- something rotten in Denmark?

Discussion in 'General Discussion' started by LoserSlick, Apr 20, 2017.

  1. LoserSlick

    LoserSlick Devotee of the Blood Owl

    i can't imagine what else you think incremental adjustments at the margin would refer to?

    maybe let's just wait for these new changes to take effect and see how they do - i'm sure we both want the same thing :)
     
  2. badgerale

    badgerale Warchief of Wrath

    I read this as if hhhhhhttttt was the sound you made while thinking. It amused me.
     
    SPiEkY and NevrGonaGivUup like this.
  3. Pipster

    Pipster I need me some PIE!

    Honestly, I've been following this thread from the beginning, and I think that's the perfect word to describe how you're being. Your insistence to use all kinds of jargon and tautology to get your point across (which is incredibly unclear now because of that fact) has been so frustrating to read, intentional or not.

    Amazed Sokolov has the patience to give you a serious reply each time.
     
    Tweek516 and SPiEkY like this.
  4. LoserSlick

    LoserSlick Devotee of the Blood Owl

    lol fair enough, but your powers of observation are underwhelming you realize (or don't). what you've just followed "from the beginning" was my overexertion in making every appropriate effort to transcend an awkward situation so as not to wrongfully lampoon a gentlemen who may well have had nothing to do with any presumed 'wrongdoing.'

    as far as serious replies go, this one's on the house. enjoy :)

    maybe do all three of us a favor and let this thread sink as it should have been a private discussion to begin with. the only thing retarding our dialogue is that it's public
     
  5. LoserSlick

    LoserSlick Devotee of the Blood Owl

    lol okay, you know what - no. you've officially flipped my 'gentleman switch' to the off position.

    since you are clearly confused here - to the point where you apparently feel obliged to insult me - maybe i can delineate what's happened here and put it in chronological terms you might actually be able to wrap your head around.

    from the top:

    a) i purchase 27 packs with zero hits, exotic or otherwise (or 'X in 27' as sok puts it, with respect at least to exotics)

    - post #7 figures that probability at 6% based on exotics alone
    - realize here that accounting for Y-leg at 5% lowers that 6% figure even further
    - post #20 provides google sheets to demonstrate X in 30 from a 100 sample (keep in mind that while to round X in 27 up would theoretically be to my advantage it does not compensate for the Y-leg at 5% margin... though this still doesn't matter with respect to the theoretical logic at issue)
    - where statistical probability and improbability are relative, that is a separate argument from whether or not something is statistically significant. that is to say that because something is simply statistically possible does not necessarily make it probable. by this reasoning, my run of zero hits at 27 X could easily have been 37 X or even 47 X and all sok would have to do is run as many tests as were necessary - and/or with larger samples - to prove that "feelsbadman" exists *feelsbadman will always exist*

    b) (here's the onion) - i make a public complaint to gripe about my "feelsbadman" 'X in 27' luck. what's more is that i did so through the veiled threat of collecting data from other players who may have had 'similar experiences.' lol obviously, that would have been completely unfair data to collect - but, as i already admitted, i was simply frustrated (and rightly so)

    c) (here's the onion of that onion) the very next pack i purchased following the previous 27 (and subsequent complaint) included 2 exotics and 2 legs

    - post #17 figures this at a 1.1% probability - tho this might only represent 4 X (as opposed to 2 X at 10% and 2 Y at 5%) in which case that figure would be even lower

    so the point of contention here is not based on the unfair premise of any specific 37-pack-long sequence of results, but rather on two back-to-back anomalies. in laymans, imagine two of those jumps from post #20 occurring not only right next to each other but specifically one followed by the other. now that would require substantially more complicated math than anything currently in this thread. the fact that i simply don't feel like doing it doesn't mean i don't have an argument - let alone that i'm wrong. lol if he wants to, sok's welcome to it

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    the cadence and subsequent progression of the exchange:

    i had no idea they had already acknowledged and taken measures to correct this problem. which is why i quickly apologized and said that this was an unnecessary thread. because this was the only point i was making. it was an attempt at constructive feedback for something i didn't know they were already aware of. get it?

    the end implication because this thing went off the rails is that team pox in all likelihood saw my feelsbadman run and intervened. the unflattering (AND NOW VERY UNNECESSARY) byproduct of that implication would be that while a), team pox isn't full of a-holes who are trying to nickle and dime us, b) they don't know what they are doing. which is unfair

    realize my sort of 'tip-toeing' around over these 4 pages was not meant to provide "Pipster" with good reading. it was meant to simultaneously close a can of worms and not thumb a nose at a bunch of guys working to constantly to improve a game we like playing.

    thanks and apologies macca/sok

    /end thread
     
  6. Sokolov

    Sokolov The One True Cactuar Octopi

    I'd like to note that that this is not what I did at all.

    I took the first 3 samples and made static tabs out of them.

    I also left a RANDOM tab that generates a new set of numbers every time you refresh the page - which in the vast majority of cases will generate at least one long run 20+ of no exotics, and sometimes two (but it mich less common to see 30 and rare to see a 40, and I have never seen a 50+).

    That's why I left the RANDOM tab there, so you can generate and look as many as you want yourself.


    I don't see how collecting data is a bad thing. If you had gotten no exotics in say, 40 or 50 packs (something MUCH, MUCH rarer and less likely, which, by the way, happens rarely compared to runs of 20-30 in the 100 pack generator), and dozens of people showed up to say "hey, we are seeing the same thing all the time" that would be a different scenario and I would have to investigate more. The fact is, when there are hundreds/thousands of packs sold weekly, it is expected that some people will be very lucky and some people will be very unlucky - even if being very lucky and very lucky aren't super common, but if the experiences of many people are experiencing are only towards one side, it would indicate a problem - but one person having a string of no exotics isn't something especially problematic.

    So again, if other people want to chime in with their experiences, they should absolutely do so. I am just doing the math here, but it doesn't mean that the system itself doesn't have an actual problem.


    You are still making the mistake of thinking the ORDER of events matter - they don't. These are independent variables.

    I have explained in various ways now and you continue to make the argument, without proof, that what you experienced is extraordinary.

    In short, there is no "auto fix" in the system. The packs generate independently of one another. You are focusing too much on the order of events - which, as demonstrated, isn't specifically related to the variances in the outcomes. The fact that you got more EXO/LEGs later isn't an anomaly on top of another anomaly, it's just a larger sample size. YOU care about the order - randomness doesn't.

    If the physical pack argument didn't convince you, I honestly don't know what will, since that clearly demonstrates that the order in which you open the pack has nothing to do with the outcome, but some orders will seem more "anomalous" than others even tho probability of each is EXACTLY the same as ANY OTHER.
     
    Last edited: Apr 29, 2017
    Etherielin, Tweek516 and SPiEkY like this.
  7. themacca

    themacca Master of Challenges

    im confused why did i get an apology.
     
    SPiEkY likes this.
  8. LoserSlick

    LoserSlick Devotee of the Blood Owl

    - i didn't mean unfair in terms of ethics; i meant unfair in terms of logic. collecting feelsbadman data solely is different from collecting data from everyone. unless your intention was to examine frequency of said behavior

    - i've conceded several times over that your argument is completely valid. lol not just that it's valid, but simply stating a fact. the impasse occurs where you rightly contend that there is no mathematical equation for "making a public complaint" - and there isn't one. the only thing you can do beyond that point is to study the behavior of results. something you're unwilling to do. there are entire college courses focused on trend and random probability.
     
  9. Sokolov

    Sokolov The One True Cactuar Octopi

    I am going to try again.
    1. You are focusing too much on your specific sequence.
    2. Any specific is rare because it's specific.
    3. Any specific sequence with the same number of EXO/LEGs will have the same % chance of occurring as any other.
    We can demonstrate this with a simple example (15% chance of EXO/LEG):

    upload_2017-4-29_23-5-6.png

    Here, we see that even tho the total number of EXOs are the same, each specific sequence of 3 3-pack results has a 12.3% chance of occurring.

    But notice that the 3 ways that these sequences can combine to make 1 EXO/LEG in a pack of 9 are exactly 1/3rd of 36.79%.

    This is because ANY of those specific sequences of 0 in 3, 0 in 3 and 1 in 3 can combine to make the 1 in 9.

    In other words, getting 1 EXO in 9 packs is ~37%, but getting it in a SPECIFIC way is much less common, relatively speaking, and ANY specific way defined is the same chance as any other way that fits the same criteria (bundling things differently WILL change things somewhat, but that's because you are changing the possible permutations to achieve the goal, but the overall % won't change once we add up all the actual permutations).
     
    Last edited: Apr 29, 2017
  10. LoserSlick

    LoserSlick Devotee of the Blood Owl

    thanks was to you

    apology was to sok

    :)
     
  11. Sokolov

    Sokolov The One True Cactuar Octopi

    This is NOT what you JUST said:

    In this quote, you specific state that you have math that can prove it, but you are unwilling to do it. Either you think the math favors you or it doesn't. You can't have it both ways and go back and forth on it depending on which argument you feel like making at the time.

    My degree is in Economics. I did this for stuff for years.
     
    Last edited: Apr 29, 2017
  12. Sokolov

    Sokolov The One True Cactuar Octopi

    Here's the bottom line:
    1. You are wrong on the math. If you disagree and I am, in fact, missing something, please do the math and show me how your situation is so anomalous, because otherwise I don't know what you are talking about.
    2. We agree that long runs of not getting anything isn't fun, and as already discussed we have plans to address this, but it isn't a math problem.
    3. I don't know why you continue to accuse me of "unwilling" to look at this issue. I am literally discussing the whole thing in this and other threads and have an entire plan to deal with this issue.
     
    Last edited: Apr 29, 2017
  13. Sokolov

    Sokolov The One True Cactuar Octopi

    This reply makes zero sense to me.

    You are the one who said YOU made a "threat" of collecting data. I am only stating that I don't consider that a threat. In fact, if people have bad experiences, I want to know about them, especially if many are having the SAME bad experience.

    Is there language barrier issue here? Because I feel like half the time you aren't even talking about the same thing I am and just talking sideways.
     
  14. SPiEkY

    SPiEkY King of Jesters

    This thread has been awesome.

    Seriously, 10/10 for the sheer effort alone, 11/10 for the execution.
     
    Last edited: Apr 30, 2017
    Etherielin, Tweek516 and badgerale like this.
  15. Sokolov

    Sokolov The One True Cactuar Octopi

    Ok, at this point, I think there's not much productive discussion here. I have explained the math/logic side enough to anyone who was interested in the topic, and continued discussion is probably not super productive and feel like I am just picking on you. So I am just going to let this go, because no matter whether I convince you on the math point, the overall feedback is appreciated and has been received.

    Maybe I am wrong on the math, but I am pretty confident I am right - because honestly, this isn't actually very complex stuff and doesn't require complex math (it is DEPENDENT event probability that is vastly more complicated since each outcome impacts all future outcomes), it's just not intuitive - which is often the problem with this sort of math.

    Regardless of who is right or wrong on the specific nature of the original issue - it is true that no one likes not getting anything over and over, which is we have the Rune Forge, and why I have outlined a plan to deal with the issue, starting with simply having smaller packs and higher gold to rare/exo/leg ratios and eventually having the "pity meter" mechanic implemented.
     
    Last edited: Apr 29, 2017
  16. LoserSlick

    LoserSlick Devotee of the Blood Owl

    k sok... i'm fairly certain you've studied advanced calculus before. applying both derivative theory and range analysis to behavior and probability is more difficult than anything you've posted on the basic mathematics of randomness. not only are you now starting to misrepresent what i said, but it is completely unfair to take a put up or shut up approach given the fact. in any case, you are grossly over-applying "random probability" to justify and dismiss literally any given thing at any given time. that is not rational.

    c'mon man...
     
  17. Sokolov

    Sokolov The One True Cactuar Octopi

    Ok, I have to respond to this at least...

    I made my arguments, then provided rigorous proof and demonstrated my claims both empirically and computationally - in multiple ways.

    You have basically repeated your claims and simply said "I am not wrong even if I won't demonstrate why my claims are correct."

    And you say *I* am being unfair to ask you to provide some demonstration of what you are saying so I can understand what you are talking about?

    And now you claim I am misrepresenting you when you say "in any case, you are grossly over-applying "random probability" to justify and dismiss literally any given thing at any given time." Which is an absurd claim. I am using math to demonstrate why I think you are wrong on the specific claim you are making that your scenario is anomalous. I am not using probability to dismiss "literally any given thing at any given time." In what way have I done that? Have I used it to dismiss the earth isn't round? Have I used it to dismiss that I am alive? Have I used it to dismiss that the Holocaust happened? No, I am not LITERALLY doing ANY of those things - therefore, I am not LITERALLY dismissing ANY thing at ANY time. Heck, I haven't even used it to dismiss that you are wrong to feel bad/weird about what happened - I have acknowledged that this is an acceptable reaction.

    Talk about misrepresentation!
     
    Last edited: Apr 29, 2017
    Etherielin and Tweek516 like this.
  18. Sokolov

    Sokolov The One True Cactuar Octopi

    And yes, there absolutely IS much more complicated math one can do with probability... but BINOMIAL distribution is sufficient in this case because it fulfills all the characteristics required:
    1. The event we are discussing X number of trials/runs
    2. We are discussing 2 outcomes per trial/run (success or failure)
    3. The probability of success is equal on every trial/run
    4. Each trial/run is independent
    We do NOT need more complicated math when something fulfills this set of criteria.
     
    Etherielin and Tweek516 like this.
  19. LoserSlick

    LoserSlick Devotee of the Blood Owl

    well there's clearly only one thing left we can do sok
     
  20. LoserSlick

    LoserSlick Devotee of the Blood Owl

    we have to arm wrestle.
     
    SPiEkY likes this.

Share This Page