I've been farming Feywild rares recently, no particular reason other than I worked out a good circuit and it was something to do. I hit about 100 rares across several characters, picked up some sentient food(between ferrocrystal weapons and superfluous minor artifacts, its decent).

Then something odd happened: I found a fey-twisted chest in the wilderness.

Then I found 5 more of them in under an hour.


Now I had heard fey-twisted chests dropped in the wilderness, but didn't really believe it based on sample size. Then I got 6 of them on the same character over the course of about an hour. On a character I hadn't yet set foot in in the feywild before. That's a pretty suspicious bumb. Obviously its statistically possible, but seems unlikely.

noamineo wrote on Mar 27^th, 2022 at 1:35pm:


If I remember right, it’s about 2% chance. We had the rate pretty well nailed down at one time because several were farming and recording data over many trials. I always preferred r10 farming that one quest with the guaranteed chest guarded by water Ellie’s. Can’t remember the name, but it’s the first one one the left from the Wynwood Hall as you enter wilderness. The one with redcaps guarding the gate. But you can bypass those, go left in the water and under. You can spot the cheat as Feytwisted or not before engaging any enemies at all. IF it’s Feytwisted, you can either solo the water Ellie’s OR post a party to come loot it and help kill them.

I did it enough times (hundreds) to be pretty sure R10 had better rates. So, that quest on r10 was the most reliable method I found. I tried to keep one piece of every twisted gear. I managed to get 3/4 of them, I think.

Each attempt was 30-45 seconds. Avg 1 twisted per 20-30 attempts = 1 chest per 10-20 mins.

Head-Meat wrote on Mar 29^th, 2022 at 9:40am:


So I've done about 70 trips into Make Believe on various reaper levels and still not seen that chest manifest as Feytwisted. If I hadn't have heard it from the Head-Meat's mouth I wouldn't believe the chest could feytwist. Going to do another 30-40 this evening when work lets out. This is still a more reliable way to farm feytiwsts than anything else.

Here's another anecdote: I did a round of R10 Feywild maybe a month ago. I think we ended up seeing 1-2 Fey chests during that round. Meaning we did the saga or whatever.

a 2% chance does not mean you will get one every 50.  It means that there is a 2% chance each time independent of the time before and the time after. 

devs have posted about their rng long long ago claiming it was good according to their tests.  everything you post about clumping is just so much bollocks you simply aren't seeing consecutive rolls of the rng, other people are getting rng rolls between your rolls. Clumping of results IS normal for an rng

Did you test 10000 times?  or did you have like 100, oh look, you have 100 from the root message.

Did you test 10000 consecutive results or was it a random 100 results over many thousands (i.e. serverwide) over which you have no idea how many. 

Your input sample is so meaningless that you cannot draw any meaningful results from it.

try this
create a spreadsheet. 
Cell A1 = Randbetween(1, 100), drag down to create 100 results.
Cell A100 = =COUNTIF(A1:A100, 1)
drag that range it across for 100 columns
gets you how many time 1 comes up, good quick estimate of 1%


Here's the 100 results I got from the countif row
1      1      2      0      0      1      0      0      0      1      0      1      0      5      1      2      1      3      1      1      0      2      0      2      1      0      1      1      0      5      1      1      4      0      1      1      1      1      0      0      0      0      0      2      2      3      0      0      0      2      1      4      2      2      0      0      2      1      1      1      3      1      3      4      1      1      0      2      1      1      3      1      1      1      1      1      1      1      0      0      3      0      1      0      1      1      1      0      0      0      2      1      1      3      4      2      1      1      1      2

Total number of times 1 happened is 118, it should be 100, and that's over 10000 results.  That's way outside of 1% 
There is a two columns with 5 in.  5 times in 100, with a 1% chance, clumping!
There is a group of 5 adjacent columns with only 0's in.  500 with not a single 1!
There is a group of 7 adjacent columns with only 1 1 in very smooth distribution, suspicious! 

Try it yourself and look at the data you get.

what you are seeing in your data is a human finding patterns in randomness

Head-Meat wrote on Mar 29^th, 2022 at 9:40am:


I asked a statustician friend for help and we came up with about 3,000 runs to get a 95% certainty this number is correct(with a 0.005% margin for error). At 45 seconds a run that'll take 37 hours. I don't think I'll proving this rate any time soon. In fact I think I'll go play some dodgeball instead.

Head-Meat wrote on Mar 29^th, 2022 at 9:40am:


Been doing R10s for a bit and not seeing anywhere close to this. But, as we've determined, a very large sample size is required and I'm getting bored.

Standing Stone Games wrote on Apr 1^st, 2022 at 1:42pm:


This is the DDO Vault. Not a spelling bee.

noamineo wrote on Apr 1^st, 2022 at 1:57pm:

I see. Googling is too hard for you, then.

You should blame your parents.

noamineo wrote on Mar 31^st, 2022 at 10:46pm:


Just don't block with your head too much.