Jump to content

2.0 Tanking Spreadsheet


Denchet

Recommended Posts

Link to the Spreadsheet.

 

First of all, credit where credit is due:

Much of the design/layout of the spreadsheet is inspired by the original Tanking Spreadsheet by LagunaD. It has unfortunately fallen out of date, so I wouldn't recommend trying to use it.

 

Basic stat formulas all come from this thread over at mmo-mechanics.com.

 

Several of the numbers, such as the percentage of attacks that are MR+KE vs FT+KE, the average value of Shadow's Kinetic Bulwark, the fact that many boss attacks only have 90% accuracy, and many more numbers and facts, come from the fantastic threads posted by KeyboardNinja and Dipstik, as well as many other contributors to those threads. Those threads can be found:

Here for KeyboardNinja's.

Here for Dipstik's.

 

How to use the Spreadsheet:

The first thing you are going to want do do in sign in to your google account (or make a google account). Then click File > Make a Copy. Go ahead and name it whatever you want. When the copy opens, it should now be editable by you.

 

The first thing you want to do is input your stat ratings into row 3. These ratings come from your character sheet. Make sure your tanking stance is on, and you remember to include the defense rating you get from your endurance stim. The stats entered by default are my current stats.

 

The second thing you are going to want to do is fill out your base stats in row 7. The base stats for each class are listed for you at the bottom left corner of the spreadsheet. These base stats assume an optimal talent build, as well as several in combat buffs/debuffs that are described in detail lower in this thread. The values entered by default are for a Shadow tank. For the acc. and damage debuffs, use the default 0.05 if your raid comp has these debuffs, use 0 if your raid comp does not have these debuffs. Acc. debuff comes from Shadow tank and Guardian tank, damage debuff comes from Shadow Tank, Vanguard tank, and possibly Watchman Sentinel if they spec into it.

 

The only other number you should ever change is A14, which is 0.79. This number is the percentage of attacks (so 79%) that bosses use on your that are melee/ranged attacks, ie dodgeable attacks. This number is important. If you assume all attacks are dodgeable you would incorrectly over inflate the value of defense. 79% is the average of all bosses in HMSV. This number will very likely change in future content, so be sure to come back here to find out what it changes to in the future. For now though, leave it at 0.79.

 

And that's it! That's all the numbers that you should have to enter. You should now see the percentage of enemy damage you take in cell A14. More importantly, you should see your stat weights for your current gear in row 27.

 

The goal of tanking is to get as many mitigation stats are you reasonably can (without sacrificing too much endurance). You then want to balance those mitigation stats properly, which is where row 27 comes in. You are optimized when these numbers are equal. Whenever you add a stat in row three, the stat weight for that stat should go down. For example, lets say you increased your defense rating by 100 in cell B3. The value of B27 should go down. You want to fiddle with the stats in row 3 until you get the values in row 27 as equal as possible. Note that for Shadows and Guardians, you will never be able to get your shield rating low enough (at last at current gear levels), so the stat weight for shield is always going to be lower than the stat weights for defense and absorb. That's just how the itemization works in this game, and there is nothing you can do about it.

 

Where the base stats come from:

All tanks get defense +5% from the fact that enemy attacks can either have 90% accuracy or 100% accuracy. A 50/50 split is assume, thus 95% accuracy, thus the 5% defense bonus.

 

Shadow:

Damage reduction: 0.02 from Jedi Resistance + 0.02 from 4-piece = 0.04

Defense: Base of 0.10 + Enemy accuracy bonus of 0.05 + Double-bladed Saber defense 0.04 + Shadowsight 0.02 = 0.21

Shield: 0.15 from tanking stance + 0.05 from equipping a shield + 0.20 from Kinetic Ward with 2-piece set bonus = 0.4

Absorb: 0.2 from equipping a shield + 0.04 from Impact Control + 0.036 from Kintetic Bulwark (on average) = 0.276

Force/tech resist: 0.02 from Shadowsight.

 

Guardian:

Damage reduction: 0.06 from tanking stance + 0.03 from Guardian Slash = 0.09

Defense: Base of 0.05 + Enemy accuracy bonus of 0.05 + 0.05 from Blade Barricade + 0.03 from Single Saber Mastery = 0.18

Shield: 0.15 from tanking stance + 0.05 from equipping a shield + 0.04 from Shield Specialization = 0.24

Absorb: 0.2 from equipping a shield

Force/tech resist: 0.05 from Blade Barricade

 

Vanguard:

Damage reduction: 0.05 from tanking stance + 0.02 from Ion Shield + 0.02 from Power armor = 0.09

Defense: Base of 0.05 + Enemy accuracy bonus of 0.05 + Deflective Guards gives 0.04 + 4-piece set bonus gives 0.02 = 0.16

Shield: 0.15 from tanking stance + 0.05 from equipping a shield + 0.02 from Shield Cycler = 0.22

Absorb: 0.2 from equipping a shield + 0.04 from Ceramic Plating + 0.15 from Power Screen/Energy Blast (on average) = 0.39

Force/tech resist: 0.02 from 4-piece set bonus

 

How all the math works:

The equations used to find row 11 just plug your stat ratings into the stat formulas, then add your base stats on top.

 

A17, your damage taken, comes from the equation:

 

(MR * (1-D-AD) * (1-S*A) (1-DR) * (1-DD)) + ((1-MR) * (1-S*A) * (1-FTR) * (1-DR) * (1-DD))

 

MR=% of attacks that are melee/ranged

D=Defense chance

S=Shield chance

A=Absorb %

DR=Damage Reduction

FTR=Force/tech resist

AD=Acc. Debuff

DD=Damage Debuff

 

Row 21 tells you what your stats would be if you added one rating to that stat. Its basically "your stats +1" plugged into the stat equations, plus base stats.

 

Row 24 is what your total damage taken would be with 1 more stat rating for the corresponding stat. It uses the same equation at A17, but uses the respective value from row 21 instead of row 11.

 

Row 27 (the stat weights) come from the difference between row 24 and A17, aka how much more total mitigation you get by adding one to each particular stat. This value is then multiplied by 100,000 so that it spits out easy to read numbers.

 

One final note. The value A27, the stat weight for armor, is fairly worthless information because it doesn't take into account any armor multipliers you have from talents and your tanking stance. This doesn't really matter though as you get armor naturally, and you can't trade armor for any of the other tanking stats anyway.

 

Updates:

5/20/13

Acc/Dmg debuffs have been removed from base stats and given their own separate cells. Should be pretty self explanatory, and there is a brief explanation in the spreadsheet.

 

Dmg debuff is now treated as a separate effect that multiplies boss damage by 0.95*, rather than just adding 0.05 to your damage reduction.

 

*Technically its multiplied by (1-DmgDebuff), which is (1-0.05)=0.95 by default.

Edited by Denchet
Link to comment
Share on other sites

Excellent work, I've been hoping one of these would get made soon :)

 

I just notice 1 very minor issue. You're calculation with the 5% damage debuff from Slow Time/Static Field has it being added to static DR whereas it actually reduces pre-mitigation damage dealt which is then reduced by static DR. The difference is pretty minor but adjusting all the base DRs down by 0.05 and adding a 0.95 scalar in the damage taken equations gives a more accurate representation.

 

Downside is when you don't have that debuff it requires some manual alterations to the spreadsheet. I might play around with it while I'm at work and add in some more tools such as auto calculating in the the class bonuses and a flag to factor in/out the 2 debuffs.

 

EDIT: I made some adjustments including adding a little functionality I really liked in the old spreadsheet such as the effect of adding/swapping augments. As well as adding in the class base stat auto-fill and both debuff flags with autofill based on class/co-tank.

 

https://docs.google.com/spreadsheet/ccc?key=0AmWVSdivhbgLdGdSZFRyTkE0MTJRTmRaUS1IRlAyZEE&usp=sharing

Edited by grallmate
Link to comment
Share on other sites

Great work! Would it be possible to also calculate the Spikiness Damage Taken (SigmaSquishiness/Spikiness on LagunaD's)?

 

If I remember correctly, SigmaSquishiness/Spikeiness was help full to check for the spikeness of the damage taken for the chosen stats.

Edited by leto_cleon
Link to comment
Share on other sites

One final note. The value A27, the stat weight for armor, is fairly worthless information because it doesn't take into account any armor multipliers you have from talents and your tanking stance. This doesn't really matter though as you get armor naturally, and you can't trade armor for any of the other tanking stats anyway.

I have to quibble a little with this point, because I, and probably many others, don't have any Underworld set armorings, whereas we can buy Verpine armorings. Actually, when I first hit 55, I only had Campaign armorings, and as Kitru and others have said elsewhere, the benefits of the set bonus outweighs the benefit of the 69/72 armorings. Even now, having the 4-piece Arkanian bonus, that's still 4 Verpine armorings I can't use.

Link to comment
Share on other sites

Great work! Would it be possible to also calculate the Spikiness Damage Taken (SigmaSquishiness/Spikiness on LagunaD's)?

 

If I remember correctly, SigmaSquishiness/Spikeiness was help full to check for the spikeness of the damage taken for the chosen stats.

 

This may not be as directly useful as before, since boss "big hits" now seem to be broken up into smaller chunks with a chance for Defense to avoid each sub-piece, thereby smoothing the damage out more -- so a more complex method (or a simulator) may be needed.

Link to comment
Share on other sites

This may not be as directly useful as before, since boss "big hits" now seem to be broken up into smaller chunks with a chance for Defense to avoid each sub-piece, thereby smoothing the damage out more -- so a more complex method (or a simulator) may be needed.

 

It depends. There are still quite a few big hits like Terminate that land on tanks in one, massive, discreet chunk. And spikiness isn't just a measure of how variable damage is across a single attack. I'm doing some of my own math to calculate spikiness by determining the chance weighted variance of the comparative mitigation factors of the various mitigation mechanisms for each tank and, no matter how you figure it, Shadows are *stupidly* spiky at the moment: you have a ~34% chance to take ~45% more damage than a VG/Guard from an individual attack (the difference is even worse for F/T attacks where a Shadows higher Defense rating isn't a factor and Shadows end up taking more damage in *every* scenario than Guard/VGs because Shield/Abs aren't high enough to offset the *massive* DR advantage that Guard/VGs get). A 34% chance for 45% more damage, even with damage distributed across multiple attacks, is still going to punch through your mitigation often enough for it to be a problem.

Link to comment
Share on other sites

Great work! Would it be possible to also calculate the Spikiness Damage Taken (SigmaSquishiness/Spikiness on LagunaD's)?

 

If I remember correctly, SigmaSquishiness/Spikeiness was help full to check for the spikeness of the damage taken for the chosen stats.

 

I honestly never gave those values much thought. I always completely ignored them when using that spreadsheet, and just focused on minimizing squishiness. I would also have no way of knowing how one should properly balance squishiness and spikiness, so I'm not sure how to give spikiness any real meaning.

 

I have to quibble a little with this point, because I, and probably many others, don't have any Underworld set armorings, whereas we can buy Verpine armorings. Actually, when I first hit 55, I only had Campaign armorings, and as Kitru and others have said elsewhere, the benefits of the set bonus outweighs the benefit of the 69/72 armorings. Even now, having the 4-piece Arkanian bonus, that's still 4 Verpine armorings I can't use.

 

You could just equip your higher level armor and see if your damage taken % on the character sheet goes up or down when losing the 4-piece set bonus pretty easily for Shadow tanks. For VGs its a bit trickier, you would have to equip the higher level armors and see what your armor rating becomes. Then plug that into the spreadsheet while lowering base defense chance by 0.02, and see if that gives a lower DT than just using the old armorings with the 4-piece. I guess you could do the same thing for the Shadow 2-piece. You could also test the value of using 2x 2-piece for Shadow that way. For Guardians I have no idea what you would do.

 

Excellent work, I've been hoping one of these would get made soon :)

 

I just notice 1 very minor issue. You're calculation with the 5% damage debuff from Slow Time/Static Field has it being added to static DR whereas it actually reduces pre-mitigation damage dealt which is then reduced by static DR. The difference is pretty minor but adjusting all the base DRs down by 0.05 and adding a 0.95 scalar in the damage taken equations gives a more accurate representation.

 

Downside is when you don't have that debuff it requires some manual alterations to the spreadsheet. I might play around with it while I'm at work and add in some more tools such as auto calculating in the the class bonuses and a flag to factor in/out the 2 debuffs.

 

EDIT: I made some adjustments including adding a little functionality I really liked in the old spreadsheet such as the effect of adding/swapping augments. As well as adding in the class base stat auto-fill and both debuff flags with autofill based on class/co-tank.

 

https://docs.google.com/spreadsheet/ccc?key=0AmWVSdivhbgLdGdSZFRyTkE0MTJRTmRaUS1IRlAyZEE&usp=sharing

 

Are we absolutely sure that's how the DR debuff works? I asked about it in another thread because I wasn't sure, and was told it just adds to your DR. If its multiplicative instead of additive that's a pretty easy thing to change. Same thing with the accuracy debuff, do we know for sure whether or not that just adds to dodge chance? I guess the same also goes with the innate 90% accuracy skills enemies have.

 

Its difficult to test in game because it makes such a small difference.

Link to comment
Share on other sites

Are we absolutely sure that's how the DR debuff works? I asked about it in another thread because I wasn't sure, and was told it just adds to your DR. If its multiplicative instead of additive that's a pretty easy thing to change. Same thing with the accuracy debuff, do we know for sure whether or not that just adds to dodge chance? I guess the same also goes with the innate 90% accuracy skills enemies have.

 

The damage debuff is multiplicative because damage reduction is applied locally to the target whereas damage dealt is applied locally by the attacker. The accuracy debuff is additive because the Defense chance is factored in by directly reducing the target's accuracy. It was tested way back when at release.

Link to comment
Share on other sites

I honestly never gave those values much thought. I always completely ignored them when using that spreadsheet, and just focused on minimizing squishiness. I would also have no way of knowing how one should properly balance squishiness and spikiness, so I'm not sure how to give spikiness any real meaning.

 

If I recall correctly, high stacking on defence rating lowered squishiness much more than stacking shield/absorb. The problem however was high defence was more spikey due to hits not hitting, and hitting. The low shield or low absorb meant low mitigation on the hits. Bad defence rolls and spike damage are more difficult to heal. To smoothen this out, points could be shifted from defence to shield/absorb (this was on my Guardian).

 

So just for illustration some pre-2.0 stats,

 

356 Def, 569 Sh, 410 Ab (0.2846 Squishiness, 0.22393 Sigma Squishiness)

On use Def Relic: 671 Def, 569 Sh, 410 Ab (0.26011 Squishiness, 0.22848 Sigma Squishiness)

Link to comment
Share on other sites

The damage debuff is multiplicative because damage reduction is applied locally to the target whereas damage dealt is applied locally by the attacker. The accuracy debuff is additive because the Defense chance is factored in by directly reducing the target's accuracy. It was tested way back when at release.

 

Alright, that should be easy enough to fix. I'll go through and fix that and also make the two debuffs separate values that you can "toggle" on and off depending on group comp rather than putting them in base stats.

If I recall correctly, high stacking on defence rating lowered squishiness much more than stacking shield/absorb. The problem however was high defence was more spikey due to hits not hitting, and hitting. The low shield or low absorb meant low mitigation on the hits. Bad defence rolls and spike damage are more difficult to heal. To smoothen this out, points could be shifted from defence to shield/absorb (this was on my Guardian).

 

So just for illustration some pre-2.0 stats,

 

356 Def, 569 Sh, 410 Ab (0.2846 Squishiness, 0.22393 Sigma Squishiness)

On use Def Relic: 671 Def, 569 Sh, 410 Ab (0.26011 Squishiness, 0.22848 Sigma Squishiness)

 

But then how do we use spikiness with squishiness in order to maximize our overall tankyness? Is it worth increasing squishiness in order to decrease spikiness? I get what the stat does, I just don't know how to apply any value to it.

Edited by Denchet
Link to comment
Share on other sites

But then how do we use spikiness with squishiness in order to maximize our overall tankyness? Is it worth increasing squishiness in order to decrease spikiness? I get what the stat does, I just don't know how to apply any value to it.

 

In general, you want to get spikiness below a certain threshold depending upon your healer's skill level and then optimize survivability while maintaining the threshold.

Link to comment
Share on other sites

But then how do we use spikiness with squishiness in order to maximize our overall tankyness? Is it worth increasing squishiness in order to decrease spikiness? I get what the stat does, I just don't know how to apply any value to it.

 

LagunaD explained SigmaSquishiness and Spikiness as:

 

"SigmaSquish" is the standard deviation of your squishiness on an attack-by-attack basis. Smaller values are better from the point of view of taking predictable damage. For a (hypothetical) pure-mitigation tank, SigmaSquish would reflect only the attacker's chance for critical hits.

 

A related value, called "Spikey-ness" is the ratio of SigmaSquish for your character in armor to a naked character with no avoidance or mitigation. This normalizes out the attacker's critical hits. For a pure mitigation tank, Spikey-ness would equal Squishiness. For a real tank, it will be higher. Smaller values are better.

 

Personally I used, Squishness/SigmaSquish to compare between different stats to see which was less squishy, and which had a smaller variance. So for example, when comparing the two Dread Guard Relics on a Guardian tank,

 

DG On Use Def Relic: 671 Def, 569 Sh, 410 Ab (0.26011 Squishiness, 0.22848 Sigma Squishiness)

DG On Use Sh/Ab Relic: 356 Def, 759 Sh, 600 Ab (0.26249 Squishiness, 0.22264 Sigma Squishiness)

 

-315 Def, +190 Sh, +190 Ab = -0.238% Squishinesss, +0.584% less variance (315 def vs 380 Sh/Ab)

 

Which is relic is better? Depends if you prefer to be less squishy or have lesser damage taken variance.

Edited by leto_cleon
Link to comment
Share on other sites

Alright, updated a bit.

 

Acc/Dmg debuffs have been removed from base stats and given their own separate cells. Should be pretty self explanatory, and there is a brief explanation in the spreadsheet.

 

Dmg debuff is now treated as a separate effect that multiplies boss damage by 0.95*, rather than just adding 0.05 to your damage reduction.

 

*Technically its multiplied by (1-DmgDebuff), which is (1-0.05)=0.95 by default.

 

Link to new version

 

Will update the OP tonight or tomorrow.

Link to comment
Share on other sites

You could just equip your higher level armor and see if your damage taken % on the character sheet goes up or down when losing the 4-piece set bonus pretty easily for Shadow tanks. For VGs its a bit trickier, you would have to equip the higher level armors and see what your armor rating becomes. Then plug that into the spreadsheet while lowering base defense chance by 0.02, and see if that gives a lower DT than just using the old armorings with the 4-piece.

 

Rather than actually buying and ripping armorings from Verpine pieces, I have assumed that AskMrRobot has the right armor ratings. Using my AMR profile, I get:

 

4 x arkanian => 6222 armor, Damage Taken = 0.33859

 

I then saved that as another profile, and swapped the 4 Arkanian armorings for Force Wielder 31's.

Using that armor rating, I amended the spreadsheet constants: DR base reduced from 0.04 to 0.02, shield base reduced from .40 to .35.

 

swap to Verp => 6370 armor, Damage Taken = 0.35751

 

-- i.e. 1.892 percent more damage taken with the 4 Verpine armorings vs. 4 Arkanian. (28 more Endurance => roughly 290 more HP, and 28 more Willpower).

 

I haven't run the numbers on the other possible combinations (combinations of Campaign and Dread Guard armorings with various numbers of Arkanian and Underworld are all possible) but at least this proves to my own satisfaction that the DR gain outweighs the armor rating, willpower and HP loss :)

Edited by Ancaglon
Link to comment
Share on other sites

×
×
  • Create New...