Page 2 of 6 FirstFirst 1234 ... LastLast
Results 11 to 20 of 55

Thread: Anniversary Chest drop rates - Are we getting fooled?

  1. #11
    Join Date
    Jan 2010
    Location
    Chimra
    Posts
    2,104
    Blog Entries
    1

    Default

    Quote Originally Posted by zserg View Post
    Yeah, what I find odd is that I have seen several people saying they got even multiple legendaries, yet that data says 0.

    Maybe people reporting on discord were somehow a little biased to mostly report negative results ?

    Edit: the common % does seem a little high though.
    not only does it seem high... that would be higher than expected common on a normal chest....
    "Nothing is true, everything is permitted"
    [High King, Lieutenant Warchief, Duke, Avenger] Keep is in Signature. Play Nice Meow ;3

  2. #12
    Join Date
    Jan 2010
    Location
    Chimra
    Posts
    2,104
    Blog Entries
    1

    Default

    Quote Originally Posted by pull lead View Post
    You are off with your calculations. The chances of getting a legendary are 1-(.99^n) where n is number of rolls attempted. Its only been about 20 years for me and formal statistics.

    You have about a 60% chance to get a legendary in 100 rolls.
    assuming both sets of data are unique with no overlap... we are talking about 1240 rolls. I don't doubt players *have* gotten them; I do however find this incredibly curious as the chances for that many rolls to have a 1/100 odds to yield a net 0 after almost 13 full sets of 1/100 probability. The odds should remain consistent as stated and they just aren't. These two guilds may have to be the *luckiest* [in being unlucky] for these results.
    "Nothing is true, everything is permitted"
    [High King, Lieutenant Warchief, Duke, Avenger] Keep is in Signature. Play Nice Meow ;3

  3. #13
    Join Date
    Jul 2010
    Location
    the merge
    Posts
    2,733

    Default

    Quote Originally Posted by shadoxfoxspirit View Post
    assuming both sets of data are unique with no overlap... we are talking about 1240 rolls. I don't doubt players *have* gotten them; I do however find this incredibly curious as the chances for that many rolls to have a 1/100 odds to yield a net 0 after almost 13 full sets of 1/100 probability. The odds should remain consistent as stated and they just aren't. These two guilds may have to be the *luckiest* [in being unlucky] for these results.
    I’m not taking a position either way because the data we have is unreliable. There are undoubtedly biases in data reporting that we don’t understand and can’t control for, let alone appreciating the veracity of the claims. When I did my Jera rolling, I posted a picture of every roll I did. I think that’s the only real way to trust the data.

  4. Default

    I've added the numbers I found here on the forum:
    Actual Drop Rate on 710 chests: 0.0% legendary 6.8% epic 6.1% rare 18.2% uncommon 69.0% common
    Advertised rates 1% legendary 7% epic 14% rare 35% uncommon 43% common

    So the number of epics is pretty close, but the number of commons is very high, at the expense of rares and uncommons. I still haven't found anyone who rolled a legendary willing to admit as to how many chests they rolled.

  5. #15

    Default

    Quote Originally Posted by Eveline Lems View Post
    I've added the numbers I found here on the forum:
    Actual Drop Rate on 710 chests: 0.0% legendary 6.8% epic 6.1% rare 18.2% uncommon 69.0% common
    Advertised rates 1% legendary 7% epic 14% rare 35% uncommon 43% common

    So the number of epics is pretty close, but the number of commons is very high, at the expense of rares and uncommons. I still haven't found anyone who rolled a legendary willing to admit as to how many chests they rolled.
    Take this data and get your money refunded for fraudulent odds. Facebook will 100% refund you if you put in a claim. This is a massive amount of data and iWhatever would work with you. I'm going on the long shot and you are doing 80 per Favor Roll and iPhone App for 10% extra, blah blah 20ish rolls per $90. Takes 35 refills of $90 a pop ... this is shy of $3000 minimum. I'd be fighting it. Every single one of you spending like this and seeing the same results should do the same.

  6. Default

    Quote Originally Posted by pull lead View Post
    You are off with your calculations. The chances of getting a legendary are 1-(.99^n) where n is number of rolls attempted. Its only been about 20 years for me and formal statistics.

    You have about a 60% chance to get a legendary in 100 rolls.
    I'd be happy to stand corrected, but every roll has .99 chance to not be a legendary, independent of how many chests you roll.

  7. Default

    Quote Originally Posted by ☾ṡἅṋḭẗḕḕ☽ View Post
    Take this data and get your money refunded for fraudulent odds. Facebook will 100% refund you if you put in a claim. This is a massive amount of data and iWhatever would work with you. I'm going on the long shot and you are doing 80 per Favor Roll and iPhone App for 10% extra, blah blah 20ish rolls per $90. Takes 35 refills of $90 a pop ... this is shy of $3000 minimum. I'd be fighting it. Every single one of you spending like this and seeing the same results should do the same.
    fortunately I have not maxed my credit cards for this. Only 14 saved up free rolls in there are mine. I have just tallied up the results I found reported.

  8. #18

    Default

    Ok, this issue is revisited regularly and I do wish now that I had documented everything. For two reasons.

    1.) The moment the server maintenance hit yesterday I won a Layla that was not present after the maintenance was over. Prompt as I was in submitting a ticket, I've been asked for screen shots, which I did not know I would need to take.

    Click, win, click again, "server maintenance."

    2.) The most important detail in running these statistics is using proper methodology.

    What happens in the "data collection" process can quickly bias and/or skew results, and so on...

    I have routinely present statistical evidence that provides the strong suggestion that the drop rates have not been what was advertised, but in this case want to encourage you all to stop with poor analysis being provided.

    What is clearly evident in the posts so far is survivors bias and skewed toward grievance.

    If you want to conduct a proper analysis you must specify who is rolling, and how many rolls are being made in advance. Then those exact results must be reported. Allowing people to roll and then report is bad methodology.

    If people want to do a serious analysis, then I'm happy to do all of the math, but I'm hesitate to deal with any of the vocal posters who are simply wrong about things like degrees of freedom given the drop rates that are clearly advertised.
    Come say hello FB link
    Web 3 Keep link

    Quote Originally Posted by VersionsOfViolence View Post
    #1 it's not nice to be mean to ppl
    #2 nobody cares about da wilbsta

  9. #19

    Default

    Quote Originally Posted by Eveline Lems View Post
    fortunately I have not maxed my credit cards for this. Only 14 saved up free rolls in there are mine. I have just tallied up the results I found reported.
    Good to hear. I have disputed before on large sums with Castle Age a long time ago when I spent over their shady practices on not being transparent or things like this and got refunded. Just those in the same boat from the data gathering might want to consider if they continue not being transparent or addressing a possible issue in the rates. 700 is a large sample. It's a large reason I departed Castle Age for awhile and upon coming back don't spend like I used too.

  10. #20

    Default

    Quote Originally Posted by Da Wimsta View Post
    Ok, this issue is revisited regularly and I do wish now that I had documented everything. For two reasons.

    1.) The moment the server maintenance hit yesterday I won a Layla that was not present after the maintenance was over. Prompt as I was in submitting a ticket, I've been asked for screen shots, which I did not know I would need to take.

    Click, win, click again, "server maintenance."

    2.) The most important detail in running these statistics is using proper methodology.

    What happens in the "data collection" process can quickly bias and/or skew results, and so on...

    I have routinely present statistical evidence that provides the strong suggestion that the drop rates have not been what was advertised, but in this case want to encourage you all to stop with poor analysis being provided.

    What is clearly evident in the posts so far is survivors bias and skewed toward grievance.

    If you want to conduct a proper analysis you must specify who is rolling, and how many rolls are being made in advance. Then those exact results must be reported. Allowing people to roll and then report is bad methodology.

    If people want to do a serious analysis, then I'm happy to do all of the math, but I'm hesitate to deal with any of the vocal posters who are simply wrong about things like degrees of freedom given the drop rates that are clearly advertised.
    I am happy to help out too.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •