Fight Analysis shows ridiculous results

I recorded six fights with an even level 4 Ash Rattler and eight fights with a lower level 3 Ember Moth. I wanted to see if my new staff with supposed "GREAT" accuracy was indeed living up to it's name.

Leather padded staff Lvl 7 stats:
Dmg: 3-17, Delay 34, Accuracy: "GREAT"

I am level 4 Druid, my stats are:

96 str
106 dex
104 agi
104 cnt
Lvl 7 Staff skill


Here's what I found:

Lvl 3 Ember Moth Fights:
Fight 1:
Total Swings: 12
Total Landed: 6
Accuracy 50%
Avg Dmg 5

Fight 2:
Total Swings: 12
Total Landed: 8
Accuracy 66.67%
Avg Dmg 6.75

Fight 3:
Total Swings: 11
Total Landed: 7
Accuracy 63.64%
Avg Dmg 7.64

Fight 4:
Total Swings: 12
Total Landed: 9
Accuracy 75%
Avg Dmg 6.92

Fight 5:
Total Swings: 8
Total Landed: 8
Accuracy 100%
Avg Dmg 10.38

Fight 6:
Total Swings: 14
Total Landed: 8
Accuracy 57.14%
Avg Dmg 4.93

Fight 7:
Total Swings: 9
Total Landed: 7
Accuracy 77.78%
Avg Dmg 8

Fight 8:
Total Swings: 20
Total Landed: 9
Accuracy 45%
Avg Dmg 3.9


Lvl 3 Ember Moth Fights:
OVERALL Average Accuracy: 66.9%
OVERALL Average Damage: 6.69



Lvl 4 Ash Rattler Fights:
Fight 1:
Total Swings: 16
Total Landed: 10
Accuracy 62.5%
Avg Dmg 4.69

Fight 2:
Total Swings: 13
Total Landed: 7
Accuracy 53.85%
Avg Dmg 5.62

Fight 3:
Total Swings: 12
Total Landed: 10
Accuracy 83.3%
Avg Dmg 6.67

Fight 4:
Total Swings: 10
Total Landed: 8
Accuracy 80%
Avg Dmg 10.2

Fight 5:
Total Swings: 12
Total Landed: 6
Accuracy 50%
Avg Dmg 5.17

Fight 6:
Total Swings: 19
Total Landed: 13
Accuracy 68.42%
Avg Dmg 4.95


Lvl 4 Ash Rattler Fights:
OVERALL Average Accuracy: 66.35%
OVERALL Average Damage: 6.21



Now tell me something.. Does an average OVERALL accuracy of 66.9% sound right for a mob that is a lvl BELOW me when I have a staff skill 3 levels ABOVE me and it's accuracy is described as "GREAT"? I don't think so!

I also think a 66% accuracy is ridiculous for an even level fight against the Ash Rattler with my 3 lvl staff skill advantage.

My overall observation is that if I didn't use magic to supplement the melee damage of the staff I would almost die almost every fight due to:
1. Average accuracy of 66%
2. Stuns, Mes's and other immobilizing effects these creatures do to me even with my 104 agi and 104 cnt ratings.
3. Other spell direct damaging effects against me

If I went pure melee, I would leave the fights with around 20-45% health. However, when I supplement my melee with regeneration and direct damage gems I can win fights with 60-80% total health.

Attached is my spreadsheet with the recorded values of each fight. I only recorded damage done, not damage taken.

Please adjust accuracy to match description of weapon! Also tune mob difficulty to appropriate levels!

[Glitchless]

Sample size of 6 and data ranging from 45% to 100%... your std deviation is huge making this not statistically significant.

That being said people are probably in general over-estimating the effect of the accuracy modifiers. If you hit 70% of the time with an accuracy Good you're probably going to hit 73% of the time with accuracy Great. The dam/delay is far more important.

The wild fluctuations in accuracy is the cause for concern. You should review the formula that calculates it. I could gather data all day but results will show that the extremes were such values, maybe even greater variations the more I collect. With more results sure I will get a better idea of overall accuracy but I would wager a bet to say that overall accuracy will NOT be "GREAT".

Tell me an appropriate sample size so I can waste a day of my life collecting results to prove you are passing off my data as statistically insignificant when in fact you should look into this further..

Accuracy modifiers? What about level difference, skill difference, weapon lvl advantage??


I don't know what your definition of "GREAT" is but here's my thoughts:

50-60% of time I would consider poor accuracy
61-70% fair
71-80% good
81-90% great
91-100% deadly

66% I hardly find GREAT.

If anyone would like to contribute their data collection results, this would be an interesting thread.

Think of it like baseball. If a batter hits the ball 66% of the time... that is DAMN good.

i gotta agree, i tend to miss at app 50-60% of the time with a weapon that hits normal even when fighting monsters 8 lvls below me. i havent done as extensive research as even noid has done, but i tend to miss alot more then monsters do

[doctor-warlord]

well my opnion is that you can make an accurate anylisis since you lvl skills change and equipment is changed pretty often so the only way you can make an accurate anylisis is basicly if you convince jeff to release the accuracy calculations which i doubt he will even consider

I have to agree with some of these posts. I have to collect some data, but I'm pretty close to positive that my main hits more often with a weapon accuracy rated lower (Good) than one that is rated higher (Excellant).

I don't know what you guys are talking about... I hit just fine.

Originally Posted by Glitchless

Sample size of 6 and data ranging from 45% to 100%... your std deviation is huge making this not statistically significant.

That being said people are probably in general over-estimating the effect of the accuracy modifiers. If you hit 70% of the time with an accuracy Good you're probably going to hit 73% of the time with accuracy Great. The dam/delay is far more important.

lol yeah ^.^ get a sample size of at least 100 and come back.

Originally Posted by Noid

The wild fluctuations in accuracy is the cause for concern. You should review the formula that calculates it. I could gather data all day but results will show that the extremes were such values, maybe even greater variations the more I collect. With more results sure I will get a better idea of overall accuracy but I would wager a bet to say that overall accuracy will NOT be "GREAT".

Tell me an appropriate sample size so I can waste a day of my life collecting results to prove you are passing off my data as statistically insignificant when in fact you should look into this further..

Accuracy modifiers? What about level difference, skill difference, weapon lvl advantage??


I don't know what your definition of "GREAT" is but here's my thoughts:

50-60% of time I would consider poor accuracy
61-70% fair
71-80% good
81-90% great
91-100% deadly

66% I hardly find GREAT.

If anyone would like to contribute their data collection results, this would be an interesting thread.

do you have low dex as a piercer/low str as a crush?

those almsot must be considered.

Originally Posted by Noid

The wild fluctuations in accuracy is the cause for concern. You should review the formula that calculates it. I could gather data all day but results will show that the extremes were such values, maybe even greater variations the more I collect. With more results sure I will get a better idea of overall accuracy but I would wager a bet to say that overall accuracy will NOT be "GREAT".

Tell me an appropriate sample size so I can waste a day of my life collecting results to prove you are passing off my data as statistically insignificant when in fact you should look into this further..

Accuracy modifiers? What about level difference, skill difference, weapon lvl advantage??


I don't know what your definition of "GREAT" is but here's my thoughts:

50-60% of time I would consider poor accuracy
61-70% fair
71-80% good
81-90% great
91-100% deadly

66% I hardly find GREAT.

If anyone would like to contribute their data collection results, this would be an interesting thread.

You need a large sample size to reduce the large fluctuation, it is recommended about 100 at minimum.

If you wanna record for 100 fights, be my guest.

oh didnt notice. i saw a link for thiss.

Originally Posted by Coccyx

this thread died a year ago.... why are we bringing it back now?

Someone bumped it I guess...if I woulda known that I wouldn't have posted