Paladin 2.4 BiS?

**SirTaint** · 10-30-2014 11:06 PM

13-14 per 1% can't be correct. At 646 parry my rate was 17%. Even considering using HA shield that's no where near 13-14 per 1%.

**Kenji1134** · 10-30-2014 11:13 PM

Originally Posted by SirTaint

13-14 per 1% can't be correct. At 646 parry my rate was 17%. Even considering using HA shield that's no where near 13-14 per 1%.

Paladin is not a good candidate to perform testing on parry returns, since blocking takes priority and will artificially deflate the parry numbers.
Lets assume that Paladin and Warrior get the same % chance returns for a given amount of Parry. Since Warrior can only parry, they will see the full rate.

That said, the base value for parry, offhand, is 341 at lvl 50 for a tank. So (646-341)/13.5 = 22.6% chance above the base value, whatever that is.
However your observed 17% is really attacks that were not blocked, but were parried. You would have also seen some attacks which were blocked, but could have "potentially" been parried had the block failed... Warrior does not have this extra probability to deal with.
---
Did some testing on War.
523 parry w/ gear. Base parry at 50 is 341.
Out of 434 attacks that landed, 97 were parried, which is 22.6%
If we go by an average of 13.5 parry for 1%, given my 22.6% observed rate, the base value should be about 9.1%, um, I'll test that in a bit.
---
And MORE data!
At 341 parry, so no parry on any gear, vs lvl 5 Yarzon Feeders... cause they were around...
472 hits, 46 parried, which gives us a rate of 9.746%, pretty close to what I calculated before, though I guess it may be closer to 14 points for 1%, rather than 13.5 or so... Anyway, the math is good enough.

That said, with NO parry on your gear, you have a ~9.7% chance to parry for -25% (assuming normal STR gear), which is 2.43% mitigation in the long run.
With my 523 parry, I saw a 22.6% chance, so that's 5.65% mitigation.
Lets say I have 700 parry, off a 341 base, at a rate of 1% chance for every 14 points, and reduce incoming by 25%... That would be 9.7%+(700-341)/14% = 35.34% chance, and 25% reduction on parry gives us 8.84% mitigation.

So NO parry is a baseline of 2.43%, a TON of parry is 8.84%... That is an effective gain of... 1-(1-.0884)/(1-.0243) = 6.57% less incoming melee damage!

Good, bad, worth it?
For tanks, especially Paladins if you consider Flash's inability to crit, the next best thing after parry would most likely be Det, then Crit. So if we drop ALL the parry gear and get Det and Crit instead, focusing on Det where possible, will that increase our damage by more than 6.57%? To at least "break" even in terms of damage gained vs mitigation lost.

#MathMan(MathTank)HathSpoken!

**MythToken** · 10-30-2014 11:16 PM

I think your base value doesn't make sense.

341/13.5 = 25%

which if we assume thats true you would see 1/4 of all attacks parried as a warrior. And increasing to 682 would take you to a theoretical 1/2 of all attacks parried.

Which would probably be the max parry you should ever shoot for, unless you want to go for 1023 parry and 3/4 of all attacks are parried.

Warrior only

**tremor24** · 11-03-2014 11:00 AM

Originally Posted by MythToken

I think your base value doesn't make sense.

341/13.5 = 25%

which if we assume thats true you would see 1/4 of all attacks parried as a warrior. And increasing to 682 would take you to a theoretical 1/2 of all attacks parried.

Which would probably be the max parry you should ever shoot for, unless you want to go for 1023 parry and 3/4 of all attacks are parried.

Warrior only

The base value of 13.5 points of parry per percent of parry rate does not imply that at 341 parry will equal 25% though it may seem that way at first glance. Since 341 parry is the "base" amount at lvl50 then values below this are not relevant, but if you work the slope back to 0 parry you will find that the % of attacks parried at 0 parry is actually about -14.3% (of course not valid below 341). Since it is not 0 you cannot ignore the intercept in your equation. This is implied in Kenji1134's values but not shown explicitly.

In other words, you can get a much more accurate representation of the scaling with: (%) Attacks Parried = .0706 * ParryValue - 14.33

You can plug in the parry values Kenji1134 has provided and find his results of: 341 parry gives 9.75% attacks parried, 523 parry gives 22.6% attacks parried, and with 682 parry (assuming the scaling holds) gives 33.8% attacks parried. These are for a Warrior or as a percentage of a Paladin's unblocked attacks.

Also, another thing to note is that as strength increases then the amount mitigated by parry also increases (at 445 strength the amount parried becomes 26% instead of 25%), and the value of parry will increase at each strength tier due to this. The difference between max and min parry can actually make a difference and maybe even more so in the future, but depends on the fight as many have noted. I'm not saying just stack parry always, but saying another secondary stat is always better is simply not true for all situations. The extremely small amount of damage a main tank is doing at critical dps periods may make just about as much difference as the amount mitigated by parry at key moments, and a lot more attacks can actually be parried than many realize.

**SirTaint** · 11-08-2014 01:12 AM

So I've been parsing my hybrid build. (i90 penta melds, Wootz penta melded) Puts me at 590 parry. My block rate and parry rate vary between 20-21%.

That fits into the ~13.5 parry per %. 590-341 = 249/13.5 = 18.4=9.75 = 28.15 80*28.15 = 22.52

I did about 10 t9s using the build and some attacks can't be parried so that supports the margin of error.

Going to use a different parry amount the next time we farm.

Thread: Paladin 2.4 BiS?

Thread Tools

Search Thread

Display

Hybrid View

Tags for this Thread