Diablo® III

Legendary "Drop Rates" Explained (TLDR Warning)

04/18/2014 10:30 PMPosted by Algestes
This is all conjecture though, and I don't think it's really true. No offense. but I was getting so many legs the past two days, averaging 1-4 a rift. and many of them were the elusive ones like ice climbers. RNG is indeed at work with drops.


Yea, and I have obtained 3 Cindercoats in the last week. I still feel like there is something going on behind the scenes.
04/18/2014 10:18 PMPosted by Thacker
It's quite easy to test Vhailor.

Play your highest DPS character and run rifts for 20 hours.

Play your lowest DPS highest toughness character and run rifts for 20 hours.

Compare the legs / hour.....you're not going to find much variance in the numbers what so ever.

My point is this.

Killing 10x as fast DOES NOT = 10x more legs...which points to other factors other than "RNG" effecting the loot drops. The easiest to pin the blame on? A time based system.


Again, whether I did or didn't, there's not much one can say based on that. It's been shown that one can examine particular patterns in a phone book and spell out major events in history, but that doesn't mean the phone book is rigged.

You're greatly missing the statistical principle. Small time windows like 20 hours say absolutely nothing about how a RNG system works. Nothing whatsoever. You need far more data than one person can amass personally (unless they work for years on the problem).

Even IF such a pattern were uncovered, I find it hard to believe that it couldn't be explained by Blizzard's admitted timer which starts to boost your Legendary chances after a certain amount of time. People killing quickly will make use of this timer far less frequently than people who kill slowly. That likely explains the difference right there.

Moreover, your system would imply that the forum moderators were outright lying when they let slip the drop percentage boosts in the first few levels of Torment. Under a theory such as yours, monsters could not in any way have drop percentages, else they would be susceptible to the same abuse as chests. I doubt the moderators would outright lie. Omit, sure, but not lie.
It's crazy how much sense this makes, and the more I think about it, the more logical it seems from a business standpoint. Dirty, but we'll never be able to prove it as there will probably never be an official statement about such a system.
Edited by illy#1454 on 4/18/2014 10:37 PM PDT
04/18/2014 10:33 PMPosted by Fatcat
Probably one of the best posts I have ever seen explain something in a game. Every point you made makes 100% sense. Give this man a Medal.


Thanks man. I didn't proof read and just copied / pasted from word... I might elaborate more if this continues to gain interest, along with doing some more detailed experimentation.
04/18/2014 09:10 PMPosted by Bizi
i read...and then i went and put on my tin foil hat


It have been confirmed that the game has "failsafe" which steps in and ensure that nobody is getting completely screwed by rng.
" We added a system in the expansion that tracks the amount of time you spend fighting creatures without finding a legendary and after a certain period of time will slowly start increasing the legendary drop rate."
http://us.battle.net/d3/en/forum/topic/12472447421?page=2#34

A issue with such a system is that drops seems to come in same intervals w/e you play on Torment 5 or torment 1 - simply because the dropchance in between those "guarenteed" drops not is high enough to override the failsafe.
A better solution would be to have the failsafe kick in shorter intervals on the higher torments.
04/18/2014 10:30 PMPosted by Algestes
This is all conjecture though, and I don't think it's really true. No offense. but I was getting so many legs the past two days, averaging 1-4 a rift. and many of them were the elusive ones like ice climbers. RNG is indeed at work with drops.


glad you got all those good loot in such a short span. Meanwhile 2 days ago I got a slur of legs between 1-3 rifts and it all ended up being trash loot. After that I was lucky to get a trash every 5 rifts or not at all. Been two days now and I get upwards of 10+ rifts with nothing, then getting a two hander. Feels great that you can be screwed to such an extent.
04/18/2014 10:35 PMPosted by Vhailor

Moreover, your system would imply that the forum moderators were outright lying when they let slip the drop percentage boosts in the first few levels of Torment. Under a theory such as yours, monsters could not in any way have drop percentages, else they would be susceptible to the same abuse as chests. I doubt the moderators would outright lie. Omit, sure, but not lie.


Hmm, what if the "drop percentages" pertain only to the TYPE of legendary that would drop, not the actual number and frequency of legendary drops?
04/18/2014 10:20 PMPosted by EarlyD3Buyer
Good post, wonder if Blizzard will answer


LOL

Blizzard would never partake in a thread such as this.

They only ever say things like "Under the hood" or "behind the scenes" when it comes to things like this, so they don't have to divulge in factual information on what type of system is in place.

Blizzard will treat this post like the Crusader forums, like the plague and avoid it at all costs.
I think the drop percentages pertain more to their timer than an actual "drop percentage"

100% bonus, sounds like their timer or "legendary meter" fills up faster in those difficulties as opposed to out.
04/18/2014 10:39 PMPosted by Myou
04/18/2014 10:30 PMPosted by Algestes
This is all conjecture though, and I don't think it's really true. No offense. but I was getting so many legs the past two days, averaging 1-4 a rift. and many of them were the elusive ones like ice climbers. RNG is indeed at work with drops.


glad you got all those good loot in such a short span. Meanwhile 2 days ago I got a slur of legs between 1-3 rifts and it all ended up being trash loot. After that I was lucky to get a trash every 5 rifts or not at all. Been two days now and I get upwards of 10+ rifts with nothing, then getting a two hander. Feels great that you can be screwed to such an extent.


Not all of it was good loot. just had a lot of ice climbers... like three pairs lol. but a lot of it was all unusable stuff like shields\2handers.
I completely agree OP. I noticed the same too and have mentioned it a few times around here and friends in my clan. I have also seen others mention the exact same thing. I know its not completely RNG and others can continue shouting it, they mean nothing..so.
Honestly, it doesn't matter if Blizzard responds to this or not. You can't expect much more than general PR from them.... I don't lose any respect for them in that regard either. From a business standpoint you can't expect anything less.

My entire point was, you can go nuts trying to completely maximize your build and be as fast as possible... play 10 hours straight and get carpal tunnel and everything else... but aside from the extra legendaries you get simply from more time played... Johnny casual can come in here, play T1.... click all his skills and kill at 1/10 your rate and end up with similar gear and have drops just as often.
04/18/2014 10:35 PMPosted by Vhailor
04/18/2014 10:18 PMPosted by Thacker
It's quite easy to test Vhailor.

Play your highest DPS character and run rifts for 20 hours.

Play your lowest DPS highest toughness character and run rifts for 20 hours.

Compare the legs / hour.....you're not going to find much variance in the numbers what so ever.

My point is this.

Killing 10x as fast DOES NOT = 10x more legs...which points to other factors other than "RNG" effecting the loot drops. The easiest to pin the blame on? A time based system.


Again, whether I did or didn't, there's not much one can say based on that. It's been shown that one can examine particular patterns in a phone book and spell out major events in history, but that doesn't mean the phone book is rigged.

You're greatly missing the statistical principle. Small time windows like 20 hours say absolutely nothing about how a RNG system works. Nothing whatsoever. You need far more data than one person can amass personally (unless they work for years on the problem).

Even IF such a pattern were uncovered, I find it hard to believe that it couldn't be explained by Blizzard's admitted timer which starts to boost your Legendary chances after a certain amount of time. People killing quickly will make use of this timer far less frequently than people who kill slowly. That likely explains the difference right there.

Moreover, your system would imply that the forum moderators were outright lying when they let slip the drop percentage boosts in the first few levels of Torment. Under a theory such as yours, monsters could not in any way have drop percentages, else they would be susceptible to the same abuse as chests. I doubt the moderators would outright lie. Omit, sure, but not lie.


I've taken several inferential stat classes. A very small sample size will be very similar to the population. However, you are right: the more data you have, the closer to the true population mean you get. But you are wrong in saying that a small sample size can not be representative of the population. All you need is a sample size of at least 10 random samples of 5 out of, let's say, a billion samples to be accurate within a .05 or .01 chance...Cause it can, and it is. And it is what most science is based on. Much with confidence interval padding, however.

To figure out if this is all true, find 10 random characters who can clear t1 on avg of 20min and find 10 random characters who clear it in 10min. Then, compare the two sample means (means being avg legs per min). you will find this to be very accurate to the mean of the whole population...you would also find it very similar to data of only 5 random samples of each. The only time you really would run into big problems is if you ran samples below 5 in this particular case.

In any case, from my experiences, I don't need any of this data. It is true. It's bs too.
Edited by AXELBRAIN#1891 on 4/18/2014 11:05 PM PDT
04/18/2014 10:44 PMPosted by ReckITT
I completely agree OP. I noticed the same too and have mentioned it a few times around here and friends in my clan. I have also seen others mention the exact same thing. I know its not completely RNG and others can continue shouting it, they mean nothing..so.


I just have no idea why Blizzard went from letting players chose any item combination they wanted anytime they wanted via the AH and trading... to completely locking us into what we find with the new BOA system.

Today on my barb I did happen to find 3 set pieces. All from different sets.

I would much rather have found 3 of the same set, so I could start working on actually using their set bonuses.

Better yet, I would have liked to have traded the other two I found for the ones I wanted.

We need to be able to make the decision, not shoehorned into the path the drop system decides we take.
I'm not sure I agree, but kudos for the thought-provoking post. What you suggest does make sense, but I also think what Vhailor said is correct, you need a much larger data pool to really validate your position.
Posted by Fatcat
Probably one of the best posts I have ever seen explain something in a game. Every point you made makes 100% sense. Give this man a Medal.
04/18/2014 10:15 PMPosted by Vhailor
Why on earth would Blizzard implement such a system, and then not disclose it?


Same EXACT reason I can roll THIRTY times on BOOTS for reduced level requirement, and never get it.

30 times. 6-ish attributes? 0/30? please, tell me again at how random that is.
Cheers Op,

me and my friends allready noticed this too, we are joking about "legendary timer" more than week allready. Dont think there is something like "legendary classes" that dont let you loot better ones often, but i would put my hand in fire for time spend / legendary control.
04/18/2014 10:45 PMPosted by Thacker
Johnny casual can come in here, play T1.... click all his skills and kill at 1/10 your rate and end up with similar gear and have drops just as often.


at least now I don't feel so pressured to tier up on my crusader now... Ill just plunk around t1, t2 and be happy if that's the truth.
Your trying to find a pattern in a random loot system. Just because this has been your experience does not scream conspiracy. I have had lucky streaks and have found multiple upgrades in 1 day. I have had dry streaks were I find nothing but 2 handers and junk legendaries. You need to look into how the loot system works, how magic find works.
This topic has reached its post limit. You may no longer post or reply to posts for this topic.

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)

Reported!

[Close]