Hey whazup guys!
In the last 2 weeks i’ve been collecting 60 RoRG for a statistical experiment about the chance to get a specific affix! We are not talking about 1000 RoRG, but I think that a full inventory is a good amount and a bit indicative for how much “is all about RNG”…
As you can see from the video I got 4 ancient RoRG out of 60 so chance is pretty low. While farming rings I’ve noticed higher chances to get ancient version of other items, like 10 amulets, 12 bracers/weapon, etc.
We are always talking about RNG, but let me say that I find it very weird…. even considering all the affixes I got!
Vitality: 19
Red. Resource Cost: 8
Area Dmg: 8
Average Dmg: 6
CDR: 5
Life%: 4
Crit Hit Dmg: 3
Life per Second: 2
All Resistance: 2
Armor: 1
Socket: 1
Crit chance: 1
4 rings out of 60 rolled wrong primary
From these results I feel it’s so weird that the most wanted affixes (ex. CDR, socket, CC,CHD) rolled so few times compared to others.
Insane ratio for Vitality… maybe Blizz have put a higher chance because is a more casual gamer friendly affix (??) I don’t know, I hope it’s not… I hope “it’s all about RNG” as they keep saying, but let me say that everything is so strange..
Ofc some of you may have found a good rolled one in very few attempts, but pls keep in mind that Blizz policy has interest in to stuck people at farming over and over, so RoRG could be their best option considering how mandatory it is..
That’s all, feel free to share your opinion!
I hope you guys are enjoying the game despite RNG ^^
Mainstat and Vitality always had the highest weight during rerolling at mystic for any type of an item. I assume that the same weight coefficients are used when the items are generated, so getting that much vitality on rings is understandable.
And, don't forget that there is a chance (15% I think) that the item will roll completely random - e.g. wrong main stat. Of course, class specific rare and legendary items still won't roll a mainstat of different class (you can still find blue wizard hats with dexterity, though).
yeah RNG doesnt mean you have the same chance for everything .. so if there are 10 possible affixes, there is not a 1/10 th chance for each affix to roll.
Its a weightet stat distribution system.
Blizzard knows which are the most desired stats, and they make them rarer than e.g vitality.
However 60 is a very smal sample size to actually get some good analysis of this weighted system..
From what i see you are completly missing Life on Hit - whereas whenever i get a rorg it has life on hit..
so maybe try to get 600 ?
anyway thanks for your dedication. every bit of data is good
Btw guys, that's the point... there should not be different weights, or at least remove default affixes (IAS,LoH in that case) and make a differentiation between defensive, offensive and utility affixes to roll 1 each (just an idea, i prefer socketed trifectas :D)
Btw guys, that's the point... there should not be different weights, or at least remove default affixes (IAS,LoH in that case) and make a differentiation between defensive, offensive and utility affixes to roll 1 each (just an idea, i prefer socketed trifectas :D)
The weighting of stats is extremely important for itemization. IF everything were equally likely you'd see items with (insert useless primary) as often as items with (whatever primary is currently best). That would royally screw itemization progression in a dozen different ways.
A 60 sample size is FAR too tiny to get meaningful data on affix weighting anyway. You came pretty close to the 10% Ancient chance (1/12 compared to 1/10) at least.
Blizzard made this Rorg easy to get, but pain in the ass to get the right stats.
Farming an RorG for about 5 Months now, i looted 27 or more.
Not ONE with cc/cd/socket .
But its fair , u get the nice secondary affix pretty easy but a good one is... Pain in the ass!!!
I still hate rng , because 27 unusable rorg is bad luck.
In 10 RoRGs I found 3 with sockets and 3 with Crit Chance. RNg is RNG
yeah RNG doesnt mean you have the same chance for everything .. so if there are 10 possible affixes, there is not a 1/10 th chance for each affix to roll.Its a weightet stat distribution system.Blizzard knows which are the most desired stats, and they make them rarer than e.g vitality.However 60 is a very smal sample size to actually get some good analysis of this weighted system..From what i see you are completly missing Life on Hit - whereas whenever i get a rorg it has life on hit..so maybe try to get 600 ? :Danyway thanks for your dedication. every bit of data is good
60 of his RoRGs had LoH. It's guaranteed and therefore omitted form his little table here.
Edit: I have some choice words ot say to whoever decided on the post formatting system here. Seriously.
As you can see from the video I got 4 ancient RoRG out of 60 so chance is pretty low. While farming rings I’ve noticed higher chances to get ancient version of other items, like 10 amulets, 12 bracers/weapon, etc.
The chance for an item to be ancient is 10% as stated by blizzard. I dont know what you hope to achieve with you miniscule sample size.
On a total of 250 bags i have found 12x anc for each kind of items but 4 rorgs, it COULD (like whole topic) means that your dear 10% is bs for that ring, do u see know? COULD
If you feel 60 is not enough data to discuss then go farm some more or just go to another topic a full inventory of RoRG takes time to be done, so considering you won't find more data about RoRG, 60 are at least enough to DISCUSS and share some ideas, i'm not stating the truth
When the expansion first came out, I ran approximately 1600 Act I Bounties in the first 40 hours (that's about 320 Caches). Some split farmed, some solo. Some in Normal, some in Master, some in T1/T2 (T6 was out of the question that early in the game). By the time the first month passed, I did around 3000 Bounties.
Not a single RoRG rolled CHC or CHD. Not sure how many rolled sockets, but back then legendary gems didn't exist so it wasn't a desired stat.
Didn't get one with CHC until June 2014.
Now with the guaranteed legendary drop per Cache on T6... too easy.
I lost count on how many RoRGs with CHC/CHD or sockets I've gotten since they made the change. One of the first RoRGs I got since 2.1.2 was a socketed ancient one, too.
Somtimes I just happy I'm so lucky with stuff like RoRG. I could not stand doing thousands of bounties and getting nothing when it normally takes me an afternoon to have perfect crit stat and whatever my other stat of choice is.
But don't be sad Jaetch. I'd played months on my non-seaosn 'sader and still have yet to find a BoP, FotF, or Furnace. We all have stuff we're unlucky with.
Hey OraK, thanks for sharing your findings, it was a lot of work and I enjoy seeing what kind of stats tend to roll on items. For myself, I've probably only got about a dozen RoRGs across all my characters, and I think one rolled CC, one with a socket, and one with CDR... maybe?
You said you only got 4 ancients out of 60, but the stated chance of getting an ancient by blizzard is 10%. Of course, this 10% rate "COULD" be wrong, but your experiment does not necessarily support this. Like others are saying, an N of 60 is much, much too low to make meaningful conclusions. To illustrate, lets do an experiment with dice. If you assume a 10% chance of getting an ancient, that's like rolling a 10 sided die and getting a roll of exactly 10 (or any other specific number, I'm just choosing 10 arbitrarily). So if you rolled 60 of these 10 sided dice, you'd EXPECT to see six 10's because 6/60 is 10%. When you simulate this, you do not get six 10's every time, because RNG. Here is an example where I simulated rolling a 10 sided die, sixty times, and then repeated this ten times, and counted the number of 10's that came up each time:
Counts of how many 10's appeared for each set of 60 dice rolls: 9, 4, 8, 7, 6, 9, 5, 5, 3, 4
Your result, 4 out of 60 ancients, appeared twice. The expected number, 6 out of 60, appeared once. The range was quite wide too, as high as 9 out of 60 (15%) and as low as 3 out of 60 (5%). However, if you add up all the dice rolls there were exactly 60 out of 600 appearances of 10, which is 10%! This seems quite lucky as I'd expect to be slightly off 10% even with 600 trials, but I guess that's RNG for you.
This doesn't prove anything about diablo 3's drop chances, it just shows that low sample sizes are expected to give highly variable results. So getting 4 ancients out of 60 legendaries is entirely consistent with a 10% drop chance. It doesn't prove it is 10% and doesn't prove that it's not 10%, but it is one result you would expect.
In the last 2 weeks i’ve been collecting 60 RoRG for a statistical experiment about the chance to get a specific affix! We are not talking about 1000 RoRG, but I think that a full inventory is a good amount and a bit indicative for how much “is all about RNG”…
So here is the video
https://www.youtube.com/watch?v=YSCMP6K3jJE
As you can see from the video I got 4 ancient RoRG out of 60 so chance is pretty low. While farming rings I’ve noticed higher chances to get ancient version of other items, like 10 amulets, 12 bracers/weapon, etc.
We are always talking about RNG, but let me say that I find it very weird…. even considering all the affixes I got!
Vitality: 19
Red. Resource Cost: 8
Area Dmg: 8
Average Dmg: 6
CDR: 5
Life%: 4
Crit Hit Dmg: 3
Life per Second: 2
All Resistance: 2
Armor: 1
Socket: 1
Crit chance: 1
4 rings out of 60 rolled wrong primary
From these results I feel it’s so weird that the most wanted affixes (ex. CDR, socket, CC,CHD) rolled so few times compared to others.
Insane ratio for Vitality… maybe Blizz have put a higher chance because is a more casual gamer friendly affix (??) I don’t know, I hope it’s not… I hope “it’s all about RNG” as they keep saying, but let me say that everything is so strange..
Ofc some of you may have found a good rolled one in very few attempts, but pls keep in mind that Blizz policy has interest in to stuck people at farming over and over, so RoRG could be their best option considering how mandatory it is..
That’s all, feel free to share your opinion!
I hope you guys are enjoying the game despite RNG ^^
OraK
www.zeroempathy.net
www.zeroempathy.net (EU)
My Youtube: https://www.youtube.com/channel/UCqhcpLBbZeLwilfRVHA0eZA
Era 3 - NS, #1 Solo WD
Era 2 - NS, #1 4man, #5 2man, #7 Solo WD
Era 1 - NS, #3 2man
And, don't forget that there is a chance (15% I think) that the item will roll completely random - e.g. wrong main stat. Of course, class specific rare and legendary items still won't roll a mainstat of different class (you can still find blue wizard hats with dexterity, though).
Its a weightet stat distribution system.
Blizzard knows which are the most desired stats, and they make them rarer than e.g vitality.
However 60 is a very smal sample size to actually get some good analysis of this weighted system..
From what i see you are completly missing Life on Hit - whereas whenever i get a rorg it has life on hit..
so maybe try to get 600 ?
anyway thanks for your dedication. every bit of data is good
Btw guys, that's the point... there should not be different weights, or at least remove default affixes (IAS,LoH in that case) and make a differentiation between defensive, offensive and utility affixes to roll 1 each (just an idea, i prefer socketed trifectas :D)
www.zeroempathy.net (EU)
My Youtube: https://www.youtube.com/channel/UCqhcpLBbZeLwilfRVHA0eZA
Era 3 - NS, #1 Solo WD
Era 2 - NS, #1 4man, #5 2man, #7 Solo WD
Era 1 - NS, #3 2man
A 60 sample size is FAR too tiny to get meaningful data on affix weighting anyway. You came pretty close to the 10% Ancient chance (1/12 compared to 1/10) at least.
In 10 RoRGs I found 3 with sockets and 3 with Crit Chance. RNg is RNG
60 of his RoRGs had LoH. It's guaranteed and therefore omitted form his little table here.
Edit: I have some choice words ot say to whoever decided on the post formatting system here. Seriously.
http://eu.battle.net/d3/en/profile/Twoflower-2131/hero/47336841
Best you can get is probably socket / ias / loh / chd and then reroll LOH to cc.
http://eu.battle.net/d3/en/profile/Twoflower-2131/hero/47336841
If you feel 60 is not enough data to discuss then go farm some more or just go to another topic a full inventory of RoRG takes time to be done, so considering you won't find more data about RoRG, 60 are at least enough to DISCUSS and share some ideas, i'm not stating the truth
www.zeroempathy.net (EU)
My Youtube: https://www.youtube.com/channel/UCqhcpLBbZeLwilfRVHA0eZA
Era 3 - NS, #1 Solo WD
Era 2 - NS, #1 4man, #5 2man, #7 Solo WD
Era 1 - NS, #3 2man
Best you can get is Socket and reroll AS or Mainstat to CC/CHD
Edit: Or any derivative of one stat and reroll.
You guys.
When the expansion first came out, I ran approximately 1600 Act I Bounties in the first 40 hours (that's about 320 Caches). Some split farmed, some solo. Some in Normal, some in Master, some in T1/T2 (T6 was out of the question that early in the game). By the time the first month passed, I did around 3000 Bounties.
Not a single RoRG rolled CHC or CHD. Not sure how many rolled sockets, but back then legendary gems didn't exist so it wasn't a desired stat.
Didn't get one with CHC until June 2014.
Now with the guaranteed legendary drop per Cache on T6... too easy.
I lost count on how many RoRGs with CHC/CHD or sockets I've gotten since they made the change. One of the first RoRGs I got since 2.1.2 was a socketed ancient one, too.
Armory | YouTube | Twitter | Clan Site
But don't be sad Jaetch. I'd played months on my non-seaosn 'sader and still have yet to find a BoP, FotF, or Furnace. We all have stuff we're unlucky with.
You said you only got 4 ancients out of 60, but the stated chance of getting an ancient by blizzard is 10%. Of course, this 10% rate "COULD" be wrong, but your experiment does not necessarily support this. Like others are saying, an N of 60 is much, much too low to make meaningful conclusions. To illustrate, lets do an experiment with dice. If you assume a 10% chance of getting an ancient, that's like rolling a 10 sided die and getting a roll of exactly 10 (or any other specific number, I'm just choosing 10 arbitrarily). So if you rolled 60 of these 10 sided dice, you'd EXPECT to see six 10's because 6/60 is 10%. When you simulate this, you do not get six 10's every time, because RNG. Here is an example where I simulated rolling a 10 sided die, sixty times, and then repeated this ten times, and counted the number of 10's that came up each time:
10 - 1 - 5 - 2 - 10 - 9 - 8 - 2 - 7 - 5 - 2 - 2 - 1 - 7 - 2 - 1 - 6 - 5 - 8 - 10 - 10 - 6 - 5 - 7 - 3 - 10 - 6 - 4 - 1 - 2 - 2 - 2 - 8 - 4 - 2 - 10 - 1 - 1 - 9 - 9 - 1 - 6 - 6 - 5 - 8 - 8 - 2 - 2 - 8 - 8 - 9 - 9 - 10 - 7 - 3 - 4 - 6 - 9 - 10 - 10
3 - 4 - 1 - 6 - 4 - 1 - 5 - 8 - 5 - 3 - 1 - 6 - 6 - 9 - 9 - 4 - 1 - 1 - 9 - 6 - 8 - 5 - 7 - 4 - 3 - 3 - 2 - 1 - 1 - 8 - 4 - 2 - 2 - 10 - 7 - 7 - 6 - 10 - 2 - 8 - 3 - 5 - 8 - 5 - 7 - 4 - 4 - 2 - 9 - 10 - 4 - 6 - 10 - 4 - 3 - 9 - 8 - 7 - 3 - 1
9 - 4 - 7 - 7 - 10 - 9 - 4 - 7 - 2 - 2 - 8 - 1 - 10 - 2 - 3 - 3 - 7 - 2 - 7 - 10 - 10 - 1 - 3 - 2 - 5 - 8 - 3 - 8 - 7 - 8 - 10 - 9 - 8 - 1 - 9 - 9 - 9 - 1 - 6 - 9 - 10 - 7 - 10 - 7 - 2 - 3 - 7 - 8 - 2 - 10 - 2 - 7 - 1 - 3 - 2 - 9 - 2 - 5 - 3 - 2
7 - 7 - 10 - 5 - 3 - 10 - 4 - 9 - 4 - 4 - 5 - 7 - 6 - 9 - 1 - 3 - 1 - 7 - 10 - 2 - 1 - 5 - 6 - 8 - 7 - 4 - 9 - 5 - 10 - 10 - 8 - 4 - 9 - 4 - 2 - 3 - 8 - 6 - 4 - 3 - 9 - 9 - 5 - 8 - 5 - 6 - 1 - 2 - 3 - 10 - 8 - 5 - 6 - 1 - 1 - 2 - 1 - 2 - 8 - 10
1 - 10 - 10 - 4 - 6 - 1 - 9 - 3 - 7 - 8 - 6 - 6 - 3 - 7 - 3 - 3 - 10 - 9 - 3 - 1 - 4 - 7 - 3 - 1 - 2 - 3 - 9 - 8 - 3 - 1 - 3 - 6 - 8 - 5 - 2 - 10 - 9 - 2 - 2 - 5 - 8 - 8 - 5 - 5 - 4 - 3 - 3 - 9 - 8 - 9 - 2 - 3 - 10 - 8 - 7 - 5 - 8 - 10 - 2 - 3
3 - 1 - 10 - 7 - 6 - 10 - 8 - 8 - 5 - 4 - 5 - 9 - 7 - 9 - 4 - 6 - 3 - 8 - 7 - 10 - 1 - 10 - 10 - 5 - 9 - 10 - 7 - 1 - 9 - 8 - 3 - 9 - 10 - 9 - 3 - 7 - 9 - 9 - 10 - 5 - 6 - 5 - 4 - 5 - 4 - 4 - 7 - 2 - 4 - 1 - 2 - 10 - 4 - 3 - 3 - 3 - 2 - 5 - 9 - 6
4 - 8 - 10 - 8 - 2 - 6 - 10 - 8 - 7 - 8 - 7 - 5 - 9 - 2 - 5 - 6 - 6 - 6 - 1 - 5 - 4 - 5 - 7 - 8 - 6 - 3 - 6 - 7 - 7 - 10 - 6 - 4 - 10 - 4 - 1 - 8 - 6 - 7 - 4 - 9 - 5 - 9 - 5 - 7 - 5 - 8 - 2 - 2 - 9 - 7 - 8 - 6 - 4 - 4 - 3 - 9 - 1 - 2 - 2 - 10
10 - 3 - 1 - 7 - 3 - 3 - 5 - 6 - 2 - 7 - 10 - 9 - 3 - 3 - 1 - 5 - 1 - 1 - 4 - 9 - 1 - 4 - 9 - 10 - 3 - 10 - 1 - 7 - 1 - 2 - 8 - 5 - 6 - 4 - 9 - 2 - 9 - 9 - 5 - 2 - 4 - 7 - 1 - 3 - 8 - 9 - 4 - 2 - 4 - 8 - 9 - 10 - 8 - 4 - 9 - 6 - 5 - 2 - 5 - 3
3 - 9 - 5 - 10 - 9 - 6 - 4 - 9 - 5 - 9 - 7 - 2 - 7 - 2 - 5 - 5 - 4 - 6 - 3 - 4 - 6 - 10 - 4 - 4 - 3 - 8 - 7 - 3 - 6 - 3 - 8 - 7 - 4 - 2 - 9 - 6 - 3 - 1 - 4 - 2 - 6 - 6 - 7 - 5 - 2 - 7 - 4 - 6 - 9 - 5 - 3 - 9 - 10 - 2 - 8 - 4 - 6 - 2 - 7 - 2
9 - 1 - 3 - 2 - 1 - 10 - 9 - 5 - 4 - 10 - 2 - 4 - 3 - 1 - 9 - 1 - 3 - 9 - 2 - 7 - 5 - 1 - 4 - 1 - 1 - 10 - 8 - 3 - 5 - 6 - 1 - 2 - 4 - 2 - 4 - 5 - 3 - 5 - 4 - 2 - 9 - 9 - 3 - 9 - 3 - 9 - 6 - 9 - 4 - 7 - 2 - 7 - 10 - 1 - 8 - 8 - 8 - 2 - 2 - 2
Counts of how many 10's appeared for each set of 60 dice rolls: 9, 4, 8, 7, 6, 9, 5, 5, 3, 4
Your result, 4 out of 60 ancients, appeared twice. The expected number, 6 out of 60, appeared once. The range was quite wide too, as high as 9 out of 60 (15%) and as low as 3 out of 60 (5%). However, if you add up all the dice rolls there were exactly 60 out of 600 appearances of 10, which is 10%! This seems quite lucky as I'd expect to be slightly off 10% even with 600 trials, but I guess that's RNG for you.
This doesn't prove anything about diablo 3's drop chances, it just shows that low sample sizes are expected to give highly variable results. So getting 4 ancients out of 60 legendaries is entirely consistent with a 10% drop chance. It doesn't prove it is 10% and doesn't prove that it's not 10%, but it is one result you would expect.
thx +1