Stat Weights

[dspguy]

Let's pretend Featherfoot was like parry. Would you use featherfoot when one mob is hitting you? Or 10? More chances to proc dodge when there are 10 mobs.

Now, is Featherfoot just like parry? Not exactly. However, if fighting one mob, you may wind up outside the bell curve for your parry rate (maybe you parry 9 of 10. Maybe you parry 0/10). But, if you have 10 mobs on you, you are very likely to hit that 25% parry rate.

And I don't think parry leads to overheals when it matters. Any time that your HP drops below what a single heal would heal you to full for, parry helps. Sure, if you have 9000/10000hp and parry blocked 250dmg and your healer was going to heal you anyway... then no. Parry does not help. But if you take 3000 damage and blocked 750 of it and you are only healed for 1500, then you are 750hp ahead of the game.

[MythToken]

Be careful how you word that, just because more mobs are hitting you at the same time, does not mean you are parrying more often. It just means you are seeing the effects of parry in a small concentrated time frame.
And while this indeed will probably help you stay alive, due to the fact you are taking a lot of damage, your parry rate will not increase just because more things are hitting you.


Parry is slightly better for warriors this is a fact.

Parry isn't all that great in general this is a fact.

[Giantbane]

As has been said, parry is only really good when you're getting hit a lot of times since it gives you a lot of chances to parry and its effect will be more dependable (law of large numbers and all that). And you will usually need a lot of consistent healing in those situations so it will have more of an effect of reducing the load on the healer.

Basically it's only really good for speed runs.

[MythToken]

No not really, you parry the same % no matter how many mobs are hitting you. The effects just seem more noticeable because in that short time frame you are incurring more attacks.

Parry is as good against 10 mobs as it is against 1 mob. It mitigates the same %.

Higher mitigation when 10 things are hitting you at once is def more helpful. But again parry's usefullness is still exactly the same, because you parry for the exact same %.

If your defense rose by 10 you would not claim that to be more useful on 10 mobs than on 1 mob. It still mitigates the same %.

The fact is that more mitigation parry or otherwise is very good for keeping you alive during 10 mob attack,due to the damage you are sustaining in a short time frame and the risk of death.
But the parry stat is not suddenly more or less useful in and of itself.

[SirTaint]

The larger the data sample the more accurate the percentage.

If you have 20% parry rate with one mob hitting you over the course of a minute the RNG will have a broad range.

If you have 10 mobs hitting you the range will be more narrow and closer to the 20% mark.

The more predictable parry is the more useful it is since it's biggest flaw is how unreliable it is.

(This still doesn't make it good but outlines why it is more useful during a duration with more mobs)

[Dyne_Fellpool]

SirTaint, in your example the actual parry rate is just as likely to be above the expected 20% than below it.
Yes a larger sample size will see the parry rate come closer to the expected amount, but that is not inherently a more favorable number.

No not really, you parry the same % no matter how many mobs are hitting you. The effects just seem more noticeable because in that short time frame you are incurring more attacks.

Parry is as good against 10 mobs as it is against 1 mob. It mitigates the same %.

Higher mitigation when 10 things are hitting you at once is def more helpful. But again parry's usefullness is still exactly the same, because you parry for the exact same %.

If your defense rose by 10 you would not claim that to be more useful on 10 mobs than on 1 mob. It still mitigates the same %.

The fact is that more mitigation parry or otherwise is very good for keeping you alive during 10 mob attack,due to the damage you are sustaining in a short time frame and the risk of death.
But the parry stat is not suddenly more or less useful in and of itself.

MythToken got it right.

[Giantbane]

No not really, you parry the same % no matter how many mobs are hitting you. The effects just seem more noticeable because in that short time frame you are incurring more attacks.

Parry is as good against 10 mobs as it is against 1 mob. It mitigates the same %.

Higher mitigation when 10 things are hitting you at once is def more helpful. But again parry's usefullness is still exactly the same, because you parry for the exact same %.

If your defense rose by 10 you would not claim that to be more useful on 10 mobs than on 1 mob. It still mitigates the same %.

The fact is that more mitigation parry or otherwise is very good for keeping you alive during 10 mob attack,due to the damage you are sustaining in a short time frame and the risk of death.
But the parry stat is not suddenly more or less useful in and of itself.

The larger the data sample the more accurate the percentage.

If you have 20% parry rate with one mob hitting you over the course of a minute the RNG will have a broad range.

If you have 10 mobs hitting you the range will be more narrow and closer to the 20% mark.

The more predictable parry is the more useful it is since it's biggest flaw is how unreliable it is.

(This still doesn't make it good but outlines why it is more useful during a duration with more mobs)

^ this

It's not about the %, it's about the reliability of it. When you're getting hit a lot, you can rely on it to mitigate a roughly consistent amount of damage most of the time. If you don't get hit very often, its performance can vary wildly. This means in a low frequency setting against 1 mob, you and your healers can't rely on it and will have to heal/use cooldowns with the expectation that you won't parry those big hits where it matters.

[Lothar808]

While PLD benefit less from Parry... Parry I believe actually puts paladin's mitigation higher than a War. Total mitigation for WAR at 600 parry and 25% reduction is roughly 12% overall mitigation. For PLD with the same stats it is rough 8%, however our shield (Novus/High Allagan) gives us roughly 9%, combine the two for a Total mitigation of ~17%... So while we benefit less from Parry overall, when stacked with shield blocking, it is still very useful for full mitigation builds.

Here is the math:

1000 hits for 100 damage = 100,000 total damage
600 Parry = 46% change to parry (600/13)
at ilvl 110 and all vit pieces = 25% reduction on parried attacks.
Novus/High Allagan = 31% block rate, 31% block chance

So Block has a chance to happen first, if it does no parry can happen.

For PLD:
1000 attacks at 100 damage = 100,000
310 attacks blocked at 31 damage reduced = 9610 damage reduced (roughly 9-10% mitigation)
690 attacks left at 100 damage = 69,000 damage left
317 attacks parried (690x46%) at 25 damage reduced = 7925 damage reduced (roughly 7-8% mitigation)

For WAR
1000 attacks at 100 damage = 100,000
460 attacks parried (1000x46%) at 25 damage reduced = 11500 damage reduced (roughly 11-12% mitigation)

[MythToken]

Nice Lothar! interesting stuff

I think back to a few other points about the reliability of parry and featherfoot. They arn't something you should be relying on tbh. They are not activated skills that you should be popping for large hits. (well featherfoot is active, but just pop it a ton)

Thier main purpose is to mitigate and the idea that a single mob doesn't attack very often is flawed as well. Auto attacks and queued skills from a boss can and do occur very often and in rapid succession.
You won't even see it normally unless you die from it. But this is where parry is useful, and mitigates damage.

I think a better word for parry would be "luck" because yes, the value you actually parry on a particular fight can vary wildly. And I have out parried a tank that has 100 more parry than me. It happens.

Parry should never be viewed as reliable mitigation for large attacks. It's great when it happens but its not what it is designed for. Parry is still useful for single mob tanking. It just isn't the best in the world.

I think having an overall understanding of parry and why you should or should not have it is helpful to this discussion. Everyone's input here has hit on some key points about parry and its reliability or lack there of.

I just don't think that parry was ever intended to be reliable. But that doesn't necessarily make it bad either.