# Thread: Strange experimental results

1. New Registrant
Join Date
Mar 2009
Posts
2

## Strange experimental results

Night Elf Warrior 80 vs. Heroic Training Dummy (Worldboss)
Defensive Stance, Impale 2/2, 1H Weapon Specialization 5/5
Red Sword of Courage (160-299 @ 1.60), Weapon Skill 400, Haste Rating 0, Attack Power 2707
Hit Rating 191, Armor Penetration Rating 0, Crit Rating 25, Expertise Rating 166
Source : Recount, Melee Damage
5832 hit (63.6%) : 174-324 (avg:281)
2197 glancing (24.0%) : 163-274 (avg:215)
701 parry (7.6%)
263 crit (2.9%) : 501-648 (avg:572)
180 miss (2.0%)

Theorically, my sword hits a worldboss for 249-325, according to these calculations :
(160+(2707/14*1.6))*0.9*1.1*(1-(13083/(467,5*80+13083-22167,5)))=249
(299+(2707/14*1.6))*0.9*1.1*(1-(13083/(467,5*80+13083-22167,5)))=325
But I don't have these stats (174-324 differs from 249-324), even if theorical average damage is near exprimental average damage (281), according to this calculation :
((160+299)/2+(2707/14*1.6))*0.9*1.1*(1-(13083/(467,5*80+13083-22167,5))=287
The difference between 287 and 281 may be explained by the fact that Recount doesn't make a difference between "hit" and "hit but blocked". Heroic Training Dummy have a Block Value of 76.

Can you explain this difference between my theorical calculations and my experimental results ? Is it an error from Recount ?

2. Established Registrant
Join Date
Dec 2008
Posts
457
I think you explained it yourself. If recount is considering "blocked hits" as "hits", then it makes sense for your minimum hit to be your theoretical minimum minus the targets block value.

Also, you are calculating your theoretical average by adding min and max dmg and dividing by two, which is correct. But if you have an actual sample of hits, you actually have to average all the recorded hits (recount does this to give you an average). A difference of 6 points of damage over a theoretical average of 281 damage is merely 2.1%, which in my opinion is within an acceptable margin of error, considering you may have block noise in your data sample.
Last edited by Molohk; 03-03-2009 at 11:46 AM.

3. Professional Casual
Join Date
Aug 2007
Posts
913
After accounting for blocks, the values seem to be perfectly right (plus/minus 1, which could be a rounding thing)?

+ Reply to Thread

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts