It is widely believed that the classes that fight up close tend to get more drops. As an elementalist, I can fight up close, or at quite a distance. For a long time I've strongly felt that I get more drops when I'm up close. I then went about testing my theory that close combat gets more drops.
To test this theory, I recorded all of my battles when in a group of human players. For a battle, I counted only those frays where we were fighting mobs of 3 or more creatures grouped together. I then recorded 20 battles from a far distance, using stuff like flare and fireball, at such a distance that I was pretty much never damaged by the creatures. Then I recorded 20 battles where I was up front taking damage using close up skills like inferno and close range fireball etc.
Out of the 20 battles where I was far away, I earned 8 drops.
Out of the 20 battles where I was up front taking the hits, I earned 17 drops.
I know that this is in no way flawless as far as scientific guidelines allow, but I did have some controls and saw quite a noticeable difference. I'm sure many of you will try to refute me and this experiment, but that's fine, I just wanted to test my feelings that close-up combat granted me more drops.
What are your reactions and ideas?
Drop Theory Experiment
Uldrath
Lasareth
Up those battles to 100 and then we'll be talking sample size reliability :P
Uldrath
Hehe yeah yeah I know, but recording 200 battles didn't seem too appealinng.
Lasareth
Nope but it's possible you had a streak in that 20 sample run. There DOES appear to be a difference, now it just needs to be further tested :P
Synncial77
Well, when you have the time, repeat the same sample (20 of each). Either that will draw a pretty clear picture or it will muddy things up a bit.
ManadartheHealer
You also should make sure it is in the exact same area(s).
velvetbunny
Drops are random, plain and simple. Nothing you do in a party will influence what drops get assigned to you. Having it assign items based on anything other then a random number would be silly. Imagine if what your implying were true, now imagine how many monks/ele's would then begin to try and tank in parties and get up close and or take damage just to try and get a drop assigned to them. Do you really think the designers would intend for that to happen? I certainly do not.
Vwoss
There are A LOT of people that can testify with the same results.
Tuna
I know if you are a certain distance away you have a 0% chance of getting a drop.
Dumachum
My warrior recieves 3 times more and better drops then my ranger or monk. That is a fact.
DrSLUGFly
yeah,
It's a clever experiment. And doing it out of a hundred doesn't mean that you have to do 100 in a row with the same people (ideally the group should be the same).
I propose that you get like 10 friends who are keen on the experiment and cover all your controls this way: (perhaps I'm a little too ambitious on this)
(with all friends fighting and zoning into the exact same area, each with a team of the exact same henchmen)
friend 1 - 10 times ranged ele, 10 times up close ele
friend 2 - 10 times ranged ranger, 10 times up close ranger
friend 3 - 10 times ranged warrior (secondary), 10 time melee warrior
friend 4 - 10 times ranged mesmer, 10 times attacking mesmer
friend 5 - 10 times ranged necro w/ minions, 10 times attacking necro
the healer
friend 6 - 10 times smiting monk, 10 times healing monk
friend 7 - 10 times healing monk, 10 times melee monk
the control
friend 8 - 20 times random class, normal playstyle (control)
friend 9 - 20 times random class, normal playstyle, not zoning to fight same mobs (control)
friend 10 - 20 times random class, normal playstyle, not zoning and using a very different map
To be totally appropriate it would be best if all characters were the same level too... likewise you could try a level 10 taking on level 10 mobs and a level 20 taking on level 10 mobs... but I doubt that this would make much difference.
It's a clever experiment. And doing it out of a hundred doesn't mean that you have to do 100 in a row with the same people (ideally the group should be the same).
I propose that you get like 10 friends who are keen on the experiment and cover all your controls this way: (perhaps I'm a little too ambitious on this)
(with all friends fighting and zoning into the exact same area, each with a team of the exact same henchmen)
friend 1 - 10 times ranged ele, 10 times up close ele
friend 2 - 10 times ranged ranger, 10 times up close ranger
friend 3 - 10 times ranged warrior (secondary), 10 time melee warrior
friend 4 - 10 times ranged mesmer, 10 times attacking mesmer
friend 5 - 10 times ranged necro w/ minions, 10 times attacking necro
the healer
friend 6 - 10 times smiting monk, 10 times healing monk
friend 7 - 10 times healing monk, 10 times melee monk
the control
friend 8 - 20 times random class, normal playstyle (control)
friend 9 - 20 times random class, normal playstyle, not zoning to fight same mobs (control)
friend 10 - 20 times random class, normal playstyle, not zoning and using a very different map
To be totally appropriate it would be best if all characters were the same level too... likewise you could try a level 10 taking on level 10 mobs and a level 20 taking on level 10 mobs... but I doubt that this would make much difference.
Dusk_
Simple fact: people remember the bad, and pass the good off as the norm.
When you actually try to prove this theory properly, you'll find the results to be less drastic. As people said before, test it more times. 20 is nothing.
Also, I'll note that you based your drops on the number of battles, not enemies. Your results will be seriously thrown off if you ended up killing more enemies close up.
When you actually try to prove this theory properly, you'll find the results to be less drastic. As people said before, test it more times. 20 is nothing.
Also, I'll note that you based your drops on the number of battles, not enemies. Your results will be seriously thrown off if you ended up killing more enemies close up.
Kazahana
i dont think drops is based on dmg or proximity(unless your far out of range) at all i never do any dmg to creatures cause i play a healer with no attack skills and i still get pretty good drops
Uldrath
Quote:
|
Originally Posted by DrSLUGFly
yeah,
..... I propose that you get like 10 friends who are keen on the experiment and cover all your controls this way: (perhaps I'm a little too ambitious on this) (with all friends fighting and zoning into the exact same area, each with a team of the exact same henchmen) |
Also, just to note, all the battles I recorded were against enemies approximately between 1-2 lvls below me and many levels higher than me, since they were all in the same area.
Lasareth
The best controlled experiment is to go out with two people--each can solo on their own. Have one solo all of them, then go yourself and solo all of them. Both of you stay together the entire time. Tally drop rate for yourself when done.
Keep the sample level and types of creatures constant.
This way, the ONLY variable is who is hitting them, and you have a superior basis for deduction.
Er, of course, this isn't your hypothesis, lol.
For your real hypothesis, have the same two-person setup but have one person kill up-close the entire time, while you vary distance.
Keep the sample level and types of creatures constant.
This way, the ONLY variable is who is hitting them, and you have a superior basis for deduction.
Er, of course, this isn't your hypothesis, lol.
For your real hypothesis, have the same two-person setup but have one person kill up-close the entire time, while you vary distance.
Ikinsey
Enemies or areas are not things you need to control. They are irrelevant. Different monsters dont change the odds of who gets drops. The only control you must have is classes, the classes of the people in your party must be the same. It would also be wise to keep constant the type of player: NPC, or PC because theoretically the chance of getting a drop could change between the two. Also I agree a higher number of recorded drops would be nice, but not necisarilly as important as some others make it seem. The best thing you could do to make this a very reputable experiment would be to do several tests in the same incriment. Lets say you tested the drops in the number of 20 as you did. Do the same thing 5 more times if the results are far in one direction everytime you may have somehting.
Cunning
I find it highly unrealistic that the developers would have actually made the drops biased like that.
I reckon the drops are infact random, pure and simple.
I reckon the drops are infact random, pure and simple.
DrSLUGFly
good point.
So rally up a team of 4 war/monks. Get 1 to always tank and do dmg, get another to always hit ranged dmg (smite), get the third to always support (heal) and the fourth to always hang back and watch. Then see what the numbers look like after 20 or 50 groups of mobs.
So rally up a team of 4 war/monks. Get 1 to always tank and do dmg, get another to always hit ranged dmg (smite), get the third to always support (heal) and the fourth to always hang back and watch. Then see what the numbers look like after 20 or 50 groups of mobs.
Uldrath
Well my theory was that distance mattered, not class. This theory arose from my experience being a class that could either fight up close or far away. I'm sure that they would not bias it on class, for that would be unfair, but I thought that there might be some programming thing that gave preference first to those who were closest to the enemy.
This would not be a strict bias, for essentially anyone could be close to the enemy. Warriors just happen to be closest most often.
This would not be a strict bias, for essentially anyone could be close to the enemy. Warriors just happen to be closest most often.
Alderman Sweet
If drop rate is dependent on proximity, Alesia should be getting most of the drops.
JoDiamonds
I know I saw somewhere (manual? official website?) that you aren't considered to be fighting a monster unless certain conditions are met. If you don't meet the conditions, you aren't fighting that monster, and you can't get drops from that monster.
I believe that the conditions are something like:
1- The monster is in your Danger Zone (as named in the manual)
2- The monster is in the Danger Zone of an ally that you heal / buff / etc.
If that's more or less true, then if a monster is fighting a warrior who is a bit out of your range (not uncommon), then likely that monster is never "fighting" you, and therefore you aren't eligible for its item drops.
This follows closely with what Uldrath has said, but I'm pretty sure I've seen it written down in some kind of official docs somewhere (trying to find it, failing).
Sadly, this would mean that Warriors are just more likely to get drops than the guys in the back, though strangely enough healers are likely to get the drops since they heal everyone all over the place.
I believe that the conditions are something like:
1- The monster is in your Danger Zone (as named in the manual)
2- The monster is in the Danger Zone of an ally that you heal / buff / etc.
If that's more or less true, then if a monster is fighting a warrior who is a bit out of your range (not uncommon), then likely that monster is never "fighting" you, and therefore you aren't eligible for its item drops.
This follows closely with what Uldrath has said, but I'm pretty sure I've seen it written down in some kind of official docs somewhere (trying to find it, failing).
Sadly, this would mean that Warriors are just more likely to get drops than the guys in the back, though strangely enough healers are likely to get the drops since they heal everyone all over the place.
Sereng Amaranth
Quote:
|
Originally Posted by Alderman Sweet
If drop rate is dependent on proximity, Alesia should be getting most of the drops.
|
The Ages
Quote:
| If drop rate is dependent on proximity, Alesia should be getting most of the drops |
Algren Cole
Quote:
|
Originally Posted by velvetbunny
Drops are random, plain and simple. Nothing you do in a party will influence what drops get assigned to you. Having it assign items based on anything other then a random number would be silly. Imagine if what your implying were true, now imagine how many monks/ele's would then begin to try and tank in parties and get up close and or take damage just to try and get a drop assigned to them. Do you really think the designers would intend for that to happen? I certainly do not.
|
Sekkira
Quote:
|
Originally Posted by velvetbunny
Drops are random, plain and simple. Nothing you do in a party will influence what drops get assigned to you. Having it assign items based on anything other then a random number would be silly. Imagine if what your implying were true, now imagine how many monks/ele's would then begin to try and tank in parties and get up close and or take damage just to try and get a drop assigned to them. Do you really think the designers would intend for that to happen? I certainly do not.
|
I also agree that it is based on distance at some points as I've once gotten stuck in the stairs when doing the divinity coast and eventually there were no drops for me ever.
Paladin_Adoni
this is how I see the loot sytem working :
more damage dealt = higher chance of loot
more damage taken = higher chance of loot
more hits landed = higher chance of loot
closer range = higher chance of loot
low number of drops so far = higher chance of loot
basically each person at the point when a monster dies has this information tallied, then randoms between the people who have had a low amount of loot so far, but have a high "number" fromt he other factors.
the obvious mold breaker is a healer. who deal no damage. in there case, it would most likely take these into account :
more damage stopped = higher chance of loot
more health healed = higher chance of loot
that sort of thing.
nothing to base this on. but its seems to fit (or something similar)
more damage dealt = higher chance of loot
more damage taken = higher chance of loot
more hits landed = higher chance of loot
closer range = higher chance of loot
low number of drops so far = higher chance of loot
basically each person at the point when a monster dies has this information tallied, then randoms between the people who have had a low amount of loot so far, but have a high "number" fromt he other factors.
the obvious mold breaker is a healer. who deal no damage. in there case, it would most likely take these into account :
more damage stopped = higher chance of loot
more health healed = higher chance of loot
that sort of thing.
nothing to base this on. but its seems to fit (or something similar)
Sentinel
When you are doing experiments you need to be unbiased. Do the experiment
and take a look at the results after. The original poster had already a "feeling" and therefore he was biased. No offence but I can't trust his results one bit.
Also when random numbers are involved then statistics come out to play and you need to run the same experiment over and over to start getting an idea of what is going one. I'm afraid 20 times is just plain inadequate. Even 100 times is not enough to write it in stone. I mean look at the real world. Scientists take thousants of samples in order to formulate an opinion about something.
Back to the game. With my W I was always complaining that I don't get any drops in FoW or UW. Some days were better than others but the bad days I remember the most. After I accepted it's random somehow the drops got better.
It was all in my mind?? YOU BET!!lol
With my monk i stay the hell out of range of the nasties while I stay close enough to heal and my drops don't suffer comparing to anyone.
Just to disprove the argument once and for all I invite you guys to come to one of my farming spots. You stay at the spawn while I go out, deep in the map, and kill 30-40 nasties. According to some of you I should get most of the drops.... say, more than half... and you guys at the spawn you'll be lucky to get anything. I highly doubt it but with you guys there checking on me I can't be biased.
Sent.-
and take a look at the results after. The original poster had already a "feeling" and therefore he was biased. No offence but I can't trust his results one bit.
Also when random numbers are involved then statistics come out to play and you need to run the same experiment over and over to start getting an idea of what is going one. I'm afraid 20 times is just plain inadequate. Even 100 times is not enough to write it in stone. I mean look at the real world. Scientists take thousants of samples in order to formulate an opinion about something.
Back to the game. With my W I was always complaining that I don't get any drops in FoW or UW. Some days were better than others but the bad days I remember the most. After I accepted it's random somehow the drops got better.
It was all in my mind?? YOU BET!!lol
With my monk i stay the hell out of range of the nasties while I stay close enough to heal and my drops don't suffer comparing to anyone.
Just to disprove the argument once and for all I invite you guys to come to one of my farming spots. You stay at the spawn while I go out, deep in the map, and kill 30-40 nasties. According to some of you I should get most of the drops.... say, more than half... and you guys at the spawn you'll be lucky to get anything. I highly doubt it but with you guys there checking on me I can't be biased.
Sent.-
JoDiamonds
If you are nowhere near the enemies being killed, you have a 0% chance of getting a drop from it. Sentinel, I'd be happy to join you for this test. I know I'll get zero drops once you are far enough away.
I'm not certain of anything else said in this thread, but many, many people have pointed out that if you are really far away, you just don't get drops. Which makes sense.
I'm not certain of anything else said in this thread, but many, many people have pointed out that if you are really far away, you just don't get drops. Which makes sense.
Algren Cole
Quote:
|
Originally Posted by Paladin_Adoni
this is how I see the loot sytem working :
more damage dealt = higher chance of loot more damage taken = higher chance of loot more hits landed = higher chance of loot closer range = higher chance of loot low number of drops so far = higher chance of loot basically each person at the point when a monster dies has this information tallied, then randoms between the people who have had a low amount of loot so far, but have a high "number" fromt he other factors. the obvious mold breaker is a healer. who deal no damage. in there case, it would most likely take these into account : more damage stopped = higher chance of loot more health healed = higher chance of loot that sort of thing. nothing to base this on. but its seems to fit (or something similar) |
Sentinel
Quote:
|
Originally Posted by JoDiamonds
If you are nowhere near the enemies being killed, you have a 0% chance of getting a drop from it. Sentinel, I'd be happy to join you for this test. I know I'll get zero drops once you are far enough away.
I'm not certain of anything else said in this thread, but many, many people have pointed out that if you are really far away, you just don't get drops. Which makes sense. |
Well.... Let's try it then. I will be interesting for sure. Let's round up a few others and let's do it. I'm in eastern time and can run this any time after 6pm est tonight. Let's do it!
Sent.-
Paladin_Adoni
Quote:
|
Originally Posted by Algren Cole
I can assure you that's not how it works...that would kill the server in 32.7981 seconds flat.
|
:P
I did say or something similar. the idea was not to take it WORD for WORD. but a type of system that takes this approach.
oh and it would be 32.7984 seconds thank you very much
Algren Cole
Quote:
|
Originally Posted by Paladin_Adoni
oh and it would be 32.7984 seconds thank you very much
|
lol...forgive me, my math has always been poor
the lag you experience is probably(if it isn't graphic)based on the fact that ArenaNet almost assuredly not using enough servers to host the gamers...hosting is EXPENSIVE...and with no monthly fee providing it for free is difficult...I'm sure the botters don't help much either seeing as they are constantly taxing the system. Bots are never not doing anything...MANY MANY players spend alot of time doing nothing...
Paladin_Adoni
its definately not my system, built for maximum doom 3 and half life 2 graphics play.
the lag is the number 1 reason I dont play PvP 1/4 of a second lag is a BAD thing.
back on topic. I get CRAP loot in groups all the time. almost never anything good. the good stuff I get solo. and im a warrior
the lag is the number 1 reason I dont play PvP 1/4 of a second lag is a BAD thing.
back on topic. I get CRAP loot in groups all the time. almost never anything good. the good stuff I get solo. and im a warrior
Diomedes
I've seen a lot of arguments flying around and it might be helpful to try and condense a few of these together.
First, I think it's a big problem that the OP used groups and not individuals. If some groups had 10 monsters and others only 5 then there will be big problems. For the rest of my post, I'm going to pretend that he used monsters and not groups, that way if the experiment is repeated properly (i.e. counting monster kills, not groups) what I say can still apply with just plug in the new numbers.
So we have 2 groups, near (N) and far (F). The frequency for drops for N is 17/20 = .85, the frequency for drops for F is 8/20 = .4
u = sum(x*(f/n))
So here's the table
Near Group
Drop outcome(x)| Measured Probability p(x) | Calculation of mean xp(x)
no drop = 0 | .15 | 0
got a drop = 1 | .85 | .85
u = .85
Far Group
no drop = 0 | .6 | 0
got a drop = 1 | .4 | .4
u = .4
Calculation of sample variance = sum((x-u)^2)*p(x)
Near group
(x-u)^2 | (x-u)^2*p(x)
.7225 | .1084
.0225 | .0191
s^2 = .1275
Far Group
(x-u)^2 | (x-u)^2*p(x)
.16 | .096
.36 | .144
s^2 = .24
(more to come)....
First, I think it's a big problem that the OP used groups and not individuals. If some groups had 10 monsters and others only 5 then there will be big problems. For the rest of my post, I'm going to pretend that he used monsters and not groups, that way if the experiment is repeated properly (i.e. counting monster kills, not groups) what I say can still apply with just plug in the new numbers.
So we have 2 groups, near (N) and far (F). The frequency for drops for N is 17/20 = .85, the frequency for drops for F is 8/20 = .4
u = sum(x*(f/n))
So here's the table
Near Group
Drop outcome(x)| Measured Probability p(x) | Calculation of mean xp(x)
no drop = 0 | .15 | 0
got a drop = 1 | .85 | .85
u = .85
Far Group
no drop = 0 | .6 | 0
got a drop = 1 | .4 | .4
u = .4
Calculation of sample variance = sum((x-u)^2)*p(x)
Near group
(x-u)^2 | (x-u)^2*p(x)
.7225 | .1084
.0225 | .0191
s^2 = .1275
Far Group
(x-u)^2 | (x-u)^2*p(x)
.16 | .096
.36 | .144
s^2 = .24
(more to come)....
Diomedes
now we can difference the means and create a confidence interval.
(X1 - X2) +/- t.025*(s1^2/n1 + s2^2/n2)^(.5)
(.85-.4) +/- 2.02*(.1275/20 + .24/20)^.5 = .45 +/- 0.273819923
Now some people claim that there is no difference between the means, but that does not fall within a 95% confidence interval.
Hence to everyone stating that you can't tell anything from a sample of 40 observations, actually you can.
However in this case, it needs to be repeated with monsters, not groups.
Anyhow, good luck guys, this is sort of a crude statistical method, but it is valid.
-Diomedes
(X1 - X2) +/- t.025*(s1^2/n1 + s2^2/n2)^(.5)
(.85-.4) +/- 2.02*(.1275/20 + .24/20)^.5 = .45 +/- 0.273819923
Now some people claim that there is no difference between the means, but that does not fall within a 95% confidence interval.
Hence to everyone stating that you can't tell anything from a sample of 40 observations, actually you can.
However in this case, it needs to be repeated with monsters, not groups.
Anyhow, good luck guys, this is sort of a crude statistical method, but it is valid.
-Diomedes
Diomedes
In case the math made your eyes glaze over, let me summarize:
(A) What the OP did was NOT too small of a sample size
(B) The OP or someone should repeat the exercise with monsters, not groups
(C) If that was monsters, not groups, you can reject the hypothesis that the two samples are the same, in other words, IF ALL OTHER THINGS ARE INDEED HELD CONSTANT, then You /DO/ get more drops from being closer.
-Diomedes
(A) What the OP did was NOT too small of a sample size
(B) The OP or someone should repeat the exercise with monsters, not groups
(C) If that was monsters, not groups, you can reject the hypothesis that the two samples are the same, in other words, IF ALL OTHER THINGS ARE INDEED HELD CONSTANT, then You /DO/ get more drops from being closer.
-Diomedes
StormWater
In the words of the noble bard, Samuel Clemmons-
"There are lies, damned lies and statistics."
I'd love to hear from Anet on the general principles for loot dropping and allocation of that loot. The rest of us can hypothesize all we want.
"There are lies, damned lies and statistics."
I'd love to hear from Anet on the general principles for loot dropping and allocation of that loot. The rest of us can hypothesize all we want.
PieXags
I just know that in a 5 man UW smite run today (just with my guild mates, not as a trapper, or anything like that, just people), I didn't get a single drop.
Not one.
I was the only ranger, so I was problably furthest away the whole time.
There are times when we'll go through all but the last fight in a mission and I'll have not gotten a drop. Sometimes I even prepare the text "I'm bound to get a drop one of these days" simply because I'm so unlucky when it comes to drops.
Damn damn damn.
Not one.
I was the only ranger, so I was problably furthest away the whole time.
There are times when we'll go through all but the last fight in a mission and I'll have not gotten a drop. Sometimes I even prepare the text "I'm bound to get a drop one of these days" simply because I'm so unlucky when it comes to drops.
Damn damn damn.
Diomedes
Quote:
|
Originally Posted by StormWater
In the words of the noble bard, Samuel Clemmons-
"There are lies, damned lies and statistics." I'd love to hear from Anet on the general principles for loot dropping and allocation of that loot. The rest of us can hypothesize all we want. |
I would like to hear from ANet too, but I doubt that they'll ever actually make a statement. In the mean time, those who are careful and use numbers wisely can learn a good deal.
-Diomedes
Sekkira
I'm neither here nor there. I would be against someone saying one thing is set in solid proof without stone... err stone without proof (is not thinking).
Anyway, I'd rather hear it from ArenaNet on how all this works.
Anyway, I'd rather hear it from ArenaNet on how all this works.