I see for example why my human with 2H axe hit is mostly 8% and i can understand that according to that data.
0,6% miss is 1 miss per ~200 hits and there can be items that give more profits than 0,6% hit.
But when i’m combine human warrior or paladin with 2H mace or sword, then hit is mostly 4% or 5% instead of 6%. 1% difference give me extra 1 miss per 100 hits.
It’s really worth of it?
That’s why I based all the rankings on a fairly full model of each spec’s damage. Hard to just say whether it’s worth it or not without doing a bit of math.
As a tank, while I do understand that my average TPS will be better, the chance to miss with my taunt should not be higher than absolutely necessary.
A customized option to set a minimum % hit would be interesting.
For example, before the actual optimization, you will set that you want a minimum of 9% hit and AMR will work around that to achieve it.
Yeah, we might add a customization option to let you force a particular amount of hit on your gear. Hit was always such an annoying mechanic for this very reason: math and averages want you to do one thing, feelings often want you to do another. Missing feels bad. People would rather do less damage than miss sometimes.
Snapshot ID: 9be810b347844d7b89c83454c432e035
So I ran a couple of tests of my own in-game.
I followed AMR down to a T and used Bone Slicing Hatchet on Main Hand and Mirah’s Song on Off Hand.
Edit: Race: Orc
Targets: Winterfall Pathfinder / Totemic / Den Watcher
Level: 53-56
On my second test, I ran with Axe of Deed Woods on Main Hand and Flurry Axe on Off Hand.
As you can see, my MISS went from 23% to 11.4% and my DPS increased dramatically.
I want to thank for all the effort put into bringing AMR to Classic, but I urge you guys to take a closer look at Weapon Skill / Hit Rating again, because as of now, AMR is giving me different suggestions from other sources and being somewhat unreliable for the time being.
Your sample size is way too small to make much of a comparison, unfortunately. That’s why we use mathematical models instead. (Just changing those weapons, per the formulas above, would not change your chance to hit by more than 10%! Adding 5 weapon skill from the orc racial would only change your chance to get hits by roughly 2.5%.)
Like I mentioned, we’re probably going to add something to let people force a certain amount of hit. I know that people don’t like the increased variance that extra misses bring to the game - it’s annoying. Especially considering how short most fights are in raids - each miss feels a lot bigger than it really is, statistically speaking.
Is it possible to implement an option that allows you to pick certain types of weapon?
For example, if I don’t have Edgemaster, I’d like to choose Axes instead of Swords, instead of having to go weapon by weapon ticking them off.
However, even after all this, there is one thing that is putting me off.
AMR suggests that I use Serathil & Bone Slicing Hatchet, yet, a spreadsheet from Fight Club along side with a Simulation for Classic tells me I will achieve a higher DPS with Deep Woods & Flurry. Would you know why that happens? I selected the same fight duration and buffs for comparison.
Snapshot ID: 2a73e3c7d03b4afebf0818482cfd33a4
The item lists can be filtered with the filter box at the top - just type in “axe” and it’ll filter the list to axes.
I’m not going to be able to answer why other sources of info are different unless I went through every detail of their spreadsheet and simulator to look at how they have implemented everything.
I’ve gone through my calculations for warrior backwards and forwards now multiple times. I just don’t see how flurry axe could calculate out to an average increase in damage compared to a lot of other weapons, assuming it works the way people say it does.
Bone Slicing Hatched compared to Axe of the Deep Woods for your main hand:
Hatchet:
40.6 DPS
1.7 Speed
13 Agility
Deep Woods:
41.5 DPS
2.7 Speed
tiny proc worth less than 3 DPS
These items are almost identical from a stats perspective. They’re both axes. The hatchet is fast, deep woods is slow - which is the main difference. All the calculations I do point towards fast weapons being better for fury warriors. Your weapon dps doesn’t matter for execute or bloodthirst, which are your two main special attacks. You dump extra rage on heroic strike. Using heroic strike costs rage, adds a flat amount of damage, and also makes you lose out on the rage that auto would have generated. When you have a fast weapon, when you use heroic strike, you don’t miss out on as much rage. Slow weapons benefit from the flurry axe more and that weird thing with off hand using yellow hit while heroic strike is queued… but really it’s all about rage generation from all the math I can do.
Hey just wanted to jump in, with your calculations is it better to purposely force the off hand extra hit or to purposely avoid and gain the extra rage? or just dump when you need to and not care either way
The DPS you gain from the extra hit chance on the off hand while main hand heroic strike is queued is very small. I don’t think it needs to be something that you specifically aim/time for. If it happens it happens and is some very slight gravy damage. Wasting any amount of rage would negate the gain.
With that formula is something wrong?
Target Defense - Your Weapon skill = D
Miss Chance = (5.6-IF(D<-10,(D+10)*0.4-1,D *0.1))/100
For 305 WS and 63 lvl target D >= -10
(5.6-(10*0.1))/100 = (5.6-1)/100 = 4.6/100 = 4.6%
I think before IF should be +
(5.6+(10*0.1))/100 = (5.6+1)/100 = 6.6/100 = 6.6%
I copied it wrong from my notes, it’s supposed to be
Your Weapon Skill - Target Defense = D
Please correct me on this if I’m wrong, but I was under the impression, that a slow weapon in main hand is something you may use to better toggle heroic strike, without actually using it (while low on rage) and thereby circumventing the 19% added miss chance on your off hand.
I would imagine that, regarding weapons, the simulation doesn’t take toggling heroic strike into account?
To make this a bit more understandable: You watch your main hand swing timer (which should be slower than your off hand), after a hit, you activate heroic strike and just before the next hit, you deactivate it, thereby saving rage, but still giving your off hand the chance to hit as if it was an ability.
We do not model heroic strike cancelling. It is an extremely small gain for a lot of extra work - and most people are not doing it. If you mess it up even once and miss out on, say, one bloodthirst because of delays from accidentally using rage, you negate the entire gain from doing it.
If you are going to do heroic strike cancelling, you’d want a slow main hand for sure.
And when using a slow weapon, it will be very hard to mess this up even once. You can simply cancel a lot sooner than would be optimal and skip some offhand-swings which might be “dangerous”. It is not only a gain in damage, it is also a gain in rage to do that (because you hit more often and the offhand can also generate rage).
While not so many people might do something like this on retail, the classic playerbase is, in my experience, a bit more extreme when it comes to using such things.
I think the key problem is current knowledge of players.
For example from Blue posts i know that hit cap is between 8% and 9%, but closer to 9%.
That players who didn’t read that still believe that hit cap is 8% like in “vanilla”, so most of players that follow class discords, wowhea, icy-veins ect… There were theorycrafters during vanilla, that calculated 8.6%, because they missed on theirs characters with 8%. Even today when you tell for example on Warriors discord that you aiming for 8.6%, then you can be very ridiculed by them.
Sadly i can’t find any official formula, so we can only assume what it should be like.
Also i did questionnaire inside my guild about hitcap and dps.
From results and comments i can deduce, that even if they can increase their dps by lowering hit chance, then still they prefer stay on hitcap, because they are more afraid of miss hit during boss encounter.
And it was a unanimous decision of all respondents.
The original post in this thread tells you exactly how to calculate the hit cap!
Calm down 
I see that formula here and that you implemented 1% ignored hit above 10 Defense, but i can’t find any similar formula in another sources.
Currently for me you are the only one reliable source that says it is 8.6% instead of 9% for 300 ws and 6.6% instead of 6% for 305 ws.
Also i checked all my wlogs and few other top ret paladins from list and i dont see any miss with 305 ws + 6% hit.
According to that and some dwarf paladin logs option…
Target Defense - Attacker Weapon Skill <= 10
Hit Cap = 5% + (Target Defense - Attacker Weapon Skill) * 0.1%
Target Defense - Attacker Weapon Skill > 10
Hit Cap = 5% + (Target Defense - Attacker Weapon Skill - 10) * 0.4% + 1%
…looks sensibly.
Please understand my doubts
Blizzard made a blue post confirming that the first 1% hit is ignored when target is more than 10 defense above you. The post isn’t available anymore for some reason though.
I don’t know for sure about the 8% miss vs 8.6% miss - back in the day it was “general knowledge” that miss was between 8 and 9%. I used a formula that I found that seemed to match what people see happen in-game and what we all remember.
I’ve said it a few different times… it doesn’t matter if you are at the hit cap. You’ll still get dodged. Being 1% below the hit cap won’t really feel any different than being at the hit cap - you can never get rid of at least a 5% chance to be dodged. If people want to be hit-capped, that’s fine… but you’ll still be avoided and have attacks that do nothing. I think people don’t realize that in their compulsion to reach the cap.
