VaultHunter101
(So long, and thanks for all the fish)
#1
I’ve been trying to figure out the probability that you’ll get a higher level item out of the grinder. It’s obviously easier to find white and green items to grind, so there are more of those in the stats than blue and purple. If you want to contribute data, just post in the thread below, indicating at least the initial rarity and the number of increased rarity to total grinds.
Update (2018/06/22):
I’ve split the data out to take into account the recent patch that increased legendary drop rates. First table is data with the new rates, while the one below is the combined data from before the patch (and before the drop rates experiment.) I’ve also included the data from @decimus28umo.
Running totals for non-moonstone grinds, post-patch
Type
Increased rarity
Total grinds
Success rate
White
770
965
79.8%
Green
653
1068
61.1%
Blue
298
755
39.5%
Purple
28
616
4.5%
Orange
13
13
100.0%
OOP
19
54
35.2%
Glitch
26
26
100.0%
Note 2: So far, three glitch or three orange rarity items always seem to give you the same rarity back; the numbers on these are pretty low still though (39/39). This seems comparable to three orange items.
Running totals for non-moonstone grinds pre-patch, including data from @BuddyPharaoh and @Lance_Chaser:
White: 273 / 344 = 79.4% success rate
Green: 246 / 409 = 60.1% success rate
Blue: 72 / 164 = 43.9% success rate
Purple: 7 / 133 = 5.3% success rate
Note 1: I’m fairly confident at this stage that it makes no real difference to the rarity of the item produced whether you use three of a kind, or two+one, or even three different items.
I’ve been keeping stats since I first started, but I haven’t been playing the game religiously enough to rack up what I’d call statistically significant numbers (mostly, I only played the grinder enough to get the Master Chef achievements on each character).
In my case, it was broken out by recipe. I can provide that data if you really want, but I don’t think it indicates anything (there’s just too few of each). I didn’t distinguish between 3oaK and 2+1; those were all lumped together. As a result, pretty much all I can supply is the by-rarity figures.
Based on my figures, I was already pretty sure of a flat 80% to get a green from 3 whites, 50% to get a blue from 3 greens, and 40% to get a purple from 3 blues.
You seem to have a significantly higher rate for green>blue, and I’m not sure why. Your blue>purple rate is lower than you report (24 / 79 = 30.4%) - not sure what’s going on there.
Like you, I’ve done nowhere near enough purples to get any meaningful results. No moonstone grinds yet.
1 Like
VaultHunter101
(So long, and thanks for all the fish)
#3
Grinded a backpack full of purple weapons and got one Legendary as a result. And this time I took notes.
All Weapons used were LV50 and no moonstones were used for the grinds.
My backpack consisted of 33 weapons with the following type of weapons used
AR: 5
Pistols: 5
SG: 5
RL: 5
SMG: 5
Snipers: 4
Lasers: 4
So did a total of 11 Grinds to empty my backpack and re-saved the new grinds for future use.
These were my grinding results:
RL1 + SG1 + SMG1 = Purple Laser
RL2 + SG2 + SMG2 = Purple SMG
RL3 + SG3 + SMG3 = Purple SMG
RL4 + SG4 + SMG4 = Purple Sniper
RL5 + SG5 + SMG 5 = Purple Laser
AR1 + Pistol1 + Sniper1 = Purple Pistol
AR2 + Pistol2 + Sniper2 = Purple SMG
AR3 + Pistol3 + Sniper3 = LV50 Trick Shot Maggie
AR4 + Pistol4 + Sniper4 = Purple Pistol
AR5 + Pistol5 + Laser1 = Purple Pistol
Laser2 + Laser3 + Laser4 = Purple Laser
So got lucky and got the Maggie in Grind #8 this time. I’ll say that was a pretty good success rate.
One time I got two legendaries when I grinded a backpack full purples. But sometimes I’ve gone days with no legendaries at all.
VaultHunter101
(So long, and thanks for all the fish)
#6
So I’m currently unsure if the permanent change to legendary drop-rates has any effect on the grinder. As a result, I’ve split my data into pre- and post-patch summaries. I’ll get a bunch more grinding done as and when I can, then update the OP with the gory details…
If you use a DLC2 legendary + random Legendary + purple class of choice = Guaranteed Legendary result. You have to have a DLC2 legendary in your recipe for a guaranteed legendary without using moonstones. You actually can’t use moonstones when you have a DLC2 legendary weapon as an ingredient.
update: this seems to have been taken away. I haven’t grinded in a while, but I tried my (what I thought was full proof) recipe of DLC2 + random legendary + purple class of choice no longer gives a guaranteed legendary. My jaw dropped when I got a purple as a result. LOL so, my many many grinds with said recipe resulting in legendary output is no longer full proof.
not sure if I was extremely lucky with all of my grinds(we’re talking close to a 100 or more since the DLC2 release) was extremely good luck(ha ha, not likely) or they modified something recently with the grinder(much more likely). sad day…LOL
1 Like
VaultHunter101
(So long, and thanks for all the fish)
#8
Update: After a long break, I’ve been playing more TPS. I’ve added more data post-hotfix to increase legendary drop rates, and included the first set of results for non-moonstone grinding of oramge-orange-purple combinations.
1 Like
Amazing_DKSWW
("...take off your pants and go hog wild.")
#9
I got a Maggie from grinding 3 random purples last night.
What I did for testing was grind blue starting bullpups into purple shotguns as a moonstone grind( cost 0 moonstones). With the resulting purples (regular grind of course) I got the 9 legendary shotguns. 2 of the 9 were exact duplicates. Used the starting and ending challenge numbers to keep track. A second character brought and drop the weapons for use, took most of the night but got some good data. Also I can probably go back and find the lunshine numbers for the blue moonstone grinds with before and after backup save challenge numbers later.
1 Like
VaultHunter101
(So long, and thanks for all the fish)
#13
Thanks for the info. I’ll add the purple stats in with mine next time I update the OP.
1 Like
VaultHunter101
(So long, and thanks for all the fish)
#14
Updated to include @decimus28umo data, as well as the first batch of orange-orange-purple grinds.
