Why Witch loot drops suffer so badly - the stats:
|
I agree, your sample size is to small.
You say "50 exalted can't drop", that's wrong. The distribution is a bell curve for drops and exalted are near the end. Yes it's statistically impossible for 50 to drop, but it is realistically possible for 50 exalted to drop, and in a row in fact. That is a difference you didn't make and so I'm making it right now. The reason people want a larger sample size is because you are trying to force a major game flaw onto us. While it is statistically unlikely that your information is wrong, it is realistically possible that it is wrong. Please conduct another study with 1000 drops. Feel free to combine both your studies. Do this, and I'm sure I'm sure you will make a much more compelling argument. I'd like to take a moment to also negatively criticize your study as well. Please check drop bias on a per area basis and then report back. See if other areas have drop bias for int gear where the chance it will drop is much higher compared to other gear. Check to see if maybe crossroads will drop lots of int gear whereas Fellshrine drops mostly dex/str. As well, conduct your study on a few different leveled areas. Maybe level 22 areas have drop bias for int gear? For what you have shown, the only argument you could make is "Drop rates for level 21 areas are biased against int items in general." This doesn't account for higher or lower level areas, area specific drop rates, or any other areas in the game. Thus, your study is extremely incomplete if you aim to make it related to the game as a whole. In no way shape or form could your study imply that drop bias for the entire game is favored against int as it currently is. Last edited by DestroTheGod#5470 on Mar 26, 2013, 11:12:01 AM
|
|
|
I see the trolls are still posting pointless posts about sample size. Oh well, good luck to them!
Back on topic: Nochalo thinks pure builds 'should' have less drops as the bi-classes items are more useful to all. Well, I disagree. The game has set up 6 basic classes, each with their own start-point and stat bias. Therefore I think that all the basic game-balance issues should begin by insuring each class has a similar, balanced, experience with the game. The option to multi-class has always been the specialty of the experienced player (in all RPGs), not the rank and file. Imml thinks that a player should use whichever item works best, whatever the RNG gives you is what you adapt to. He, for example, is prepared to wear a crap Chestplate simply because it's the one with linked sockets. Now, I agree with this sentiment on a balanced playing field of total pure RNG on an individual level basis, but this is not how the rest of the game is set up to be played. If we all started with one class and built randomly from there, fine, but, since we're choosing a class, and since we're being encouraged to make 'builds', the idea of just taking anything that looks nice as being 'the point of it all' is somewhat absurd IMO... ... to which we come to fsg, who wants to make a pure Strength Witch. To which I refer him to the answer I gave to Nochalo. Yes, Imml and Bru741 are correct, by Witch gear, I mean pure INT gear and diverting your witch to non-INT gear is simply multi-classing rather than multi-tasking. A witch specialising in two-handed weapons is simply a gimped Marauder rather than an interesting Witch build IMO. So, yes derbefnier, by Witch gear, I do mean pure INT gear, and I feel that's what any non-pedantic personality would assume as well (most people). I agree with most of the things Tabx says, obviously, as having the weighting as class-required rather than level-required would reduce the 'need' to trade and multibox to a great degree and reduce that activity to mainly currency items and perfection items (which it kind of is anyway). The 'more' even split effect would, I believe, simply make more happy gamers rather than more grumpy gamers. |
|
|
So we're the trolls by suggesting that you need to do a lot more work to build out your model? Good to know, because by the same token I can say that yours is the troll post by not finishing the job. Go take a class on statistics, then you'll see how flawed your initial argument is simply by lack of data.
|
|
|
while some who use that line have indeed adopted a troll-ish tone, i believe the bulk of the posters who have pointed this out in a reasoned manner and have pointed towards statistical evidence to prove their point in a logical manner.
lumping everyone pointing this, imo valid, point into a "troll" badge, calling their posts pointless, and then not addressing their issue makes me doubt your intent of actually wanting to have a full and thorough debate here. i doubt you'll convince many people of your view until you respond in a far more appropriate manner. " Last edited by dlrr#2847 on Mar 26, 2013, 11:19:10 AM
| |
|
I've stated my answer on sample size. Please keep it on topic. Sample size is not the topic. Feel free to start a new thread complaining how this thread does not have the correct sample size. Thanks :)
Because all the 'relevant' posts are getting buried by the pointless repetition, which is unfair to those who wish to make their view felt on the topic. Someone has stated the sample size isn't enough, job done, opinion known. My opinion has been replied, please stop drowning the thread (which is what I'm guessing you want). Thanks :). |
|
" I think you are wrong. Drops do not follow a normal distribution. It is more like a dice roll where values are then assigned as 'drops', ie. http://en.wikipedia.org/wiki/Uniform_distribution_%28discrete%29 It seems to me that you didn't come away with the fundamental point of statistics- of using numbers to prove something rather than intuitive guessing. Proving that something follows a certain distribution with statistical certainty is pretty hard (I have no clue how to do it even for this very simple distribution). It is very presumptuous to make claims you have no clue about, if I had to guess, I would say that 433 trials would be more than enough to prove a uniform distribution for a die. IGN: Achernan Last edited by howlinghobo#7728 on Mar 26, 2013, 11:31:03 AM
|
|
"it is not an opinion? | |
" And int-related items had exactly 33% in that sample. Which is where you would expect all the stats to be, in a larger sample size. But, more importantly, you seem to have some working definition of 'Witch item' that is probably not shared by a large number of the players in this thread. My witch, for example, wears a lot of STR/INT items, because I have the stats to support it, and I happen to have found useful rares. And at 28, I was still wearing pure DEX gloves because I hadn't found a rare who stats made me willing to change. If I have two fish, and you have three fish, how many fish do we have?
None. These are MY fish. |
|
|
This is foolish. Cronk goes out of their way to do something contributive with their time, and what's this? 95% of the people don't know how to respond except with short, useless statements or annoying "You need more", as if they did anything themselves. It's fucking pissing me off; every day, this nonsense prevails over everyone's hard work, and I am starting to really blow a gasket over this one. You fuckers better respect the OP's contribution, and work with it. If you want a larger sample number, why don't you go get your fucking asses out there and contribute, and stop being sorry ass children.
You will never see a man faking anger, passion and relentless behavior.
You will always see a man faking love, politeness and respectful behavior. |
|
|
one sample of ~400 is not enough, cause you have no idea what your error on a sample to sample basis is. (i'm sorry i don't know the proper terminology in english)
100 samples of 400 would be better. |
|
