Crown Crate Drop Success Rate Thread (Live Server)

lehran · December 2016

Korah_Eaglecry wrote: »

From what I was reading on Reddit that seems to be pretty accurate. They found that the Apex mounts were dropping about 1% of the time. Which makes sense considering its the top tier of the crates. You arent going to sell alot of these crates if everyones getting the Apex rewards every time or even close to every time.

Those statistics are not accurate for live; drop rates were improved compared to the PTS. If reddit is seeing apex on average 1% of the time, then there is some serious sampling/reporting bias going on there (which doesn't surprise me one bit, making this into a scientific endeavor has required me to throw out the majority of data sources I found for crates; self-reporting simply isn't cutting it here for whatever reason).

Chance for apex on live is coming out to just under 3% based on my data (650 crates). This would mean the probability to get at least one Apex for N crates is the following (calculated via sum(.97077^x * 0.02923) for x = 0 to N-1 -- aka the probability that you open on your first plus the probability you don't get it on 1 but do get it on 2, plus the probability you don't get it on 1 or 2 but do get it on 3, and so on up to no for first N-1 but getting it on N):

35.917% chance of opening at least 1 Apex in 15 crates
50.933% chance of opening at least 1 Apex in 24 crates
58.933% chance of opening at least 1 Apex in 30 crates
73.683% chance of opening at least 1 Apex in 45 crates
83.135% chance of opening at least 1 Apex in 60 crates
90.113% chance of opening at least 1 Apex in 78 crates

EDIT: and since this is the top of a new page, might as well re-share my numbers and other observations so people know what I'm basing stuff on

You are guaranteed to get at least one Superior quality or better card in every crate. My per-crate rates below do not reflect this (they were computed using the per-card probabilities); in other words, you'll see a per-crate rate for 4 Normal cards, even though such is actually impossible. The numbers are provided anyway so you can see that all of them add up to 100%.
Apex rewards only appear in the bonus card slot, they never appear as one of the 4 normal cards.
Bonus cards are always Epic quality or better.
Color guide (for easy reference): Normal = White, Fine = Green, Superior = Blue, Epic = Purple, Legendary = Yellow, Apex = Orange
Total crates: 650
Chance of bonus card: 15.538%
Chance per card (normal): Normal = 47.923% Fine = 24.731% Superior = 23.962% Epic = 2.038% Legendary = 1.346%
Chance per card (bonus): Epic = 57.426% Legendary = 23.762% Apex = 18.812%
Chance per crate (number of Normal cards): 0=7.355% 1=27.073% 2=37.371% 3=22.927% 4=5.274%
Chance per crate (number of Fine cards): 0=32.097% 1=42.184% 2=20.790% 3=4.554% 4=0.374%
Chance per crate (number of Superior cards): 0=33.430% 1=42.138% 2=19.918% 3=4.184% 4=0.330%
Chance per crate (number of Epic cards): 0=83.875% 1=15.199% 2=0.902% 3=0.024% 4=0.000% 5=0.000%
Chance per crate (number of Legendary cards): 0=91.226% 1=8.477% 2=0.293% 3=0.005% 4=0.000% 5=0.000%
Chance per crate (number of Apex cards): 0=97.077% 1=2.923%
Chance per crate (at least this number of Normal cards): 0=100.000% 1=92.645% 2=65.572% 3=28.201% 4=5.274%
Chance per crate (at least this number of Fine cards): 0=100.000% 1=67.903% 2=25.718% 3=4.928% 4=0.374%
Chance per crate (at least this number of Superior cards): 0=100.000% 1=66.570% 2=24.432% 3=4.514% 4=0.330%
Chance per crate (at least this number of Epic cards): 0=100.000% 1=16.125% 2=0.927% 3=0.025% 4=0.000% 5=0.000%
Chance per crate (at least this number of Legendary cards): 0=100.000% 1=8.774% 2=0.298% 3=0.005% 4=0.000% 5=0.000%
Chance per crate (at least this number of Apex cards): 0=100.000% 1=2.923%

Raw data and the script used to parse said data is here.

Recremen · December 2016

lehran wrote: »

Korah_Eaglecry wrote: »

From what I was reading on Reddit that seems to be pretty accurate. They found that the Apex mounts were dropping about 1% of the time. Which makes sense considering its the top tier of the crates. You arent going to sell alot of these crates if everyones getting the Apex rewards every time or even close to every time.

Those statistics are not accurate for live; drop rates were improved compared to the PTS. If reddit is seeing apex on average 1% of the time, then there is some serious sampling/reporting bias going on there (which doesn't surprise me one bit, making this into a scientific endeavor has required me to throw out the majority of data sources I found for crates; self-reporting simply isn't cutting it here for whatever reason).

Chance for apex on live is coming out to just under 3% based on my data (650 crates). This would mean the probability to get at least one Apex for N crates is the following (calculated via sum(.97077^x * 0.02923) for x = 0 to N-1 -- aka the probability that you open on your first plus the probability you don't get it on 1 but do get it on 2, plus the probability you don't get it on 1 or 2 but do get it on 3, and so on up to no for first N-1 but getting it on N):
35.917% chance of opening at least 1 Apex in 15 crates

50.933% chance of opening at least 1 Apex in 24 crates

58.933% chance of opening at least 1 Apex in 30 crates

73.683% chance of opening at least 1 Apex in 45 crates

83.135% chance of opening at least 1 Apex in 60 crates

90.113% chance of opening at least 1 Apex in 78 crates

EDIT: and since this is the top of a new page, might as well re-share my numbers and other observations so people know what I'm basing stuff on
You are guaranteed to get at least one Superior quality or better card in every crate. My per-crate rates below do not reflect this (they were computed using the per-card probabilities); in other words, you'll see a per-crate rate for 4 Normal cards, even though such is actually impossible. The numbers are provided anyway so you can see that all of them add up to 100%.

Apex rewards only appear in the bonus card slot, they never appear as one of the 4 normal cards.

Bonus cards are always Epic quality or better.

Color guide (for easy reference): Normal = White, Fine = Green, Superior = Blue, Epic = Purple, Legendary = Yellow, Apex = Orange

Total crates: 650

Chance of bonus card: 15.538%

Chance per card (normal): Normal = 47.923% Fine = 24.731% Superior = 23.962% Epic = 2.038% Legendary = 1.346%

Chance per card (bonus): Epic = 57.426% Legendary = 23.762% Apex = 18.812%

Chance per crate (number of Normal cards): 0=7.355% 1=27.073% 2=37.371% 3=22.927% 4=5.274%

Chance per crate (number of Fine cards): 0=32.097% 1=42.184% 2=20.790% 3=4.554% 4=0.374%

Chance per crate (number of Superior cards): 0=33.430% 1=42.138% 2=19.918% 3=4.184% 4=0.330%

Chance per crate (number of Epic cards): 0=83.875% 1=15.199% 2=0.902% 3=0.024% 4=0.000% 5=0.000%

Chance per crate (number of Legendary cards): 0=91.226% 1=8.477% 2=0.293% 3=0.005% 4=0.000% 5=0.000%

Chance per crate (number of Apex cards): 0=97.077% 1=2.923%

Chance per crate (at least this number of Normal cards): 0=100.000% 1=92.645% 2=65.572% 3=28.201% 4=5.274%

Chance per crate (at least this number of Fine cards): 0=100.000% 1=67.903% 2=25.718% 3=4.928% 4=0.374%

Chance per crate (at least this number of Superior cards): 0=100.000% 1=66.570% 2=24.432% 3=4.514% 4=0.330%

Chance per crate (at least this number of Epic cards): 0=100.000% 1=16.125% 2=0.927% 3=0.025% 4=0.000% 5=0.000%

Chance per crate (at least this number of Legendary cards): 0=100.000% 1=8.774% 2=0.298% 3=0.005% 4=0.000% 5=0.000%

Chance per crate (at least this number of Apex cards): 0=100.000% 1=2.923%

Raw data and the script used to parse said data is here.

This highlights precisely why I cannot support this gambling feature in any capacity. Even after buying 78 gambling boxes, we will still wind up with about 1 in 10 people not getting a single Atronach mount. It is completely ludicrous. I know some people just can't see past the "Well at least I got mine" mentality, but these are downright terrible chances for getting something that should be directly purchasable.

JimT722 · December 2016

Recremen wrote: »

lehran wrote: »

Korah_Eaglecry wrote: »

From what I was reading on Reddit that seems to be pretty accurate. They found that the Apex mounts were dropping about 1% of the time. Which makes sense considering its the top tier of the crates. You arent going to sell alot of these crates if everyones getting the Apex rewards every time or even close to every time.

Those statistics are not accurate for live; drop rates were improved compared to the PTS. If reddit is seeing apex on average 1% of the time, then there is some serious sampling/reporting bias going on there (which doesn't surprise me one bit, making this into a scientific endeavor has required me to throw out the majority of data sources I found for crates; self-reporting simply isn't cutting it here for whatever reason).

Chance for apex on live is coming out to just under 3% based on my data (650 crates). This would mean the probability to get at least one Apex for N crates is the following (calculated via sum(.97077^x * 0.02923) for x = 0 to N-1 -- aka the probability that you open on your first plus the probability you don't get it on 1 but do get it on 2, plus the probability you don't get it on 1 or 2 but do get it on 3, and so on up to no for first N-1 but getting it on N):
35.917% chance of opening at least 1 Apex in 15 crates

50.933% chance of opening at least 1 Apex in 24 crates

58.933% chance of opening at least 1 Apex in 30 crates

73.683% chance of opening at least 1 Apex in 45 crates

83.135% chance of opening at least 1 Apex in 60 crates

90.113% chance of opening at least 1 Apex in 78 crates

EDIT: and since this is the top of a new page, might as well re-share my numbers and other observations so people know what I'm basing stuff on
You are guaranteed to get at least one Superior quality or better card in every crate. My per-crate rates below do not reflect this (they were computed using the per-card probabilities); in other words, you'll see a per-crate rate for 4 Normal cards, even though such is actually impossible. The numbers are provided anyway so you can see that all of them add up to 100%.

Apex rewards only appear in the bonus card slot, they never appear as one of the 4 normal cards.

Bonus cards are always Epic quality or better.

Color guide (for easy reference): Normal = White, Fine = Green, Superior = Blue, Epic = Purple, Legendary = Yellow, Apex = Orange

Total crates: 650

Chance of bonus card: 15.538%

Chance per card (normal): Normal = 47.923% Fine = 24.731% Superior = 23.962% Epic = 2.038% Legendary = 1.346%

Chance per card (bonus): Epic = 57.426% Legendary = 23.762% Apex = 18.812%

Chance per crate (number of Normal cards): 0=7.355% 1=27.073% 2=37.371% 3=22.927% 4=5.274%

Chance per crate (number of Fine cards): 0=32.097% 1=42.184% 2=20.790% 3=4.554% 4=0.374%

Chance per crate (number of Superior cards): 0=33.430% 1=42.138% 2=19.918% 3=4.184% 4=0.330%

Chance per crate (number of Epic cards): 0=83.875% 1=15.199% 2=0.902% 3=0.024% 4=0.000% 5=0.000%

Chance per crate (number of Legendary cards): 0=91.226% 1=8.477% 2=0.293% 3=0.005% 4=0.000% 5=0.000%

Chance per crate (number of Apex cards): 0=97.077% 1=2.923%

Chance per crate (at least this number of Normal cards): 0=100.000% 1=92.645% 2=65.572% 3=28.201% 4=5.274%

Chance per crate (at least this number of Fine cards): 0=100.000% 1=67.903% 2=25.718% 3=4.928% 4=0.374%

Chance per crate (at least this number of Superior cards): 0=100.000% 1=66.570% 2=24.432% 3=4.514% 4=0.330%

Chance per crate (at least this number of Epic cards): 0=100.000% 1=16.125% 2=0.927% 3=0.025% 4=0.000% 5=0.000%

Chance per crate (at least this number of Legendary cards): 0=100.000% 1=8.774% 2=0.298% 3=0.005% 4=0.000% 5=0.000%

Chance per crate (at least this number of Apex cards): 0=100.000% 1=2.923%

Raw data and the script used to parse said data is here.

This highlights precisely why I cannot support this gambling feature in any capacity. Even after buying 78 gambling boxes, we will still wind up with about 1 in 10 people not getting a single Atronach mount. It is completely ludicrous. I know some people just can't see past the "Well at least I got mine" mentality, but these are downright terrible chances for getting something that should be directly purchasable.

They could also change them next season for any reason. I don't think they said they never would. I doubt people will run these sort of tests every season and since they don't give us the drop rates we would never know.

ninti · December 2016

No offense to Enodoc's hard work, but I am working on a spreadsheet of my own, one that is vetted and verifiable. I have 811 crates in it right now, and the odds I am seeing are much worse then Enodoc's spreadsheet.

Apex 0.65%
Legendary 2.58%

Chance of a fifth card is 12.45% per crate.

The spreadsheet is at https://docs.google.com/spreadsheets/d/1YE8TpQxwJvRBdIoHKKFSnt5nl2qttvFPhdQly_xz6y8/edit?usp=sharing

Edit: I accidentally put the superior number as the Epic number. I have now erased it.

lehran · December 2016

ninti wrote: »

No offense to Enodoc's hard work, but I am working on a spreadsheet of my own, one that is vetted and verifiable. I have 811 crates in it right now, and the odds I am seeing are much worse then Enodoc's spreadsheet.

Apex 0.65%
Legendary 2.58%
Epc 22.76%

Chance of a fifth card is 12.45% per crate.

The spreadsheet is at https://docs.google.com/spreadsheets/d/1YE8TpQxwJvRBdIoHKKFSnt5nl2qttvFPhdQly_xz6y8/edit?usp=sharing

I find it interesting that you and I are getting vastly different data. I'm working on putting additional crates into my dataset. I'm not using any of the data from Enodoc's sheet, it's 100% input by me based on my own 205 crates as well as youtube videos of crate openings; and for said videos I only count them if they're opening 15+ crates in a single video or starting at 0 crown gems and opening at least 4 crates, and the videos cannot have any skips or cuts in them. This helps me ensure that I'm seeing raw data rather than only the crates that the person decided were worthy to show, either because they were really lucky or really unlucky. I've found that on these forums at least, self-reported data tends to largely be in one of those two camps, and I don't see many middle-of-the-road reports. Screenshots are similarly not verifiable, as they may also be cherry-picked. When I get 800+ crates in my dataset I'll run my numbers again and see if it trends down towards what you have or stays relatively steady; my guess is going to be the later.

My preference is that ZOS would just release the drop rate numbers so we didn't have to do this, as well as any methodology or special behavior to the crates. Furthermore, I'd like them to run several million simulations of crate openings using the live server RNG to prove that the actual numbers match up with what ZOS claims them to be (aka make sure the RNG is actually fair).

JimT722 · December 2016

lehran wrote: »

ninti wrote: »

No offense to Enodoc's hard work, but I am working on a spreadsheet of my own, one that is vetted and verifiable. I have 811 crates in it right now, and the odds I am seeing are much worse then Enodoc's spreadsheet.

Apex 0.65%
Legendary 2.58%
Epc 22.76%

Chance of a fifth card is 12.45% per crate.

The spreadsheet is at https://docs.google.com/spreadsheets/d/1YE8TpQxwJvRBdIoHKKFSnt5nl2qttvFPhdQly_xz6y8/edit?usp=sharing

I find it interesting that you and I are getting vastly different data. I'm working on putting additional crates into my dataset. I'm not using any of the data from Enodoc's sheet, it's 100% input by me based on my own 205 crates as well as youtube videos of crate openings; and for said videos I only count them if they're opening 15+ crates in a single video or starting at 0 crown gems and opening at least 4 crates, and the videos cannot have any skips or cuts in them. This helps me ensure that I'm seeing raw data rather than only the crates that the person decided were worthy to show, either because they were really lucky or really unlucky. I've found that on these forums at least, self-reported data tends to largely be in one of those two camps, and I don't see many middle-of-the-road reports. Screenshots are similarly not verifiable, as they may also be cherry-picked. When I get 800+ crates in my dataset I'll run my numbers again and see if it trends down towards what you have or stays relatively steady; my guess is going to be the later.

My preference is that ZOS would just release the drop rate numbers so we didn't have to do this, as well as any methodology or special behavior to the crates. Furthermore, I'd like them to run several million simulations of crate openings using the live server RNG to prove that the actual numbers match up with what ZOS claims them to be (aka make sure the RNG is actually fair).

I read in an article about lockboxes that said the companies releasing these things probably don't want the drop rates known. Many people would look at the actual rates and say forget it. This would make sense to me.

Elsonso · December 2016

lehran wrote: »

My preference is that ZOS would just release the drop rate numbers so we didn't have to do this, as well as any methodology or special behavior to the crates. Furthermore, I'd like them to run several million simulations of crate openings using the live server RNG to prove that the actual numbers match up with what ZOS claims them to be (aka make sure the RNG is actually fair).

LOL.

ZOS has people spending hundreds on Crowns to do this research. What possible reason would drive them to just give us this data?

lehran · December 2016

Update for 832 crates on my dataset, still haven't seen any Apex outside of the bonus card slot (as demonstrated below). I'm continually tweaking the formatting for both readability and so that the post doesn't get too long. If you have any suggestions on how to better format it, please let me know (like, if there's a table bbcode I'm unaware of).

Total crates: 832
Note: Per-crate percentages are listed as X% (Y%) where X% is the theoretical chance based on the per-card rates and Y% is the experimental chance computed by counting how many crates in the dataset contained that property.
Note: Percentages are rounded to nearest thousandth. Theoretical percentages listed below as 0.000% are still possible, but have less than a 0.0005% chance of occurring. Experimental percentages listed as 0.000% are actually 0%.
Number of each card (normal): Common=1611 Fine=798 Superior=795 Epic=72 Legendary=52 Apex=0
Number of each card (bonus): Common=0 Fine=0 Superior=0 Epic=69 Legendary=34 Apex=28
Chance of bonus card: 15.745%
Chance per card (normal): Common=48.407% Fine=23.978% Superior=23.888% Epic=2.163% Legendary=1.562%
Chance per card (bonus): Epic=52.672% Legendary=25.954% Apex=21.374%
Chance per crate (number of Common cards): 0=7.085% (2.524%) 1=26.591% (24.639%) 2=37.424% (49.519%) 3=23.409% (23.317%) 4=5.491% (0.000%)
Chance per crate (number of Fine cards): 0=33.400% (30.529%) 1=42.140% (45.553%) 2=19.937% (21.394%) 3=4.192% (2.524%) 4=0.331% (0.000%)
Chance per crate (number of Superior cards): 0=33.559% (9.375%) 1=42.131% (85.697%) 2=19.835% (4.928%) 3=4.150% (0.000%) 4=0.326% (0.000%)
Chance per crate (number of Epic cards): 0=84.024% (84.135%) 1=15.031% (14.904%) 2=0.919% (0.841%) 3=0.026% (0.120%) 4=0.000% (0.000%) 5=0.000% (0.000%)
Chance per crate (number of Legendary cards): 0=90.058% (90.144%) 1=9.555% (9.375%) 2=0.380% (0.481%) 3=0.007% (0.000%) 4=0.000% (0.000%) 5=0.000% (0.000%)
Chance per crate (number of Apex cards): 0=96.635% (96.635%) 1=3.365% (3.365%)
Chance per crate (at least this number of Common cards): 0=100.000% (100.000%) 1=92.915% (97.476%) 2=66.324% (72.837%) 3=28.900% (23.317%) 4=5.491% (0.000%)
Chance per crate (at least this number of Fine cards): 0=100.000% (100.000%) 1=66.600% (69.471%) 2=24.460% (23.918%) 3=4.523% (2.524%) 4=0.331% (0.000%)
Chance per crate (at least this number of Superior cards): 0=100.000% (100.000%) 1=66.441% (90.625%) 2=24.310% (4.928%) 3=4.476% (0.000%) 4=0.326% (0.000%)
Chance per crate (at least this number of Epic cards): 0=100.000% (100.000%) 1=15.976% (15.865%) 2=0.945% (0.962%) 3=0.026% (0.120%) 4=0.000% (0.000%) 5=0.000% (0.000%)
Chance per crate (at least this number of Legendary cards): 0=100.000% (100.000%) 1=9.942% (9.856%) 2=0.387% (0.481%) 3=0.007% (0.000%) 4=0.000% (0.000%) 5=0.000% (0.000%)
Chance per crate (at least this number of Apex cards): 0=100.000% (100.000%) 1=3.365% (3.365%)
Chance to get at least 1 Apex in 15 crates: 40.160%
Number of crates to have 50% chance to get at least 1 Apex (rounded up): 21
Number of crates to have 75% chance to get at least 1 Apex (rounded up): 41
Number of crates to have 90% chance to get at least 1 Apex (rounded up): 68
Raw data and script used to generate the above statistics available on GitHub.

AdamBourke · December 2016

I got 4 crates, and got the Soul-Shriven armour 3 times... I got a dog as well. The rest were consumables. Including 1 riding lesson and 1 Mimic Stone, which I kept.

Everything else I traded in and got a grand total of... 19 gems.

lehran · December 2016

@driosketch I've finally found an Apex in one of the four normal slots, so it is possible on the live server (similar to how it was on PTS). The chance is just abysmal.

Total crates: 1012
Note: Per-crate percentages are listed as X% (Y%) where X% is the theoretical chance based on the per-card rates and Y% is the experimental chance computed by counting how many crates in the dataset contained that property.
Note: Percentages are rounded to nearest ten-thousandth. Theoretical percentages listed below as 0.0000% are still possible, but have less than a 0.00005% chance of occurring. Experimental percentages listed as 0.0000% are actually 0%.
Number of each card (normal): Common=1959 Fine=968 Superior=976 Epic=83 Legendary=61 Apex=1
Number of each card (bonus): Common=0 Fine=0 Superior=0 Epic=94 Legendary=41 Apex=33
Chance of bonus card: 16.6008%
Chance per card (normal): Common=48.3943% Fine=23.9130% Superior=24.1107% Epic=2.0504% Legendary=1.5069% Apex=0.0247%
Chance per card (bonus): Epic=55.9524% Legendary=24.4048% Apex=19.6429%
Chance per crate (number of Common cards): 0=7.0924% (2.5692%) 1=26.6040% (23.7154%) 2=37.4227% (51.2846%) 3=23.3959% (22.4308%) 4=5.4850% (0.0000%)
Chance per crate (number of Fine cards): 0=33.5151% (30.2372%) 1=42.1333% (46.3439%) 2=19.8628% (20.9486%) 3=4.1617% (2.4704%) 4=0.3270% (0.0000%)
Chance per crate (number of Superior cards): 0=33.1683% (8.8933%) 1=42.1513% (85.7708%) 2=20.0877% (5.3360%) 3=4.2547% (0.0000%) 4=0.3379% (0.0000%)
Chance per crate (number of Epic cards): 0=83.4974% (83.7945%) 1=15.5413% (15.0198%) 2=0.9354% (1.0870%) 3=0.0255% (0.0988%) 4=0.0003% (0.0000%) 5=0.0000% (0.0000%)
Chance per crate (number of Legendary cards): 0=90.2946% (90.4150%) 1=9.3386% (9.0909%) 2=0.3601% (0.4941%) 3=0.0066% (0.0000%) 4=0.0001% (0.0000%) 5=0.0000% (0.0000%)
Chance per crate (number of Apex cards): 0=96.6436% (96.6403%) 1=3.3532% (3.3597%) 2=0.0033% (0.0000%) 3=0.0000% (0.0000%) 4=0.0000% (0.0000%) 5=0.0000% (0.0000%)
Chance per crate (at least this number of Common cards): 0=100.0000% (100.0000%) 1=92.9076% (97.4308%) 2=66.3036% (73.7154%) 3=28.8809% (22.4308%) 4=5.4850% (0.0000%)
Chance per crate (at least this number of Fine cards): 0=100.0000% (100.0000%) 1=66.4849% (69.7628%) 2=24.3516% (23.4190%) 3=4.4887% (2.4704%) 4=0.3270% (0.0000%)
Chance per crate (at least this number of Superior cards): 0=100.0000% (100.0000%) 1=66.8317% (91.1067%) 2=24.6804% (5.3360%) 3=4.5926% (0.0000%) 4=0.3379% (0.0000%)
Chance per crate (at least this number of Epic cards): 0=100.0000% (100.0000%) 1=16.5026% (16.2055%) 2=0.9613% (1.1858%) 3=0.0259% (0.0988%) 4=0.0003% (0.0000%) 5=0.0000% (0.0000%)
Chance per crate (at least this number of Legendary cards): 0=100.0000% (100.0000%) 1=9.7054% (9.5850%) 2=0.3669% (0.4941%) 3=0.0067% (0.0000%) 4=0.0001% (0.0000%) 5=0.0000% (0.0000%)
Chance per crate (at least this number of Apex cards): 0=100.0000% (100.0000%) 1=3.3564% (3.3597%) 2=0.0033% (0.0000%) 3=0.0000% (0.0000%) 4=0.0000% (0.0000%) 5=0.0000% (0.0000%)
Chance to get at least 1 Apex in 15 crates: 40.0380%
Number of crates to have 50% chance to get at least 1 Apex (rounded up): 21
Number of crates to have 75% chance to get at least 1 Apex (rounded up): 41
Number of crates to have 90% chance to get at least 1 Apex (rounded up): 68
Raw data and script used to generate the above statistics available on GitHub.

snakester320 · December 2016

lehran wrote: »

ninti wrote: »

No offense to Enodoc's hard work, but I am working on a spreadsheet of my own, one that is vetted and verifiable. I have 811 crates in it right now, and the odds I am seeing are much worse then Enodoc's spreadsheet.

Apex 0.65%
Legendary 2.58%
Epc 22.76%

Chance of a fifth card is 12.45% per crate.

The spreadsheet is at https://docs.google.com/spreadsheets/d/1YE8TpQxwJvRBdIoHKKFSnt5nl2qttvFPhdQly_xz6y8/edit?usp=sharing

I find it interesting that you and I are getting vastly different data. I'm working on putting additional crates into my dataset. I'm not using any of the data from Enodoc's sheet, it's 100% input by me based on my own 205 crates as well as youtube videos of crate openings; and for said videos I only count them if they're opening 15+ crates in a single video or starting at 0 crown gems and opening at least 4 crates, and the videos cannot have any skips or cuts in them. This helps me ensure that I'm seeing raw data rather than only the crates that the person decided were worthy to show, either because they were really lucky or really unlucky. I've found that on these forums at least, self-reported data tends to largely be in one of those two camps, and I don't see many middle-of-the-road reports. Screenshots are similarly not verifiable, as they may also be cherry-picked. When I get 800+ crates in my dataset I'll run my numbers again and see if it trends down towards what you have or stays relatively steady; my guess is going to be the later.

My preference is that ZOS would just release the drop rate numbers so we didn't have to do this, as well as any methodology or special behavior to the crates. Furthermore, I'd like them to run several million simulations of crate openings using the live server RNG to prove that the actual numbers match up with what ZOS claims them to be (aka make sure the RNG is actually fair).

Sorry but why would they ever publish the drop rates they force this disgusting business method on us that's clearly a cash grab and nothing more and then tell you how much there screwing out of ppl by publishing the drop rates I think not!

driosketch · December 2016

lehran wrote: »

@driosketch I've finally found an Apex in one of the four normal slots, so it is possible on the live server (similar to how it was on PTS). The chance is just abysmal.

Total crates: 1012

Note: Per-crate percentages are listed as X% (Y%) where X% is the theoretical chance based on the per-card rates and Y% is the experimental chance computed by counting how many crates in the dataset contained that property.

Note: Percentages are rounded to nearest ten-thousandth. Theoretical percentages listed below as 0.0000% are still possible, but have less than a 0.00005% chance of occurring. Experimental percentages listed as 0.0000% are actually 0%.

Number of each card (normal): Common=1959 Fine=968 Superior=976 Epic=83 Legendary=61 Apex=1

Number of each card (bonus): Common=0 Fine=0 Superior=0 Epic=94 Legendary=41 Apex=33

Chance of bonus card: 16.6008%

Chance per card (normal): Common=48.3943% Fine=23.9130% Superior=24.1107% Epic=2.0504% Legendary=1.5069% Apex=0.0247%

Chance per card (bonus): Epic=55.9524% Legendary=24.4048% Apex=19.6429%

Chance per crate (number of Common cards): 0=7.0924% (2.5692%) 1=26.6040% (23.7154%) 2=37.4227% (51.2846%) 3=23.3959% (22.4308%) 4=5.4850% (0.0000%)

Chance per crate (number of Fine cards): 0=33.5151% (30.2372%) 1=42.1333% (46.3439%) 2=19.8628% (20.9486%) 3=4.1617% (2.4704%) 4=0.3270% (0.0000%)

Chance per crate (number of Superior cards): 0=33.1683% (8.8933%) 1=42.1513% (85.7708%) 2=20.0877% (5.3360%) 3=4.2547% (0.0000%) 4=0.3379% (0.0000%)

Chance per crate (number of Epic cards): 0=83.4974% (83.7945%) 1=15.5413% (15.0198%) 2=0.9354% (1.0870%) 3=0.0255% (0.0988%) 4=0.0003% (0.0000%) 5=0.0000% (0.0000%)

Chance per crate (number of Legendary cards): 0=90.2946% (90.4150%) 1=9.3386% (9.0909%) 2=0.3601% (0.4941%) 3=0.0066% (0.0000%) 4=0.0001% (0.0000%) 5=0.0000% (0.0000%)

Chance per crate (number of Apex cards): 0=96.6436% (96.6403%) 1=3.3532% (3.3597%) 2=0.0033% (0.0000%) 3=0.0000% (0.0000%) 4=0.0000% (0.0000%) 5=0.0000% (0.0000%)

Chance per crate (at least this number of Common cards): 0=100.0000% (100.0000%) 1=92.9076% (97.4308%) 2=66.3036% (73.7154%) 3=28.8809% (22.4308%) 4=5.4850% (0.0000%)

Chance per crate (at least this number of Fine cards): 0=100.0000% (100.0000%) 1=66.4849% (69.7628%) 2=24.3516% (23.4190%) 3=4.4887% (2.4704%) 4=0.3270% (0.0000%)

Chance per crate (at least this number of Superior cards): 0=100.0000% (100.0000%) 1=66.8317% (91.1067%) 2=24.6804% (5.3360%) 3=4.5926% (0.0000%) 4=0.3379% (0.0000%)

Chance per crate (at least this number of Epic cards): 0=100.0000% (100.0000%) 1=16.5026% (16.2055%) 2=0.9613% (1.1858%) 3=0.0259% (0.0988%) 4=0.0003% (0.0000%) 5=0.0000% (0.0000%)

Chance per crate (at least this number of Legendary cards): 0=100.0000% (100.0000%) 1=9.7054% (9.5850%) 2=0.3669% (0.4941%) 3=0.0067% (0.0000%) 4=0.0001% (0.0000%) 5=0.0000% (0.0000%)

Chance per crate (at least this number of Apex cards): 0=100.0000% (100.0000%) 1=3.3564% (3.3597%) 2=0.0033% (0.0000%) 3=0.0000% (0.0000%) 4=0.0000% (0.0000%) 5=0.0000% (0.0000%)

Chance to get at least 1 Apex in 15 crates: 40.0380%

Number of crates to have 50% chance to get at least 1 Apex (rounded up): 21

Number of crates to have 75% chance to get at least 1 Apex (rounded up): 41

Number of crates to have 90% chance to get at least 1 Apex (rounded up): 68

Raw data and script used to generate the above statistics available on GitHub.

Okay, well so far the only confirmed change from PTS crates are the consumable to gem conversion as well as adding them to the gem store. It looks like maybe they did bump the drop rate, but at this point we don't know if that's x2 or a x10 boost. In either case, it appears crates only have better value over the direct store if you don't care what you get.

Concenaros · December 2016

TBH I really have no issue with the Crown Crates.

Reason? If you look at any mobile game, they have a Gacha system which is basically this. Recently, in the past few years these systems have started to bridge over into other genres because it's a decent way to provide players with content and of course is lucrative from a business standpoint.

The odds (of course) are always set low for high rarity or wanted items and honestly, so long as ZoS continues to only provide aesthetic items and content within the crown crates, then there's no issue. It's a buy and try at your own risk kind of deal. At the end of the day, ZoS needs money to support the game. This system is just another that compounds on top of the already exisiting VIP system and Crown Store.

It's not underhanded or unusual, it's business as usual for most of the gaming industry now (unfortunately).

I'm not defending Crown Crates, I'm just saying that this topic has been brought up to the point of exhaustion in a number of titles I play and it never goes anywhere, people try to guess the Gacha drop rates and they fail. This is mainly because the odds are adjustable and can be tweaked at any time. Some companies also have full on weight adjusters built into games in order to have more targetted offers based on certain spending criteria and progress and to help them create interesting LO deals that earn the game more money.

Also, they aren't legally obligated to provide the drop rates since it's a Gacha system and Gacha is still not really considered as gambling since you're paying direct cash for a chance at a good or high rarity prize. Either way you still receive something even if it's not what you wanted. So it's technically a purchase. It all really depends on your gambling laws and how most mobile titles are handled within your region. Chances are, if you have mobile games that use gacha within your region and they don't have to post drop chances, then neither does ZoS with ESO.

JimT722 · December 2016

Concenaros wrote: »

TBH I really have no issue with the Crown Crates.

Reason? If you look at any mobile game, they have a Gacha system which is basically this. Recently, in the past few years these systems have started to bridge over into other genres because it's a decent way to provide players with content and of course is lucrative from a business standpoint.

The odds (of course) are always set low for high rarity or wanted items and honestly, so long as ZoS continues to only provide aesthetic items and content within the crown crates, then there's no issue. It's a buy and try at your own risk kind of deal. At the end of the day, ZoS needs money to support the game. This system is just another that compounds on top of the already exisiting VIP system and Crown Store.

It's not underhanded or unusual, it's business as usual for most of the gaming industry now (unfortunately).

I'm not defending Crown Crates, I'm just saying that this topic has been brought up to the point of exhaustion in a number of titles I play and it never goes anywhere, people try to guess the Gacha drop rates and they fail. This is mainly because the odds are adjustable and can be tweaked at any time. Some companies also have full on weight adjusters built into games in order to have more targetted offers based on certain spending criteria and progress and to help them create interesting LO deals that earn the game more money.

Also, they aren't legally obligated to provide the drop rates since it's a Gacha system and Gacha is still not really considered as gambling since you're paying direct cash for a chance at a good or high rarity prize. Either way you still receive something even if it's not what you wanted. So it's technically a purchase. It all really depends on your gambling laws and how most mobile titles are handled within your region. Chances are, if you have mobile games that use gacha within your region and they don't have to post drop chances, then neither does ZoS with ESO.

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

Concenaros · December 2016

JimT722 wrote: »

Concenaros wrote: »

TBH I really have no issue with the Crown Crates.

Reason? If you look at any mobile game, they have a Gacha system which is basically this. Recently, in the past few years these systems have started to bridge over into other genres because it's a decent way to provide players with content and of course is lucrative from a business standpoint.

The odds (of course) are always set low for high rarity or wanted items and honestly, so long as ZoS continues to only provide aesthetic items and content within the crown crates, then there's no issue. It's a buy and try at your own risk kind of deal. At the end of the day, ZoS needs money to support the game. This system is just another that compounds on top of the already exisiting VIP system and Crown Store.

It's not underhanded or unusual, it's business as usual for most of the gaming industry now (unfortunately).

I'm not defending Crown Crates, I'm just saying that this topic has been brought up to the point of exhaustion in a number of titles I play and it never goes anywhere, people try to guess the Gacha drop rates and they fail. This is mainly because the odds are adjustable and can be tweaked at any time. Some companies also have full on weight adjusters built into games in order to have more targetted offers based on certain spending criteria and progress and to help them create interesting LO deals that earn the game more money.

Also, they aren't legally obligated to provide the drop rates since it's a Gacha system and Gacha is still not really considered as gambling since you're paying direct cash for a chance at a good or high rarity prize. Either way you still receive something even if it's not what you wanted. So it's technically a purchase. It all really depends on your gambling laws and how most mobile titles are handled within your region. Chances are, if you have mobile games that use gacha within your region and they don't have to post drop chances, then neither does ZoS with ESO.

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

More like soul-shriven (Lol) from seeing it happen all to often in platforms other than mobile and watching as the community tries to battle against it's addition, only to watch it stick around with constantly re-tooling from the devs.

driosketch · December 2016

It's not underhanded

I'm just saying that this topic has been brought up to the point of exhaustion in a number of titles I play and it never goes anywhere, people try to guess the Gacha drop rates and they fail. This is mainly because the odds are adjustable and can be tweaked at any time.

You and I have a different definition of "underhanded".

To the larger issue, the international port of what was once the most popular mobile TCG shut down completely at the start of this year. They had about a 3 year run. Comparisons of ESO to mobile games is not reassuring.

Elsonso · December 2016

JimT722 wrote: »

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

You must have misread them. I am pretty sure they were comparing ESO to a mobile game.

Concenaros · December 2016

lordrichter wrote: »

JimT722 wrote: »

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

You must have misread them. I am pretty sure they were comparing ESO to a mobile game.

To be fair I wasn't comparing, ESO to Mobile or vice versa. What I was saying is they've adopted a Gacha system FROM mobile into ESO.

Which a lot of non-mobile games are doing now. Since Gacha is the norm on mobile I thought I'd at least touch on where the system sort-of originated. Lol.

But yes, he misread me.

Rofl.

lehran · December 2016

MMOs have had gacha systems long before smartphones were a thing, it's not a remotely new phenomenon.

While they have no obligation to publish drop rates (I'm assuming the legal team has done their homework on this before making the feature live), with enough recorded trials we can get a pretty darn good estimate on what those rates are. While I expect that exact rates may vary from season-to-season, I don't expect rates to change within a season; at least, not without being advertised in patch notes. As such, collecting this data and computing these figures allows people who may consider purchasing crates for whatever reason to make a more informed decision than going in blind. Some people may want everything or most everything, and are interested in purchasing the least number of crowns needed to have a good shot at that. Others may want one particular thing, but don't want to spend more than $X on it, so they can see from the outset whether or not they can realistically get it by spending $X, and if not, holding off on their purchase.

From my own observations, there is no sort of weighting going on behind-the-scenes between crates per account; each crate's chances are computed independently of other crates you may open. This is why we see people like Deltia who open dozens of crates and don't get anything compared with other people who got 3 Apexes within a 15-crate pack, which would be very rare if they were manipulating the odds based on what was previously opened. Otherwise, I'd expect rates more similar to MtG for example where each box is pretty much guaranteed to have exactly 4 mythic rares, and each case is pretty much guaranteed to have exactly 1 foil mythic rare. It's not guaranteed, but the vast majority of boxes/cases are like that simply due to how the cards are printed. (For those not familiar with MtG, it stands for Magic the Gathering and it's a trading card game. A box is 36 booster packs and a case is 6 boxes.)

threefarms · December 2016

It's always a 50% chance of getting an Apex. You either get or you don't.

Stupid ZOS.

Voxicity · December 2016

45 crates no atro, not buying any more in the future

rotaugen454 · December 2016

threefarms wrote: »

It's always a 50% chance of getting an Apex. You either get or you don't.

Stupid ZOS.

Then I'm going to play the lottery. I figure that I have a 50% chance!

Concenaros · December 2016

lehran wrote: »

While they have no obligation to publish drop rates (I'm assuming the legal team has done their homework on this before making the feature live), with enough recorded trials we can get a pretty darn good estimate on what those rates are. While I expect that exact rates may vary from season-to-season, I don't expect rates to change within a season; at least, not without being advertised in patch notes. As such, collecting this data and computing these figures allows people who may consider purchasing crates for whatever reason to make a more informed decision than going in blind. Some people may want everything or most everything, and are interested in purchasing the least number of crowns needed to have a good shot at that. Others may want one particular thing, but don't want to spend more than $X on it, so they can see from the outset whether or not they can realistically get it by spending $X, and if not, holding off on their purchase.

From my own observations, there is no sort of weighting going on behind-the-scenes between crates per account; each crate's chances are computed independently of other crates you may open. This is why we see people like Deltia who open dozens of crates and don't get anything compared with other people who got 3 Apexes within a 15-crate pack, which would be very rare if they were manipulating the odds based on what was previously opened. Otherwise, I'd expect rates more similar to MtG for example where each box is pretty much guaranteed to have exactly 4 mythic rares, and each case is pretty much guaranteed to have exactly 1 foil mythic rare. It's not guaranteed, but the vast majority of boxes/cases are like that simply due to how the cards are printed. (For those not familiar with MtG, it stands for Magic the Gathering and it's a trading card game. A box is 36 booster packs and a case is 6 boxes.)

They may not be doing adjusted weighting to start, as they have no reason to target specific users.

Usually once an average amount of converted users has been found, they can start breaking down those users and start planning crates with added adjusters that will make those users spend again and can propose crates that will potentially convert non-spenders.

I'm guessing we'll 'see' what they do in the future with the crates.

I'm not against you guys attempting to pinpoint those odds, I'm just saying that they can and probably will adjust those drop chances constantly, especially if you guys get close to the actual rate. They don't have to say they've adjusted the rates within the crates in patch notes either, they might, but I doubt they will. I've seen many a thread similar to this one posted in other forums for other titles and the users usually give up because they chances and weighting begin to fluctuate so much it's hard to get a good approximation on the real values without knowing what's going on behind the scenes.

Might also be worth looking into if they've implemented streak breakers too, to at least guarantee a prize of a certain rarity if the user has opened a pre-determined amount of crates without receiving one of the featured or rarer rewards.

Anyhow, keep on keeping on. Good luck to you all if you choose to keep this thread alive and constantly investigate the drop chances to the best of your abilities.

snakester320 · December 2016

Concenaros wrote: »

lordrichter wrote: »

JimT722 wrote: »

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

You must have misread them. I am pretty sure they were comparing ESO to a mobile game.

To be fair I wasn't comparing, ESO to Mobile or vice versa. What I was saying is they've adopted a Gacha system FROM mobile into ESO.

Which a lot of non-mobile games are doing now. Since Gacha is the norm on mobile I thought I'd at least touch on where the system sort-of originated. Lol.

But yes, he misread me. Rofl.

these systems started a while ago first I saw it was mass effect 3 multi player prob a idea of the witch herself... but they had the good sense to allow you to buy crates for in game credits as well where as with eso they have taken it to disgusting lvls my guess the witch is on her broom again this time with more sting!

Prof_Bawbag · December 2016

KaiVox22 wrote: »

45 crates no atro, not buying any more in the future

Could you not even afford one with all those gems?

JimT722 · December 2016

Prof_Bawbag wrote: »

KaiVox22 wrote: »

45 crates no atro, not buying any more in the future

Could you not even afford one with all those gems?

You can get as low as 3 gems a crate.

driosketch · December 2016

JimT722 wrote: »

Prof_Bawbag wrote: »

KaiVox22 wrote: »

45 crates no atro, not buying any more in the future

Could you not even afford one with all those gems?

You can get as low as 3 gems a crate.

Hell, 45 times 8 gems still only gets you to 360 gems. And if I'm reading lerah's data right, you can pull as low as 2 or even 1 gem from a crate.

ninti · December 2016

lehran wrote: »

I find it interesting that you and I are getting vastly different data.

It's not that far off. You are giving odds by crate, I was giving odds by card. By crate, my current spreadsheet gives 2.8%, while yours is 3.3%.

lehran wrote: »

I'm working on putting additional crates into my dataset. I'm not using any of the data from Enodoc's sheet, it's 100% input by me based on my own 205 crates as well as youtube videos of crate openings; and for said videos I only count them if they're opening 15+ crates in a single video or starting at 0 crown gems and opening at least 4 crates, and the videos cannot have any skips or cuts in them. This helps me ensure that I'm seeing raw data rather than only the crates that the person decided were worthy to show, either because they were really lucky or really unlucky.

Yes, I am using similar criteria, although maybe not as strict. Unless I have any reason to believe they are not reporting all the crates they opened, and they provide ALL of the data on their drops, I have been using it.

Post your full data when you can please, i would like to see it. You have been paying much more attention to the bonus cards than I have, which is a better method than my ignoring them and just folding it in. Feel free to look at mine if you want to see if I have any sources you missed. https://docs.google.com/spreadsheets/d/1YE8TpQxwJvRBdIoHKKFSnt5nl2qttvFPhdQly_xz6y8/edit#gid=0

Concenaros · December 2016

snakester320 wrote: »

Concenaros wrote: »

lordrichter wrote: »

JimT722 wrote: »

I never thought I would see the day someone compares a AAA MMO with mobile games, and say it's okay. Low standards?

You must have misread them. I am pretty sure they were comparing ESO to a mobile game.

To be fair I wasn't comparing, ESO to Mobile or vice versa. What I was saying is they've adopted a Gacha system FROM mobile into ESO.

Which a lot of non-mobile games are doing now. Since Gacha is the norm on mobile I thought I'd at least touch on where the system sort-of originated. Lol.

But yes, he misread me. Rofl.

these systems started a while ago first I saw it was mass effect 3 multi player prob a idea of the witch herself... but they had the good sense to allow you to buy crates for in game credits as well where as with eso they have taken it to disgusting lvls my guess the witch is on her broom again this time with more sting!

I've seen them in plenty of games before ME3. But like I said it's something that's been going on for awhile within Console and other platforms. How long ago? I have no idea but I know that FIFA has used a similar system with FUT, I know that Gears 4 is using this for Horde mode upgrades and stuff, so yeah I agree with your and my previous statement in that it's been happening more and more in places other than mobile.

It's kind of sad to see, but companies need to make money. So I'm not surprised it made the jump over.

JimT722 · December 2016

What would be the worst case scenario for 56 gems?