Now before I say anything; I know D3, like many other looters before it, is based in RNG. But how much of this is really RNG?
I ask the members of diablofans.com to put some of the numbers they may have tracked (hell, ill take rough estimates if you gottem). I know in my experiance and ALL of my friends and randoms I have encoutered all seem to say the same thing 'rigged game is rigged'. And I would herald the RNG horn all day long. Until the numbers I started tracking when I hit inferno got me to notice a thing or two.
My numbers have been averaged with friends of mine that have trying to track. The sample size is me and 8 others that reported
LOOT: I started tracking most dropped affixes when i was about 50, thinking that early gear didnt matter much and at this time was finaly swayed to track what I see when I could (98% of the time). And with no suprise throns/healthglobe were an easy 1st and 2nd either one showing on loot 68% of the time and both 41%. Again no suprise, but then something happened with loot about 1 week ago.
1.5 weeks ago: 1st, thorns/HG 68%....6th, resist all 17%
This week alone: 3rd, resist all 34%....thorns and globe around the same
My friends and I were thinking quick % hotfix on afffix chance of resist all to help players get into inferno easier. Again small sample size but the increase in resist all was too obviouse.
ELITE PACKS: I play a DH and figured I would split the already small sample size even smaller to two class sets, melee and ranged. Also numbers were only tracked while playing solo or with other like class sets. (IE double ranged players ect.)
again like the loot before it, there seemed to be a change a little over a week ago.
1.5 weeks ago: Ranged 1ST, mortor 2ND jailer 3RD teleport 4TH reflect dmg, LAST PLACE missle damp
Melee 1st Motlen, 2ND jailer, 3RD knockback, 2ND TO LAST PLACE REFLECT DMG
This week alone Ranged, 1ST REFLECT DMG, 2ND electric, 3rd immune minions
This week alone Melee, 1ST knock back, 2ND molten, 3rd jailer
seems that they learned we just outrange the shit out of mortor (shown in the response from the blizz staff on reddit) and it was nothing close to the ranged killer.
I dont really like to be the one to start a conspiracy thread. But numbers dont lie, even on a small sample size like this one you can see an attempt to shifting trends, making glass gannons blow themselves up for starters, or really locking down on what fucks a melee in the ass and turning the dial to 11. Granted Inferno was ment to be hard, and I can assume thats why I have seen maybe enough missle damps to count on one hand the entire time I have played since launch (been farming act 3 inferno for 2 weeks + now) considering a little feed of tenticles will sit in there and kill it in baout 5 secs. It just looks like they have kinda tuned the mobs and market on the fly. Not that I am totaly against it, more about 99% against it with no notice. Im sure they are just trying to keep the integrity of the 'RNG' at hand by maintaining that it is nothing more than RNG. Granted im sure some people got into inferno act 2 abit quicker than the firsts this week thanks to the flood of resist all but then again maybe thats was just the RNG for my small sample. I just find it hard to belive that we all had 'like' numbers (ranged with ranged and melee with melee). And I know im not the only one that has read an article concerning Blizzard rigging raid loot back in the BC days to encourage replay, and getting caught red handed. (If i find a link regaurding the article I will post here, cant find atm)
So if you got some numbers, post them. If there is an article like this I have missed, please post a link to the forum here please.
Random numbers doesn't mean what you think it means.
Some affixes have a higher drop-rate than others. This is established and expected. One affix dropping more often than another is working as intended. No, I don't have a source, it's an understood principle of statistical distributions. The "source" is the natural logarithm
Secondly, the beauty of random is that literally anything can happen. People have a tendency to believe that if you pick a random number 1-100 lots of times, there will be an equal distribution of all the numbers from 1 to 100. This isn't the way it works. Randomly, the number can be 73 every time. That's still random. Each time you select a new random number, it has an certain chance (usually approximating an equal chance) to be any of the numbers, but that's true every time you select a number. There's no bias away from a number you've already selected, no matter how many times it gets picked. Probabalistically you'll see all the numbers distributed somewhat evenly, in a way, over large samples. Even then: random is random. If it was true that you'd never see more 73's than 61's in a large sample, it wouldn't be random. There would exist within that RNG a bias towards equalized distribution. That's called a "pattern" and it isn't the same as random, it's an even distribution. People have a tendency to assume that "random" means "evenly distributed", but it doesn't. It means random, and that means that heads can come up 1000 times in a row and there's no problem. It might be weird and a little unlikely, but heads is an equally valid "next flip" no matter how many times it has come up heads in the past.
People have trouble getting past this idea, but it's true. Random is random, not perfectly heterogeneous. The two concepts have little to do with one another.
Random numbers doesn't mean what you think it means.
Some affixes have a higher drop-rate than others. This is established and expected. One affix dropping more often than another is working as intended. No, I don't have a source, it's an understood principle of statistical distributions. The "source" is the natural logarithm
Secondly, the beauty of random is that literally anything can happen. People have a tendency to believe that if you pick a random number 1-100 lots of times, there will be an equal distribution of all the numbers from 1 to 100. This isn't the way it works. Randomly, the number can be 73 every time. That's still random. Each time you select a new random number, it has an certain chance (usually approximating an equal chance) to be any of the numbers, but that's true every time you select a number. There's no bias away from a number you've already selected, no matter how many times it gets picked. Probabalistically you'll see all the numbers distributed somewhat evenly, in a way, over large samples. Even then: random is random. If it was true that you'd never see more 73's than 61's in a large sample, it wouldn't be random. There would exist within that RNG a bias towards equalized distribution. That's called a "pattern" and it isn't the same as random, it's an even distribution. People have a tendency to assume that "random" means "evenly distributed", but it doesn't. It means random, and that means that heads can come up 1000 times in a row and there's no problem. It might be weird and a little unlikely, but heads is an equally valid "next flip" no matter how many times it has come up heads in the past.
People have trouble getting past this idea, but it's true. Random is random, not perfectly heterogeneous. The two concepts have little to do with one another.
This topic was more about people posting what they have seen in solo games (or 'liked' grouped). I just find it hard to think mortor left and right at launch, followed by damage reflect left and right (once they figured that ranged found mortor a non issue) and at the same time the easiest by far (missle damp) I have only seen so many times I can count it on one hand.
Sure loot is loot, and I can agree and think it to be obviouse that garbage affixes have a greater chance to drop IE; you tend to see them alot. By that same exstention the game is not flooded with 'god-like' gear and everyone waiting on an exspansion and/or content patch. Cool. But were we really the only ones that saw the flood of res. all this last week or so. Again I wouldnt even have mentioned anything such as this if there wasnt enough people, friend and random alike, taking notice.
But I would say im still very interested in a concept of fixed elite affixes, and probably in the same sense as loot; easy and god like stats- hard to roll, potential pains in the ass and garbage stats- common rolled.
Ok, random is random, and 9 is not a sample size to tell *anything* from.
OTOH, Blizzard *can* tweak the distribution/probability of affixes showing up at whatever level they wish; i.e. it's still random, but they can put limits on it or change the probabilities. However, the OP's post is not near enough data to say yes, they tweaked it, or no, they didn't. And so what if they did? (or didn't?) That's just part of managing things to have more happy players and a better RoI for the game.
5 million players, this guy tries to calculate rng by 9 of them.
GG with that
Nerdfact: For large populations, accuracy mostly just depends on the sample size. A sample of 10,000 is pretty much the same accuracy for a population of 1 million or 1 billion. On the other hand, you still need a lot more than 9.
Random numbers doesn't mean what you think it means.
Some affixes have a higher drop-rate than others. This is established and expected. One affix dropping more often than another is working as intended. No, I don't have a source, it's an understood principle of statistical distributions. The "source" is the natural logarithm
Secondly, the beauty of random is that literally anything can happen. People have a tendency to believe that if you pick a random number 1-100 lots of times, there will be an equal distribution of all the numbers from 1 to 100. This isn't the way it works. Randomly, the number can be 73 every time. That's still random. Each time you select a new random number, it has an certain chance (usually approximating an equal chance) to be any of the numbers, but that's true every time you select a number. There's no bias away from a number you've already selected, no matter how many times it gets picked. Probabalistically you'll see all the numbers distributed somewhat evenly, in a way, over large samples. Even then: random is random. If it was true that you'd never see more 73's than 61's in a large sample, it wouldn't be random. There would exist within that RNG a bias towards equalized distribution. That's called a "pattern" and it isn't the same as random, it's an even distribution. People have a tendency to assume that "random" means "evenly distributed", but it doesn't. It means random, and that means that heads can come up 1000 times in a row and there's no problem. It might be weird and a little unlikely, but heads is an equally valid "next flip" no matter how many times it has come up heads in the past.
People have trouble getting past this idea, but it's true. Random is random, not perfectly heterogeneous. The two concepts have little to do with one another.
This. I was going to post a whole response about how you are confusing "random" with "equal probability", but this post covers it.
You do realize that for the elite pack affixes, Blizzard only said it was randomly generated. No where did it also state that all affixes are equal in distribution.
And since we don't know the formula being used for RNG, you're assuming every affix has an equally random chance to occur. The truth is that an aspect of the formula might differ based on the the first letter of the day, or the current server time. The science of RNG (especially in a game like diablo) is not as simple as rolling a dice.
As others have pointed out it is possible to roll the exact same number 20 times in a row.
What? Missile dampening? Am I the only one that hasn't seen this affix before?
I saw it the first time after about 60 hours of play time (late inferno act 1 on my first char) so I think its safe to say that this one is extremely rare. It might be an inferno only affix, I dont know but I had still been farming the rare in a cave that is needed for some quest for many hours.
I saw missile dampening many times and as far back as nightmare. Cool affix really. The biggest thing is random is random. We tend to look for patterns where none exist. And if you take a large enough sample size you will always see an uneven distribution. Thats just the way it works in real life.
Also we don't know the algorithm for generating items. It could be a crap shoot or there could actually be higher % chance for some affixes. without Blizzard saying how it works we are just guessing.
Am I the only one who actually read the post and saw him ask for more data? "I see this trend, what have you guys seen so we might confirm or deny it"? Everyone bashing him on the stats, you could have just as easily mentioned your experiences, or gone into the game and done a couple of siegebreaker runs, keeping track of what affixes you got so we could begin to form a data set of the proper magnitude.
Seriously, yelling about stats 101 when you don't even understand the scientific method, which I'm fairly sure they still teach in grade school, is kind of pathetic.
Am I the only one who actually read the post and saw him ask for more data? "I see this trend, what have you guys seen so we might confirm or deny it"? Everyone bashing him on the stats, you could have just as easily mentioned your experiences, or gone into the game and done a couple of siegebreaker runs, keeping track of what affixes you got so we could begin to form a data set of the proper magnitude.
Seriously, yelling about stats 101 when you don't even understand the scientific method, which I'm fairly sure they still teach in grade school, is kind of pathetic.
I dont really like to be the one to start a conspiracy thread. But numbers dont lie, even on a small sample size like this one you can see an attempt to shifting trends.
Statements like these don't help his case. I don't feel the need to throw in any data considering we have no idea how affixes are distributed in mode, level, monsters, item type etc… If you want serious consideration he could create a template and have people add data using the same template… possibly post in theory crafting section might help too.
Rollback Post to RevisionRollBack
Playing Diablo since 97. I know nothing and having nothing good to say, I be a troll.
What? Missile dampening? Am I the only one that hasn't seen this affix before?
It's extremely rare - but it looks very cool and the effect is kinda neat to look at. Possibly the most rare to pop up. On some mobs, it'd make the creature near unkillable. On a fast mob that likes to get your back, you'd never be able to hit it. While on some mobs, it's a who cares affix - like skeletons who stand still all the time.
I've been of the opinion that the game recognizes your class, and tries to build inferno elites to counter you. I feel like I get a lot more "mortar/teleport/vortex/fast" mobs than "plague/molten/nightmare/extra health" mobs as a ranged player.
Statements like these don't help his case. I don't feel the need to throw in any data considering we have no idea how affixes are distributed in mode, level, monsters, item type etc… If you want serious consideration he could create a template and have people add data using the same template… possibly post in theory crafting section might help to.
Then stay out of the thread. He obviously came here to see what other people were experiencing. We all know random is random, but a lot of us have noticed trends as well. Just because he doesn't have 10k samples doesn't mean what he's seen doesn't indicate that a trend exists.
I love how everyone here is more or less being dickish to the OP, yes he only has a sample size of 9 at the moment but if you read his last statement it seem he was not necessary looking for opinions on if he is right but TRYING to get a larger sample size by asking others if they have noticed anything or got a larger percentage of one particular affix or not. Help him if you want but leave your nonconstructive criticisms off and let the topic die if you don't like it.
As for you OP, I dont have any numbers but I have to say As a Barb player I seem to notice way more jailer,plagued and desecrater than anything else. It would be interesting to see if they are manipulating the RNG based on the class.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I ask the members of diablofans.com to put some of the numbers they may have tracked (hell, ill take rough estimates if you gottem). I know in my experiance and ALL of my friends and randoms I have encoutered all seem to say the same thing 'rigged game is rigged'. And I would herald the RNG horn all day long. Until the numbers I started tracking when I hit inferno got me to notice a thing or two.
My numbers have been averaged with friends of mine that have trying to track. The sample size is me and 8 others that reported
LOOT: I started tracking most dropped affixes when i was about 50, thinking that early gear didnt matter much and at this time was finaly swayed to track what I see when I could (98% of the time). And with no suprise throns/healthglobe were an easy 1st and 2nd either one showing on loot 68% of the time and both 41%. Again no suprise, but then something happened with loot about 1 week ago.
1.5 weeks ago: 1st, thorns/HG 68%....6th, resist all 17%
This week alone: 3rd, resist all 34%....thorns and globe around the same
My friends and I were thinking quick % hotfix on afffix chance of resist all to help players get into inferno easier. Again small sample size but the increase in resist all was too obviouse.
ELITE PACKS: I play a DH and figured I would split the already small sample size even smaller to two class sets, melee and ranged. Also numbers were only tracked while playing solo or with other like class sets. (IE double ranged players ect.)
again like the loot before it, there seemed to be a change a little over a week ago.
1.5 weeks ago: Ranged 1ST, mortor 2ND jailer 3RD teleport 4TH reflect dmg, LAST PLACE missle damp
Melee 1st Motlen, 2ND jailer, 3RD knockback, 2ND TO LAST PLACE REFLECT DMG
This week alone Ranged, 1ST REFLECT DMG, 2ND electric, 3rd immune minions
This week alone Melee, 1ST knock back, 2ND molten, 3rd jailer
seems that they learned we just outrange the shit out of mortor (shown in the response from the blizz staff on reddit) and it was nothing close to the ranged killer.
I dont really like to be the one to start a conspiracy thread. But numbers dont lie, even on a small sample size like this one you can see an attempt to shifting trends, making glass gannons blow themselves up for starters, or really locking down on what fucks a melee in the ass and turning the dial to 11. Granted Inferno was ment to be hard, and I can assume thats why I have seen maybe enough missle damps to count on one hand the entire time I have played since launch (been farming act 3 inferno for 2 weeks + now) considering a little feed of tenticles will sit in there and kill it in baout 5 secs. It just looks like they have kinda tuned the mobs and market on the fly. Not that I am totaly against it, more about 99% against it with no notice. Im sure they are just trying to keep the integrity of the 'RNG' at hand by maintaining that it is nothing more than RNG. Granted im sure some people got into inferno act 2 abit quicker than the firsts this week thanks to the flood of resist all but then again maybe thats was just the RNG for my small sample. I just find it hard to belive that we all had 'like' numbers (ranged with ranged and melee with melee). And I know im not the only one that has read an article concerning Blizzard rigging raid loot back in the BC days to encourage replay, and getting caught red handed. (If i find a link regaurding the article I will post here, cant find atm)
So if you got some numbers, post them. If there is an article like this I have missed, please post a link to the forum here please.
And happy RNG
Some affixes have a higher drop-rate than others. This is established and expected. One affix dropping more often than another is working as intended. No, I don't have a source, it's an understood principle of statistical distributions. The "source" is the natural logarithm
Secondly, the beauty of random is that literally anything can happen. People have a tendency to believe that if you pick a random number 1-100 lots of times, there will be an equal distribution of all the numbers from 1 to 100. This isn't the way it works. Randomly, the number can be 73 every time. That's still random. Each time you select a new random number, it has an certain chance (usually approximating an equal chance) to be any of the numbers, but that's true every time you select a number. There's no bias away from a number you've already selected, no matter how many times it gets picked. Probabalistically you'll see all the numbers distributed somewhat evenly, in a way, over large samples. Even then: random is random. If it was true that you'd never see more 73's than 61's in a large sample, it wouldn't be random. There would exist within that RNG a bias towards equalized distribution. That's called a "pattern" and it isn't the same as random, it's an even distribution. People have a tendency to assume that "random" means "evenly distributed", but it doesn't. It means random, and that means that heads can come up 1000 times in a row and there's no problem. It might be weird and a little unlikely, but heads is an equally valid "next flip" no matter how many times it has come up heads in the past.
People have trouble getting past this idea, but it's true. Random is random, not perfectly heterogeneous. The two concepts have little to do with one another.
D3 Channel: OnetwoD3
^. This. +1 one. What he said. Ect, ect.
Sure loot is loot, and I can agree and think it to be obviouse that garbage affixes have a greater chance to drop IE; you tend to see them alot. By that same exstention the game is not flooded with 'god-like' gear and everyone waiting on an exspansion and/or content patch. Cool. But were we really the only ones that saw the flood of res. all this last week or so. Again I wouldnt even have mentioned anything such as this if there wasnt enough people, friend and random alike, taking notice.
But I would say im still very interested in a concept of fixed elite affixes, and probably in the same sense as loot; easy and god like stats- hard to roll, potential pains in the ass and garbage stats- common rolled.
BUT WTF!? WTB MORE MISSLE DAMP PACKS
OTOH, Blizzard *can* tweak the distribution/probability of affixes showing up at whatever level they wish; i.e. it's still random, but they can put limits on it or change the probabilities. However, the OP's post is not near enough data to say yes, they tweaked it, or no, they didn't. And so what if they did? (or didn't?) That's just part of managing things to have more happy players and a better RoI for the game.
Nerdfact: For large populations, accuracy mostly just depends on the sample size. A sample of 10,000 is pretty much the same accuracy for a population of 1 million or 1 billion. On the other hand, you still need a lot more than 9.
If every affix had equal weighting, the item hunt wouldn't last nearly as long.
This. I was going to post a whole response about how you are confusing "random" with "equal probability", but this post covers it.
Get a sample of 1000 and then apply a margin of error, until then take your grade 3 math else where, thanks.
And since we don't know the formula being used for RNG, you're assuming every affix has an equally random chance to occur. The truth is that an aspect of the formula might differ based on the the first letter of the day, or the current server time. The science of RNG (especially in a game like diablo) is not as simple as rolling a dice.
As others have pointed out it is possible to roll the exact same number 20 times in a row.
I saw missile dampening many times and as far back as nightmare. Cool affix really. The biggest thing is random is random. We tend to look for patterns where none exist. And if you take a large enough sample size you will always see an uneven distribution. Thats just the way it works in real life.
Also we don't know the algorithm for generating items. It could be a crap shoot or there could actually be higher % chance for some affixes. without Blizzard saying how it works we are just guessing.
Seriously, yelling about stats 101 when you don't even understand the scientific method, which I'm fairly sure they still teach in grade school, is kind of pathetic.
Statements like these don't help his case. I don't feel the need to throw in any data considering we have no idea how affixes are distributed in mode, level, monsters, item type etc… If you want serious consideration he could create a template and have people add data using the same template… possibly post in theory crafting section might help too.
I've been of the opinion that the game recognizes your class, and tries to build inferno elites to counter you. I feel like I get a lot more "mortar/teleport/vortex/fast" mobs than "plague/molten/nightmare/extra health" mobs as a ranged player.
Then stay out of the thread. He obviously came here to see what other people were experiencing. We all know random is random, but a lot of us have noticed trends as well. Just because he doesn't have 10k samples doesn't mean what he's seen doesn't indicate that a trend exists.
As for you OP, I dont have any numbers but I have to say As a Barb player I seem to notice way more jailer,plagued and desecrater than anything else. It would be interesting to see if they are manipulating the RNG based on the class.