The difference now is that Shadows make up for that high average survivability by having a high variance in spikiness. After 2.5, that spikiness is being leveled off but they still have the best average survivability, which is where it becomes a problem.

The non-OP Warding relic is still BiS (dipstik's calculations were based on the intended behaviour). Right now, it's just super OP and even better. I believe it's absorbing all damage in the 6s window that it is active.

1. Nothing. It used to be proc dmg everywhere and dipstik changed it to warding to be more clear. He probably missed that one. 2. Yes. 3. Pre-mitigation, which you can see for each boss in DF/DP here.

Yessir.

Updated the known attack types, and added more DTPS and Ratios based on the data provided by KBN and insaneric (thanks!) Comparing their fights against each other and my own personal data, it looks like all ratios are within 2% of each other. I'd call that close enough to start making gearing decisions on

You are absolutely right. I'll leave the numbers as they are then.

I didn't pick through your parse thoroughly, but I did notice a couple discrepancies for some attacks like Squash where I calculated a higher pre-mitigation number than yours. Are you using the maximum value for all attacks or are you averaging them out? Also, Dread Charge is M/R. I double-checked my logs and I've parried it before. Not that any of these will make a significant difference...

Awesome, thanks KBN. Bestia ratios added.

The follow-up to that is whether the Conquerer Warding relic is better than the Underworld FR and SA relics. I'm assuming the Obroan one definitely is.

The difference is 39, which should answer your question Although FR does look like its second best for all tanks anyway, so it's probably safe to include it into the max mitigation value.

It's there and listed as BiS, although I think dipstik should probably explain it a bit better. The last number at the end of each relic is the amount of pre-mitigation damage taken per second you would need to receive for that relic to be better than the warding relic. Since most of them have numbers above 7000, it's nearly always BiS or second best.

As a Juggernaut? Chilling Scream and run away. Force Push works too. It's not really that hard to avoid, and it helps save healer resources. Carbonize works fairly well for Powertechs too.

Grob'Thok numbers are up.

Awesome, thanks I guess the only real practical question regarding Blade Barrier is how the 4-piece bonus compares to upgrading to a new level of armoring. That seems like the only real-world example of where there could be potential trade-offs, since in all other situations Strength and Power are just by-products of other gearing decisions. I'm assuming the extra armor value isn't high enough to justify breaking up a set prematurely, but there is likely a way to actually demonstrate it.

This ended up being more of a practice vs theory issue though. Ideally, those formulas would work with a really large sample size, but there were a couple problems I ran into. 1. Some attacks are rare, so lucky (or unlucky) mitigation could skew the post-mitigation DTPS for them pretty heavily. 2. Even if I were to collect data over a number of weeks, most of the mitigation percentages from gear would be changing on a regular (if not fight-to-fight) basis. The math would become a lot more convoluted very quickly, and I'd also need to be storing a lot more data about how my stats change over time than I generally do. So, acknowledging the above issues, I went the with the "shortcut" route. From my analysis, it looked like the damage ranges on most boss attacks are pretty small, so finding the highest unmitigated damage of each attack and treating it like the "worst case" scenario for all unshielded, post K/E damage reduction attacks of that type gives a good enough value for calculating pre-mitigation DTPS.

Posting my methodology for peer review. To determine DTPS: Take the largest value for an unmitigated attack type in the combat log Multiply it by the number of times that attack was used If the attack was K/E, then also divide that number by my K/E damage reduction to find the pre-mitigation damage taken Repeat for all attacks in the fight Add up the totals of each damage type Divide the total for each damage type by the duration of the fight to find the DTPS Calculate the ratio of each damage type to the total DTPS So, using Nefra as an example: The highest damage I received from Twin Attack was 4468 I was hit by Twin Attack 171 times. 171 * 4468 = 764,028 damage taken My K/E damage reduction during the fight was 50.52%. 764,028 / (1 - 0.5052) = 1,544,144.794 damage taken pre-mitigation Voice of the Masters hit me for 1763 damage I was hit 19 times by Voice of the Masters. 19 * 1763 = 33,497 My I/E damage reduction during the fight was 24%. 33,497 / (1 - 0.24) = 44,075 damage taken pre-mitigation The fight took a total of 192.621 seconds M/R K/E DTPS = 1,544,144.794 / 192.621 = 8016.49 F/T I/E DTPS = 44,075 / 192.621 = 228.82 Total DTPS = 8016.49 + 228.82 = 8245.31 M/R K/E Ratio = 8016.49 / 8245.31 = 0.97225 F/T I/E Ratio = 228.82 / 8245.31 = 0.02775 Does this method seem correct? I'm open to suggestions if there is some other way I should be calculating it.

Gate Commander Draxus updated. I removed Mass Affliction and Subteroth explosion damage from the ratio calculations since technically those are avoidable when the fight is done 100% correctly.

What's the method for calculating it in the squish formula?

You just end up using Assault more. Continue to use Enrage, Sunder and Force Choke to build up rage, and use the rest of your abilities as you can (Crushing Blow > Force Scream > Smash > Vicious Slash is generally a good priority for off-tank damage, with Ravage on cooldown). You're going to be more rage-starved, but that's about all you can do. Switching stances isn't super reliable, especially since you lose your rage when you do so.

I'm assuming this has to do with Blade Barrier, but I'm not entirely clear on the math to figure out how to optimally use DTPS to figure out the value of Strength and Power (especially in relation to other defensive stats). I've always just treated those stats more like a bonus.

Dipstik, just to clarify, in your initial post when you wrote that Juggernauts gain 3.13E-5*dps mitigation per second per point of mitigation, that was Damage Taken per Second, right?

Okay, I'll add those as well for the fights I have.

Fair enough. I'll include them in my list since I like calculating my total squishiness. Plus, it's fairly easy to cancel it back out. I figure more information is better than less

Oh, ok. Are any of those attacks from the adds? I've been splitting them up in the list. It will let us create ratios based on role as well.

Also, what are you doing to arrive at your pre-mitigation numbers? Just finding non-shielded attacks and extrapolating using your armor damage reduction?