IF the vault key drop is 7%, then i finished with 7.25 % with 300 dropped keys.
IF the vault key drop is supposed to be 10%… it’s not.
My Epic drop, which should be 10%, only made 8.6 percent with 26 epics.
Are you counting your epic keys for vault key drop rate as well? Initially, I was not and I think it makes a difference in the long run. Also are you excluding music gnomes (since they cannot drop vault keys by design)?
Let’s say that you have seen 110 gnomes, which 10 of them were music gnomes, and you got 9 regular and 1 epic vault key. Then (I believe), you should compute the rates as:
- total number of eligible gnomes: 110 - 10 = 100
- total keys: 9 + 1 = 10, key rate: 10 / 100 = 10%
- total epic keys: 1. Epic key rate: 1/10 = 10%
Note that epic key rate is also computed over the total number of keys you got, not related to gnomes. In other words, if the key drop rate is 8%, then 10% epic key drop rate (over this) would give you 8% * 10% = 0.8% drop rate for the epic keys.
No, because technically, they are a separate entity. Although, I can understand the argument for including them in the count.
Gotcha. I just mentioned this because at the beginning I was doing the same and given the same feedback. I think the way the game works (I understand very little about these so someone else feel free to correct me).
- Gnome is dead. It rolls a random number. You have X% to get “any” key.
- If you do get a key, then you have Y% to get an epic (vs. regular).
Nope. VK and EVK have different item IDs. That’s pretty much what caused EVKs to almost not show up in frist 1-2 Vault weekends after their introduction → their drop rates did not increase during Vault Weekends. That was fixed.
From that observations I’ve assumed they are having different items ID and that DEVs had to manualy increase EVK drop rates during Vault Weekend.
But I wouldn’t be surprised if drop rates(during VW) where sth like:
X for VK and Y for EVKs,
We also have a piece of information that Y ~ 0.1X (something arround 0.1 chance for dropping EVK instead of VK)
so that cummulative chance for VKs would be 1.1X, where X is unknown. Your data shows that as number of gnomes increased:
8144, ~7%, ~0.9% (567 keys, 76 Ekeys)
???, ~7%,~1%
14112, ~8%, ~0.9% (1113-119= 994 keys, 119 Ekeys
rate of EVK were somewhere between 0.9- 1% and rate of single VKs increased from a little bit under 7%, up to 8.06%
We could say, that based on your data, X is probably ~8 and EVKs Y= ~0.95%, trend was increasing.
I did 3 GaPs during weekend, but didn’t bother checking exact data… (my gnome counter was sth arround 40+/75 → anyone can count it). didn’t check VKs before weekend (probably < 30), but ended up with 95VKs (including those from tracker) and 7 EVKs
Thanks for clarification. It looks like you have a better understanding of internals and likely better than I am with probability and statistics. I will leave the latest values here so that you (or someone else) can crunch them further (if that is desirable):
- Number of regular battles: 6473.
- Number of GaP battles: 4308.
- Number of gnomes: 16691
- Number of gnomes from regular battles (important for verses): 666 (10.29%)
- Number of regular keys (excluding pity keys): 1236 (8.15%)
- Number of epic keys: 149 (0.88%)
- Number of verses from regular battles: 344 total (51.65%)
- Verse distribution: 85/104/72/83
- Number of verses from GaP battles: 17 total (0.1%)
Combining all of these, what I believe:
- Regular battles, gnome appearance rate: 10%
- Regular battles, music gnome appearance rate: 5% (50% of all gnomes)
- Key drop rate: 8% (of all eligible gnomes)
- Epic key drop rate: 0.9% (of all eligible gnomes)
- I expect the verse drops are equal between verses.
In other words, on average:
- You need 10 battles to see 1 gnome.
- You need 20 battles to see a music gnome. 80 battles to collect all verses for a single GaP.
- You need to kill 12.5 gnomes to get a key. Note that these cannot be music gnomes. So, in other words, you need to kill 25 gnomes during regular fights or 12.5 gnomes during GaP fights. In other words, you will see a key drop every 3 GaP battles (give or take) whereas you need ~250 battles on average non-GaP to get a key.
- You need to kill 111 gnomes to get an epic key. Note that these cannot be music gnomes. SO, in other words, you need to kill 222 gnomes during regular fights or 111 gnomes during GaP fights. In other words, you will see an epic key drop every 28 GaP battle where as you need ~2220 battles on average non-GaP to get an epic key.
(some of the statistics of math might be off, please feel free to correct)
You are correct or very close on all of these.
#1 It’s actually 9% (3% outside vault events)
#2 No idea if it’s actually 50% or not but sounds plausible
#3 Correct
#4 Each time a vk drops there is a 10% chance it will be a evk (I remember a dev saying this somewhere sorry to tired to go looking)
#5 We can only hope lol
Salty or Kafka said that, when EVKs were introduced (that EVKs have ~10% chance of droping compared to standard VK drop chance). Later community found out, that EVKs drop rate weren’t increased during Vault Weekend, which was fixed some time ago…
I wouldn’t be surprised if actual drop rates were like 9/100 and 1/100.
IMO, these rates are idependent → there aren’t 2 checks → one for VK and 2nd to see if it’s EVK. Game just rolls either VK or EVK (2 separate idependent items), and both of these items have a greater chance to drop during Vault Weekend.
This post is about the fix and is what we have now.
How did you obtain these numbers?
Did a few more GAPs after my above post. And a few hours of gnome hunting as well.
Finally tally of gnomes defeated: 4437.
Final # of VKs dropped: 325. Final Percentage: .073%
Final # of EVKs dropped: 29. Final Percentage: .089%
And for the “small percentage” crowd: If the drop rate had held at 9% & 10% respectively, then I would have gotten 399 VKs & 39 EVKs. Shorted 74 VKs & 10 EVKs.
.
More importantly, did anyone with a large # of gnomes killed achieve 12% or above drop rate? Or even higher, since I have guildmates that had even worse luck with GAPs than I did.
Number of vk dropped from ELIGIBLE gnomes is the figure you need to pay attention to as the new band gnomes cannot drop a vk. From your 4437 surely there was maybe 100 band gnomes? And the vk drop rate is estimated (from many data collection threads) to be 8%. Not 9%. So actually you are almost right on the average expected. A tiny bit below yes but nothing to complain about.
Also let’s not forget that not long ago we could only farm about 1 vk per hour during a vault event. Now players are farming hundreds per day. We should all be very happy with how vault events currently are.
Try maybe a dozen, at most. Gnomes were very scare outside the GAP. And, I got ZERO keys from outside the GAP as well.
Ok so 4425 gnomes and 354 vks (epic plus normal)
= exactly 8% ironically.
thats pretty good datas bud.
Except it’s been established that Epic drops are a seperate entity. 325/4425 = .073.
By who? Where was this?
Just for fun I worked out the % on my weekend data. Guess what? It was almost the exact same stats as you just a larger sample size.
18,374 gnomes (I estimated about 250 music gnomes and subtracted those from 18,624)
1337 vk plus 141 evk for a total of 1478 (no pity keys included obviously)
1337/18374 = 7.28% (practically same as your 7.3%)
Guess what happens when I include evks?
1478/18374 = 8.04% (practically same as your 8%)
We had basically the exact same droprate of vk’s over the weekend. I’m over the moon with how many vk’s I found but you’re complaining you didn’t get enough. Make of that what you will.
Long story short, the droprate of vk’s from eligible gnomes is 8%. Since the introduction of epics 10% (of the 8%) will be epic. You didn’t get robbed, you got exactly the average expected.
How did you obtain these numbers?
(Just reposting in case you missed my original post earlier.)
Iirc it was Sirrian himself who once divulged the 3% figure. And we know that during events the rate is x3 because during a special event we were told the special rate would be x5 instead of regular event x3.
So 8% now = 9%? Ok. Sure.