I don't think there is a single tank out there that would wear 100+ hit rating on a regular basis considering what you give up in terms of survivability to attain it. Even if there is a 1% resist rate, it really doesn't matter because it isn't practical to gear to avoid it.

 

foxglove

If this is the underlying assumption for the statement that +hit operates the same way on melee attacks and taunts, has this itself been confirmed? I remember that there was a lot of speculation in the 2.3 changes thread about the change implying the miss rate would be reduced from 17% to 9%, but (for the same reason) there were only a couple people testing in hour-long samples (a few hundred taunts).

Edit:

I'd disagree. As a feral offtank, I wear significant hit and expertise (~95 hit rating normally I think, and I make sure to cap +hit at 142 on Leotheras) for purposes of reliable snap threat. How this +hit applies to taunt is worth knowing, for those mobs/bosses which are tauntable. Edit: Consider the bear boss in ZA. The bear doesn't hurt a T5-geared tank that badly when you follow the strat, so lowering your mitigation a bit is just fine, but a resisted taunt is going to screw up your attempt and may cause a wipe. In this situation, your co-tank's +hit is a form of mitigation.

 

Nite_Moogle

Taunt - Spells - World of Warcraft
School: Physical

I sincerely doubt that they changed the miss rate of this attack to conform to its previous miss rate. It is reasonable to draw the conclusion that as a physical attack, it will obey the same rules as other physical attacks; namely 9% miss against Boss NPCs.

 

foxglove

I'd agree that it's reasonable. However, taunt resists still log as "resist" and not "miss," and taunts cannot be dodged/blocked/parried as far as anyone has seen, so there's still something a bit different about it mechanically from your standard melee attack.

 

Nite_Moogle

It's not really a special case mechanic as precedence for the things that make it non-standard already exist in other abilities. Overpower and Shiv cannot be dodged or parried, and overpower cannot be blocked. Thunder Clap uses normal hit mechanics and lists its failed hits as resists. They could have messed with its miss rate, but this seems unlikely at best.

 

◊ Penguin

Ok, so this has been bugging me now. I set up my 2nd box with my resto shaman, hiked over to ZG and settled in for some testing on High Priest Venoxis.

I geared myself with 143 hit rating, 9.07% according to the character sheet. Only buffs were Fort, Healing Stream, Stoneskin, Grace of Air, and Earthshield. I was also running with 10 Expertise value (42 rating).

Taunted 987 times
Successful 898 times
Resisted 89 times
For a miss rate of 9.02%.

Sadly, I lack any sort of stats background to do more than that simple calculation. I've attached the combat log if anyone wants to do anything more. Interesting enough, 3 that I noticed (and probably 1 or 2 more that I didn't) of the resists happened while he was casting Holy Fire. I believe it's been shown that mobs in cast animation can't dodge or parry, and can only miss.

I'll most likely be going back in later tonight or tomorrow with a much higher hit rating, to see if we can pin this down. Any suggestions are welcome, as well as anything I may have overlooked about High Priest Venoxis being a special case.

[E] Also, Nite_Moogle, T6 tanking gear is just loaded with hit rating. In my 'Max Stamina' set I have 156 hit rating just off items alone.

Attached Files

taunt test.txt (2.02 MB, 135 views)

 

◊ Penguin

Did a second round of testing. Exactly the same test setup the only difference was that I bumped the hit rating up to 215, 13.63% on the character sheet. I decided to add half the miss rate seen in the previous test, in case anyone's wondering why I chose that particular number. Nothing else significant changed, still 10 expertise value.

Taunted 891 times
Successful 864 times
Resisted 27 times
For a miss rate of 3.03%.

So a hit rating of ~9% produced ~9% miss, and a hit rating of ~13.5% produced ~3% miss, putting the base miss rate of the taunt ability in the ~16.5% to ~18% range. This falls well in line with the known mechanics for spell resists, which have a base miss rate of 17%. Interestingly, melee gains less from +hit rating than casters, making it harder to 'cap' (15.8 rating = 1% hit, rather than the caster 12.6 rating = 1% hit). Further testing will probably show there is the same 99% hit cap as spell hit, meaning that taunt would always have a chance to fail. With such a high base miss chance, I'm willing to speculate that Nalorakk in ZA is a special case for taunt hit.

Once again, the combat log is attached if anyone wants to do more with the data.

Attached Files

taunt test 2.txt (1.38 MB, 60 views)

 

Nite_Moogle

That's quite interesting, it's pretty clear they did muck with the base hit chance. Unfortunately that's still a pretty small sample and we don't know for sure if Venoxis has any special case elements either, so it would be good to do testing on another boss if possible.

There are no physical attacks that have an unavoidable failure chance, so I suspect with sufficient amounts of hit (though it's looking like that's an absurd amount, probably north of 250 rating) you could get taunt unresistable.

 

nfw

Doesn't the game round down? So 13.63% is really 13%?

 

Nite_Moogle

Only for expertise, for other stats you do receive that partial benefit.

 

Sarkan-ZdC

That is very interessting. Thank you for this test!

Just to be sure it is not this boss, possible to test any other boss in Zul?

 

◊ Penguin

I can't seem to let this go for some reason, so did a third round of tests. This time I did my testing on Bloodlord Mandokir to make sure that High Priest Venoxis wasn't a special case. Other than a change in the mob it was the same test setup as last time, with my hit rating at 215, 13.63% on the character sheet.

Taunted 1441 times
Successful 1404 times
Resisted 37 times
For a miss rate of 2.57%.

Combining the two tests done at 215 hit rating gives the overall result of:

Taunted 2332 times
Successful 2268 times
Resisted 64 times
For a miss rate of 2.74%.

This gives an observed base miss rate of 16.37%. Given the sample size and the fact that two separate mobs produced the results I think it's safe to say that taunt follows the same base miss rate as spells do, at least versus level 73 mobs. I see no reason why Blizzard would modify the rest of the table, so it should look like:
vs. 70: 4% (64 rating to cap)
vs. 71: 5% (79 rating to cap)
vs. 72: 6% (95 rating to cap)
vs. 73: 17% (269 rating to cap)

Assuming that is the case, the various anecdotal resists reported from people with high hit ratings on trash mobs (70 - 72) seems to indicate that there is a 1% miss rate that can't be overcome. Testing this for sure would mean gearing up to more than 270 hit rating, and taunting a boss mob until one resist was seen. I may try this some time this weekend, just to finish this up for good.

Once again, the combat log is attached if anyone wants to do more with the data.

Attached Files

taunt test 3.txt (1.82 MB, 73 views)

 

◊ Quigon

Sounds like taunt behaves as it did on 4H - 17% spell miss rate, that now benefits from hit rating.

I'd say the only question is can you get enough hit rating to ensure that it never misses, ala melee, something which you cannot do with spells.

 

Xav

This really makes me wonder if melee hit now applies to all spell hit chances, or if it was just coded to work that way for the taunt-like abilities! I suppose this could be answered pretty easily by an enhancement shaman.

 

◊ Penguin

So, answered the last two questions. I went back to ZG this time to Gahz'ranka, all the other decent bosses being dead this reset. Did the test while wearing 326 hit rating (20.63% on the character sheet). Based off all prior testing this should have been more than enough to be over the cap.

Taunted 345 times
Successful 344 times
Resisted 1 times

Percentage on that test doesn't matter so much, as the single resist above the cap shows that there is a 1% chance to miss that can't be overcome. Taunt follows spell hit mechanics exactly. Also during the test I did a lot of Thunderclaps and Demo Shouts, and the resist rate on those shows no significant change from any other hit rating. So melee +hit only affects taunt, and not other warrior/feral druid spells.

Apologies for hijacking a general thread onto a specific topic.

[E] Combatlog attached.

Attached Files

taunt test 4.txt (825.9 KB, 50 views)

 

◊ Quigon

Yep, 1 resist is an absolute result unless something weird was happening. If it was 100%, it could never resist, ever.
1 in 350 is definitely within statistical bounds.

Thanks Penguin - I'll add this to my own guide.

I think we can assume then that you need 16% to hit to reach cap.