Page 3 of 3
Re: 92 Magery
Posted: Thu Aug 27, 2009 1:32 am
by son
With 92.3 magery at OSI you never fizzled recall. I do it daily and its not because I am opening containers.
Re: 92 Magery
Posted: Thu Aug 27, 2009 6:41 am
by archaicsubrosa77
You guys just have bad Karma....
Re: 92 Magery
Posted: Thu Aug 27, 2009 9:01 am
by poogoblin
I haven't fizzled recall since about 64 magery
Re: 92 Magery
Posted: Thu Aug 27, 2009 10:17 am
by Hoots
On the 8ths, i want to do some more testing but i have been doing scribing runs of 100 ev's and not seeing 50% averages.
I need to look at it more, but it certainly *feels* like casting and scribing 8ths has a lot more 0-4's to 0-7's compared to 4-0 or even 3-0 success.
Could be the along the same lines of *risk* dice though... for those of you who play that game you never remember going 8-0, but you will tell your grandchildren about the time you lost 10 attackers to a single...
Scribing success should be the same as casting.
Ill try to report numbers on 1000 ev's and see how close we are to 50%
Re: 92 Magery
Posted: Thu Aug 27, 2009 11:47 am
by Mikel123
Again, I'm pretty sure it was never 50% but instead 45%.
Re: 92 Magery
Posted: Mon Sep 14, 2009 11:21 pm
by Mikel123
Decided to test this out, after I fizzled Ressurrection 9 times in a row.
I cast Earthquake in my house 100 times, and saw 55 successes and 45 failures, for a success rate of 55%. Remember, my post said that there should be a 45% chance of success.
So I went to a binomial calculator site (
http://www.stat.tamu.edu/~west/applets/ ... ldemo.html) and plugged in this data. Given a 45% success rate, the chance of succeeding at 55 out of 100 trials is 2.84%. So, if the 45% number is correct, I had a pretty rare run. I'll try this again, but I'd suggest other folks do it as well. I made a macro that cast EQ if my mana was above 90, then meditated if it was below 90. That way you succeed the med every time, if you're GM.
Re: 92 Magery
Posted: Tue Sep 15, 2009 10:46 am
by Derrick
Mikel123 wrote:So I went to a binomial calculator site (
http://www.stat.tamu.edu/~west/applets/ ... ldemo.html) and plugged in this data. Given a 45% success rate, the chance of succeeding at 55 out of 100 trials is 2.84%. So, if the 45% number is correct, I had a pretty rare run.
This is kind of a statistical misinterpretation. The chance of succeeding exactly 55 is rare on a 50/50 here but the change of succeeding 55% or more times is 18%, the chance of succeeding more than 50 times is 46%
If the intent was to succeed exactly 55 times, your chance would be about 3%, but that's a very specific case.
Thanks for doing this test, I'd say that your results verify that this mechanic is correct, 50% chance of 8th at GM magery.
Re: 92 Magery
Posted: Tue Sep 15, 2009 11:50 am
by Mikel123
Sorry, I should have been clearer, I plugged in the chance to hit AT LEAST 55 out of 100. The chance of hitting EXACTLY 55 is 1.08%. The chance of hitting AT LEAST 55 is 2.84%.
I'm sure there's a better test I could run (i.e. What is the chance that 55/100 is due to chance, given an expectation of 45?) but I haven't found a calculator online yet for that.
I'll probably test this a couple more times, but yeah - I'm reasonably satisfied right now.
Again, I think it should be 45% success rate at GM Magery (and that's what I stat-tested against. Assuming 50% expectation, if that's what is coded in, the probability of hitting 55 or more out of 100 is over 18%). My rationale for 45% is earlier in this thread, but in any case it's not a big deal. It would take thousands of trials to really see in practice if we're at 45% or 50%. And really, the only major difference is a few more or less lost reagents in fizzles. Not like people are PvPing too much with level 8 spells where success rate is crucial...