raid mobs and accuracy

Discussion in 'The Veterans' Lounge' started by josh, May 31, 2018.

  1. josh Augur

    I've made a couple posts in threads that i'd like to compile into one thread about accuracy and avoidance and get peoples opinions on. It's gonna be a long one so if you are interested, strap in.

    Current raid mobs have relatively low accuracy. This seems to be an intentional design choice by the devs to compensate for the fact that they were also given 100% strikethrough for dodge/parry/riposte/block skill checks.

    We don't defend against raid mobs, this does not mean that those skills don't activate, you still riposte and do counterattack damage, the mob just hits you anyway, and any skills dependent on those abilities firing, like blessed gaurdian, still trigger.

    We do avoid damage with our avoidance AC. The way miss chance is calculated is fairly straightforward, i haven't figured out exactly what it is but i have worked out a simple way to approximate it. It's likely missing some factors and it isn't the correct way to do this but it works. here's what i worked out.

    Here's a pvp parse of me against a bot with 1065 avoidance AC vs my 1066 accuracy

    Total Melee --- Damage: 1637067 --- Avg hit: 338 --- Attempts: 9536 --- Missed: 4686/9536 [49.14%] --- Hits: 4850/9536 [50.86%] --- Real Hits: 4850/4850 [100%]

    a slight boost to 1263 avoidance, no change to accuracy

    Total Melee --- Damage: 1126974 --- Avg hit: 307 --- Attempts: 8717 --- Missed: 5052/8717 [57.96%] --- Hits: 3665/8717 [42.04%] --- Real Hits: 3665/3665 [100%]

    big boost to 2138 avoidance

    Total Melee --- Damage: 626009 --- Avg hit: 276 --- Attempts: 8940 --- Missed: 6672/8940 [74.63%] --- Hits: 2268/8940 [25.37%] --- Real Hits: 2268/2268 [100%]

    this follows a fairly simple formula, 50 * accuracy/avoidance. so 50 * 1066/1263 = 42 For this to actually work though you would need to have it be the lowest of the two divided by the largest. so if i had 2000 accuracy and you had 1000 avoidance it would be 50 * 1000/2000. There is definitely some truncation occurring as well so this simple formula can give some inaccurate results in some cases. but its good enough for this post.

    so, i decided to test this idea with a real mob. i picked a functionary in plane of life because it would be the easiest recent mob to go afk on for a long parse. Here's my initial parse with 3600 avoidance

    Punch --- Damage: 20137763 --- Avg hit: 4512 --- Attempts: 6639 --- Missed: 2176/6639 [32.78%] --- Hits: 4463/6639 [67.22%] --- Real Hits: 4463/4463 [100%]

    using the formula 50 * 3600/x = 32.78 solving for x gives us an accuracy on that mob of 5500 maybe as low as 5300 depending how/when/if things are truncated.

    So, to try and confirm this number i debuffed the mob with somnolence and did another parse. for anyone unaware, somnolence reduces accuracy by the stated percentage, so with max somnolence of 24% if you had 1000 accuracy you would be dropped down to 760. Easily tested by casting it on yourself and seeing how it affects your accuracy. and the following parse supports that this is also the way it works when cast on npc's. same mob as last time, just now with somnolence the whole fight.

    Punch --- Damage: 16328483 --- Avg hit: 4494 --- Attempts: 6336 --- Missed: 2703/6336 [42.66%] --- Hits: 3633/6336 [57.34%] --- Real Hits: 3633/3633 [100%]

    50 * 3600/(5500 * .76) = 43. so, it's pretty close. I decided to test this against a raid mob, so here's the ghost of the magma lord with and without somnolence.

    Total Melee --- Damage: 25168119 --- Avg hit: 12019 --- Attempts: 3005 --- Riposted: 11/3005 [0.37%] --- Parried: 6/2994 [0.2%] --- Dodged: 10/2988 [0.33%] --- Blocked: 12/2978 [0.4%] --- Defended: 39/3005 [1.3%] --- Missed: 776/2966 [26.16%] --- Hits: 2190/2966 [73.84%] --- Absorbed: 96/2190 [4.38%] --- Real Hits: 2094/2190 [95.62%]

    I defend a couple times in this parse because i used shield flash at some point. they can be ignored as they don't have any real impact on your miss chance since gamparse calculates your miss % on the hits that make it to the miss check. 2966 of 3005 attempts made it to miss.

    We can calculate this guys accuracy to be somewhere around 7k

    This next parse is the same mob except with 24% somnolence.

    Total Melee --- Damage: 35640703 --- Avg hit: 12110 --- Attempts: 4476 --- Riposted: 10/4476 [0.22%] --- Parried: 8/4466 [0.18%] --- Dodged: 15/4458 [0.34%] --- Blocked: 7/4443 [0.16%] --- Defended: 40/4476 [0.89%] --- Missed: 1461/4436 [32.94%] --- Hits: 2975/4436 [67.06%] --- Absorbed: 32/2975 [1.08%] --- Real Hits: 2943/2975 [98.92%]

    again 50 * 3600/(7000 * .76) = 33.83. It's not too far off, however, calculating this guys accuracy is further complicated by the fact that he is dual wielding and using 2 different skills. presumably 1hb and hand to hand. not to mention kicks and bashes. Here's the individual breakdowns for both skills.

    Crush --- Damage: 19465709 --- Avg hit: 12197 --- Attempts: 2410 --- Riposted: 6/2410 [0.25%] --- Parried: 4/2404 [0.17%] --- Dodged: 9/2400 [0.38%] --- Blocked: 3/2391 [0.13%] --- Defended: 22/2410 [0.91%] --- Missed: 778/2388 [32.58%] --- Hits: 1610/2388 [67.42%] --- Absorbed: 14/1610 [0.87%] --- Real Hits: 1596/1610 [99.13%]

    Hit --- Damage: 15228157 --- Avg hit: 12993 --- Attempts: 1798 --- Riposted: 4/1798 [0.22%] --- Parried: 4/1794 [0.22%] --- Dodged: 4/1790 [0.22%] --- Blocked: 3/1786 [0.17%] --- Defended: 15/1798 [0.83%] --- Missed: 607/1783 [34.04%] --- Hits: 1176/1783 [65.96%] --- Absorbed: 4/1176 [0.34%] --- Real Hits: 1172/1176 [99.66%]

    He could potentially have different accuracy for each skill.

    Now, back to my original point that raid mobs have relatively low accuracy, here is a parse against a bunch of trash mobs in howling stones, i had the same avoidance AC as all the other parses and no accuracy decreasing debuffs were used.

    --- Total Melee --- Damage: 41638954 --- Avg hit: 11762 --- Attempts: 5229 --- Riposted: 201/5229 [3.84%] --- Parried: 192/5028 [3.82%] --- Dodged: 326/4836 [6.74%] --- Blocked: 33/4510 [0.73%] --- Defended: 752/5229 [14.38%] --- Missed: 900/4477 [20.1%] --- Hits: 3577/4477 [79.9%] --- Absorbed: 37/3577 [1.03%] --- Real Hits: 3540/3577 [98.97%]

    This trash has an accuracy somewhere around 9000, 2000 more than the magma lord ghost.

    There are still some uncertainties here, and it would be cool if people could fill in the gaps in what I've done, even just more data on more mobs would be nice. for instance i rarely tank raid bosses so i don't really even know for sure what boss accuracy is like, i have yet to tank any trash that misses me less than 20% of the time though.

    alright, what does this mean for increasing avoidance AC by maxing out HAgi. from what I've seen, going from all ROS HSta augs to all HAgi augs will give you about a 400 point increase to your avoidance AC. Using the magma lord as a baseline, if it is not debuffed, and if you had 3200 avoidance ac vs my 3600 you would probably get hit 23 percent of the time. So adding 400 avoidance to hit 3600 gives you 26%, roughly 3 % increase. However, with somnolence on him at 3200 avoidance you get 30% and at 3600 you get almost 34%. potentially 4% in that scenario.

    not a ton, but neither is the 10,000 hp you get from the HSta augs. I hit about 330k hp without undying life on raids, I'm sure warriors get much higher than this, oddly, the 10k hp increase would be roughly a 3% increase for me.

    However, the 10k hp is a known quantity, as we've seen there is definitely some truncation going on in our avoidance calculation. so, it's possible that you get hosed by getting your miss chance to 20.99% but it gets truncated down to 20%. It's hard to be certain what you are getting with HAgi.

    The final point for HSta is it helps with spell damage, HAgi doesn't

    In my opinion, HSta and HAgi are both perfectly viable, my personal opinion is I'm sticking with HAgi though, it's more important to me to take less damage and require less healing than have more hp for healers to top off.

    To all the people who are going to say, none of this matters, using your abilities appropriately is whats important, while this is true, and is even highlighted in this post, somnolence does more for your miss chance than maxing out HAgi ever will. All the little things add up to big things though, and there are a lot of little things in this game. While it's true that no amount of HAgi is ever going to save you if you aren't using your abilities properly, it doesn't hurt.

    In addition, I've shown pretty clearly that these accuracy decreasing debuffs are strong and you ignore them at your own peril, enchanters should be crippling to get somnolence on mobs even if shaman cripple is already on it. warriors should be gut punching as much as they can, if these things aren't being used in your raid guild, and you're guild is struggling, they can help.
    Sancus and gotwar like this.
  2. Hickers Elder

    Just throwing this out there:

    Accepted wisdom is that tanks stack either HAgi or HSta. However, HStr is used in the calculation for shield AC, and it’s well documented that shield AC is important for tanks. Just wondering if anyone has ever stacked HStr and seen if it makes any parseable difference?
  3. p2aa Augur

    Many tanks have swapped from Hagi to Hsta because after switching they saw no difference in miss %, they weren't missing less. Which leads me to the conclusion that avoidance AC plays hardly any impact on raid bosses. I lol at saying a warrior AA is responsible of lowering raid mobs accuracy. If it had really an impact, do you think devs would have let such an overpower ability in game without nerfing it ?
  4. gotwar Gotcharms

    This is a good post.
    Sancus likes this.
  5. Koryu Professional Roadkill

    [51053/12802] Gut Punch XII
    Classes: WAR/254
    Skill: Melee
    Target: Single
    Range: 100'
    Resist: Physical -45
    Reflectable: No
    Focusable: Yes
    Casting: 0s
    Duration: 6s+ (1 ticks), Dispelable: Yes
    Hate: 6500
    5: Decrease Chance to Hit by 50%
    6: Decrease AC v2 by 120

    Patch notes link for Gut Punch being nerfed.
  6. mackal Augur

    My guess is how a hit is actually done is that the numbers are just rolled against each other. If accuracy is higher than the avoidance roll, it's a hit. Very simple algorithm, and a test program matches your results roughly. SPA 184 is a percentage mod to the accuracy number basically (I would have to look up order of effects again or at least what I think they are :p )
  7. Wulfhere Augur


    It was un-nerfed again 2 years later (2014) from 25% back to it's current 50%.

    The thing is that it's only a 1 tick base duration debuff, so it's power might be overlooked. Likewise for Shadowknight's Veil of Darkness AA ,which is arguably more powerful depending on cool downs.

    [48702/41815] Veil of Darkness II
    Classes: SHD/254
    Skill: Melee
    Target: Single
    Range: 200'
    Resist: Corruption -2000, No Sanctification
    Reflectable: No
    Focusable: No
    Casting: 0s
    Duration: 12s (2 ticks), Dispelable: Yes
    Hate: 35000
    4: Decrease Chance to Hit by 75%

    While researching SPA 184, I found two items that could be part of this discussion (especially for paladins since they are the only tank that lacks an SPA 184 debuff ability).

    Lance of Emptiness

    Clay Smoke Bomb
  8. Derresh Augur

    Tanks stack hsta now because of the mod 2 changes that allows them to go past 35 melee shielding. Its a safer more consistent mitigation.
  9. Fooba Augur

    Dat 40 melee shielding at 2000 hsta. Thread over.
  10. josh Augur


    I've done some math on that and i agree, that is actually significant. Most people seem to dismiss it as not impactful but from what I've seen it might actually be worth it. I would have to say that i think HSta is almost certainly best for warriors. Primarily because shielding stacks completely with final stand discs. Knights have guardian now, and DP so it might tip the scales for us too. But, you absolutely want to hit the 2k mark if you are going HSta
  11. p2aa Augur

    In theory it does this, but I doubt it ends on the raid mob hitting 50 % less. I'm pretty sure there is some other mechanism that the mob has so that it's not ending with 50 %. Do you really think devs would let an ability with 10ish sec refresh make raid mob chance to hit less by a flat 50 % ?
  12. p2aa Augur

    Honestly going from 38 % to 39 % shielding is not doing less than going from 39 % to 40 % shielding.
    If melee shielding is your goal to gear Hsta, you risk to be disappointed. 1% shielding every 400 points.
    It means a mob instead of hitting you for 20 000, will hit you for 19 800. If it hit you for 40 000, it will hit you for 39 600.Not groundbreaking.
    At least for myself, I have chosen to stack Hsta (while also keeping a few type 7/8 aug with some Hagi / Hdex on them) because the other 2 don't work on raid mobs, and because the added HP is handy for the harsh AE of actual raids.
  13. Xorsazis Augur


    Decreasing Accuracy by 50% doesn't mean +50% end result in misses. The formula you are looking at is most likely far more complicated than just straight up Accuracy vs. Avoidance. Nobody has accounted for level difference. I would expect Gut Punch is doing it's part, but when you're 5+ levels on the players, that swings the accuracy more in your favor. God only knows how many other factors there are.

    As far as heroics, there's nothing showing that any heroic is going to make you a superstar vs. any of the others. 90% of this game is reacting to the environment and using the correct abilities at the correct moment.

    Foob's point is spot on.
  14. josh Augur


    Level difference has no effect on your miss chance. my formula is accurate. The bot that i was fighting against was level 85, and i did a parse with another bot who was 110 with the same avoidance and it made no difference, i missed the same amount. It's literally your avoidance vs their accuracy and nothing else. There is a ton that goes into calculating your avoidance and accuracy, including buffs and debuffs, but at the end of the line they appear to be all that is used to determine your miss chance.


    It wouldn't make the mob hit 50% less. It lowers its accuracy by 50%. a raid boss with 10k accuracy against a tank with 3200 avoidance would hit the tank 16% of the time. 50 * 3600/(10000) = 16% with gut punch active it would be 32% of the time 50 * 3600/(10000 * .5) = 32%

    The only thing i might be missing here, is it's possible that raid mobs have an innate SPA 184 bonus, like for instance, the raid mob has 7000 accuracy but a 50% innate bonus that gives it 10500, in this case, since it's innate and not a buff, the 50% from gut punch would be subtracted from the 50% innate, so instead of lowering the accuracy to 5k, it ends up lowering it to 7000.

    I thought i might see that on the magma lord but i don't. somnolence does very nearly what i would expect it to, I know you can't accept this, but it's true, gut punch is that strong, somnolence is that strong, and avoidance works very closely to how I've described. adding avoidance AC works just fine and other parses you and others have done have been mixed up by gut punch and other accuracy decreasing debuffs.

    The devs miss things that are too strong all the time, i don't think gut punch is too strong, you are the one saying that. so if it gets nerfed everyone blame him and not me lol. As far as I can tell, it works as I've described.
  15. roth Augur

    That’s not really how Shielding works. Unless things have changed in the years since I’ve learned this, without my knowing, anyway ...

    Every physical hit in the game has two components : Damage Bonus, and Damage Interval. Damage Bonus is a fixed amount, dependent upon the mob. Damage Interval is a range of fixed amounts, again dependent upon the mob. Shielding reduces the Damage Bonus ONLY. So if a mob would hit you for 2000 and Shielding reduces it to 19800, then that same mob hitting you for 40000 would only be reduced to 39800.

    A different mob, with different DB and DI values, would have a different amount of damage mitigation per hit, based upon the Shielding value vs the Damage Bonus.

    Shielding’s value is great against high DB, low DI mobs. Low DB, high DI mobs, Shielding isn’t nearly so valuable.
    Koryu likes this.
  16. p2aa Augur

    Once again, if all these accuracy debuff abilities were as strong as you list them, if you think that devs missed them and they just discovered it thanks to you, you are very delusional.
    You call parses of others "false" when they don't suit your goal. With way less avoidance AC than you on the same raid mob mini, I missed the same %. This is in line with all the parses I did back in TBM that showed that loosing lot of avoidance AC didn't make me miss less. Things didn't change in RoS. As you didn't like this result on the raid mob mini, you tried to invent this story of gut punch being the explanation.
  17. p2aa Augur

    I know how shielding works and how damage works. This is a flat damage reduction of the same amount to every hit at the end. If you want to play with words, I should have said "a different mob hitting for 40000 this time".
  18. mackal Augur

    Time to have chain Gut Punch on mobs! Way more fun than chain cheal!
  19. josh Augur


    holy crap dude, you're the delusional one. Do you think the raid mob that i hit with somnolence hit me almost 7% less often because of? what? i can't even imagine what you think caused that if it wasn't the debuff i used.

    Total Melee --- Damage: 25168119 --- Avg hit: 12019 --- Attempts: 3005 --- Riposted: 11/3005 [0.37%] --- Parried: 6/2994 [0.2%] --- Dodged: 10/2988 [0.33%] --- Blocked: 12/2978 [0.4%] --- Defended: 39/3005 [1.3%] --- Missed: 776/2966 [26.16%] --- Hits: 2190/2966 [73.84%] --- Absorbed: 96/2190 [4.38%] --- Real Hits: 2094/2190 [95.62%]

    Total Melee --- Damage: 35640703 --- Avg hit: 12110 --- Attempts: 4476 --- Riposted: 10/4476 [0.22%] --- Parried: 8/4466 [0.18%] --- Dodged: 15/4458 [0.34%] --- Blocked: 7/4443 [0.16%] --- Defended: 40/4476 [0.89%] --- Missed: 1461/4436 [32.94%] --- Hits: 2975/4436 [67.06%] --- Absorbed: 32/2975 [1.08%] --- Real Hits: 2943/2975 [98.92%]

    what explains this to you? I'll do it a 100 times if you want me to, I've parsed against the magma lord a dozen times and he hits me roughly 26% of the time every time without somnolence. He has never even come close to 27%, yet with somnolence it drops to 33%. weird? truly weird how that number also matches so closely with the math of all of this as I've described it. Such a coincidence there's no way I'm right though because it's impossible that, that SPA works the way I've described because you say it doesn't so it obviously doesn't.

    Gut punch and or somnolence is what caused your parse to be similar to mine, if you seriously can't accept this than you can't be helped.
  20. p2aa Augur

    Nope.