City of Heroes: Live Forum Archive

Overwhelming Force drop rates

Athena Six

By Athena Six,
Posted on: 2012-07-11 11:29 in The Market and Inventions

 

Posted

Quote:
Originally Posted by Bloodspeaker View Post
Has anyone compiled all the results from all the posters here? Many small samples can be compiled into a larger sample, as long as everyone's numbers are accurate.
Not since post #19, but the data since then hasn't looked like it will dramatically alter the result.

 

Posted

the game has been trolling me hard. 12 runs and no proc lol.


 

Posted

Final Results:

Acc/Dam - 6
Acc/Dam/End - 6
A/D/E/R - 8
Dam/End/Rec - 7
End/Rec 17
Proc - 12

Two of the Acc/Dam came on the last day, so the RNG gods smiled on me at last.

 

Posted

Quote:
Originally Posted by Ironblade View Post
I mock your disappointment. I've run it 23 times and zero procs.
Ha! Well good sir I would like to inform you that after 26 runs I was able to land the proc again! Then I promptly collapsed on my bed...because I could not take any more lol. My eyes were bleeding lol.


 

Posted

Quote:
Originally Posted by Ironblade View Post
I mock your disappointment. I've run it 23 times and zero procs.
I didn't join on in this event until the last minute. I ran it seven times last night before I got too tired and had to go to sleep.

I got the proc on the very first run, and I got all but one of the enhancements (I missed the acc/dam one).

 

Posted

Quote:
Originally Posted by Sailboat View Post
Updated total:

Acc/Dam -9
Acc/Dam/End - 11
A/D/E/R - 6
Proc - 2
Dam/End/Rech - 4
End/Rech - 4

Two procs in 36 runs so far.
My final tally for entire event (including the numbers quoted above):

Acc/Dam - 11
Acc/Dam/End - 17
A/D/E/R - 15
Proc - 5
Dam/End/Rech - 11
End/Rech - 7

Five procs in 66 runs (13.2 runs per proc). But I got 3 of them in 11 runs the day before the event expired, bringing my proc production up sharply; if you subtract that day, it would be 2 procs in 55 runs. Heh.

If we are to die, let us die like men. -- Patrick Cleburne
----------------------------------------------------------

The rule is that they must be loved. --Jayne Fynes-Clinton, Death of an Abandoned Dog

 

Posted

Quote:
Originally Posted by Sailboat View Post
2 procs in 55 runs. Heh.
Ouch.


 

Posted

My final count, including the drops I already posted:

Code: 
      10 Overwhelming Force: Accuracy/Damage.
     20 Overwhelming Force: Accuracy/Damage/Endurance.
      8 Overwhelming Force: Accuracy/Damage/Endurance/Recharge.
      6 Overwhelming Force: Damage/Chance
      9 Overwhelming Force: Damage/Endurance/Recharge.
      7 Overwhelming Force: Endurance/Recharge.

 

Posted

Quote:
Originally Posted by Muon_Neutrino View Post
Acc/Dmg 0
Acc/Dmg/End 1
Acc/Dmg/End/Rch 2
Dmg/Proc 2
Dmg/End/Rch 3
End/Rch 2
My final numbers including the above:

Acc/Dmg 7
Acc/Dmg/End 7
Acc/Dmg/End/Rch 6
Dmg/Proc 11
Dmg/End/Rch 5
End/Rch 9

I think I made out pretty well - 11 procs in 45 runs is a 24.4% drop rate!

Quote:
Originally Posted by Sailboat View Post
2 procs in 55 runs. Heh.
I agree, ouch! I still say that I ended up with all your luck.

@MuonNeutrino
Student, Gamer, Altaholic, and future Astronomer.

This is what it means to be a tank!

 

Posted

Here's my personal final tally

Code: 
Acc/Dmg		11
Acc/Dmg/End	17
Acc/Dmg/End/Rch	9
Dmg/Proc		7
Dmg/End/Rch	10
End/Rch		8
But you don't care about that. Let's look at the sum total.

Code:
Acc/Dmg	172
Acc/Dmg/End	205
Acc/Dmg/End/Rch	161
Dmg/Proc		151
Dmg/End/Rch	150
End/Rch		165
That's a total of 1004, so, if each had a 1/6 chance of dropping, we'd expect to see 167.333 of each.

I ran the Chi square test and the result was
Quote:
By conventional criteria, this difference is considered to be statistically significant.
Emphasis mine. So, according to this calculator http://www.graphpad.com/quickcalcs/chisquared1.cfm
the results we got were NOT from a uniform distribution where probability of a drop (p(x)) = 1/6.

I welcome anyone to validate or run their own tests. I don't want to say how many years it's been seen I last had to get into the details of this sort of analysis, so I'm rusty, to say the least.

Global = Hedgefund (or some derivation thereof)

 

Posted

Although it's a useful analysis, I seem to recall that the chi-square test is pretty substantially affected by sample size (as in, more likely to return significant results when you have a large sample, even if the practical difference isn't very large). All this one tested was whether the entire distribution of drops was a consistent 1/6 per item. The weirdness of the a/d/e numbers makes that kind of a foregone conclusion, but if you look at the rest they do kind of cluster around the target value.

I think what people are likely to be more interested in is whether there were significantly fewer procs dropped than we'd expect. So 151/1004 vs. 1/6, rather than the entire distribution vs. 1/6. What would that be, a binomial test of 15% vs. an expected of 16.67%?

I'm sure there are some math majors around who'd be happy to help out with that.

My postings to this forum are not to be used as data in any research study without my express written consent.