So sick of rigged drop rates

Replies

  • Monel
    2786 posts Member
    Options
    ZAP wrote: »
    Ultra wrote: »
    I find it ridiculous that cg takes the shak ti shards out of store nearer the event start for gas so that you have to try and buy crystals (or use up stock if you have enough) to buy those gamble packs where most of the time it rewards 5 shards, or spend £19.99 for 15 shards... yet when the event was not in the list you was putting shak ti shards in the stores. 3-4 days now no shak ti shards in store, absolute con. If I miss getting gas this time round so be it, but it’s not just coincidence that the shards disappear from normal stores when event is in events list. I’ve spoken to a few people and they’ve had the same with similar events, they pull the shards from stores and try selling packs. I know it’s a business but at least make it slightly fair and put the shards in the normal stores
    She showed up in shipments 6 hours ago for me

    Do you have her 7* already, cause if you do, then you seeing her in the shop means nothing in this convo.

    Again, I’m not agreeing with hell8mysoul, only thing I’ve said is that both sides have anecdotal arguments and zero independent data.

    And I think it was lead that said the likelihood of coordinating tracking of that data would be a nightmare and not very likely to happen.

    You would need an independent third party tracking system. Otherwise any knucklehead could just say they 25/100 or 13/100 or 256/1253748.
    It's just your belief system base on your perception of what's happening which is usually determined by your mood.
  • Ultra
    11502 posts Moderator
    edited March 2020
    Options
    There are so many constraints to Hell8MySoul’s argument

    “She’s not showing up in shipments”
    “If you have her at 7* it doesn’t count”
    “If you are 100 shards away it doesn’t count”

    You have to be at a 30 shard difference for him to find it acceptable which is dumb

    Shipments aren’t catered to each individual based on roster history
  • leef
    13458 posts Member
    Options
    Ultra wrote: »
    There are so many constraints to Hell8MySoul’s argument

    “She’s not showing up in shipments”
    “If you have her at 7* it doesn’t count”
    “If you are 100 shards away it doesn’t count”

    You have to be at a 30 shard difference for him to find it acceptable which is dumb

    Shipments aren’t catered to each individual based on roster history

    * during an event in which the character in question is required.
    Save water, drink champagne!
  • Options
    I finished the Shaak Ti farm a couple of days ago with 2 then 3/5 drops, and she was in the shipments at the time. GAS had been announced and was visible in the events tab. What means that I don't fall under the spell of CG's conspiracy?
  • ImaSmakya
    1068 posts Member
    Options
    Ultra wrote: »
    There are so many constraints to Hell8MySoul’s argument

    “She’s not showing up in shipments”
    “If you have her at 7* it doesn’t count”
    “If you are 100 shards away it doesn’t count”

    You have to be at a 30 shard difference for him to find it acceptable which is dumb

    Shipments aren’t catered to each individual based on roster history

    Or are they?

    Sorry, all the cool kids are saying it so I thought I’d get in on the action ;)
    https://swgoh.gg/p/319514721/
    DISCLAIMER: Post is subject to change.
  • TVF
    36603 posts Member
    Options
    Ultra wrote: »
    There are so many constraints to Hell8MySoul’s argument

    “She’s not showing up in shipments”
    “If you have her at 7* it doesn’t count”
    “If you are 100 shards away it doesn’t count”

    You have to be at a 30 shard difference for him to find it acceptable which is dumb

    Shipments aren’t catered to each individual based on roster history

    Occam's Razor would seem to apply.
    I need a new message here. https://discord.gg/AmStGTH
  • Tezza23
    88 posts Member
    Options
    TBH I can believe some of his numbers cause after getting 0 shards from 45 attempts and on top of that this particular character has just had packs released in store seems very suspicious, after all this is ea we are talking about , the company that Riggs football games in Fifa to balance games and entice spending on packs to win(momentum) .

    While I agree the numbers can work themselves out to 33% I also agree with this guy that sometimes they don’t and can’t deny some coincidental shady crap goes on. E.g 1 shard needed to 7* then 700 crystals later u get 3.
  • Options
    ZAP wrote: »
    leef wrote: »
    ZAP wrote: »
    TVF wrote: »
    TVF wrote: »
    TVF wrote: »
    TVF wrote: »
    I’ve spoken to a few people and they’ve had the same with similar events, they pull the shards from stores and try selling packs.

    This is a perfect example of "anecdotal."

    It's also wrong.

    Everyone has their own opinion and I respect that... but it’s what I’m experiencing and what others I’ve spoke to have also experienced

    This is still the perfect example of anecdotal.

    Your experience, and that of others you've spoken to, is nothing more than RNG.

    As always... you have no thought process other than “I’m right and your wrong” I assume you get updates when someone posts something on forums so you can straight away disagree with anything they say. I’ll wait for the “no your wrong” response

    The devs have stated in Q&A that they don't do things like this.

    If you think they're lying, ok. But your claim is baseless without data, of which you have none. It's 100% anecdotal. Sorry you disagree with that, but it doesn't make you right.

    And your evidence for them saying they don’t do this is them saying it in a q&a?

    I mean, also because I see shards show up in the stores during events, even have bought some myself.

    So now who's right?

    I’m not disagreeing with you on whether what hell8mysoul is wrong, but technically all you have is anecdotal evidence as well. The devs saying something in a Q&A is not independent data.

    They’ve done this exact type of thing in other games they’ve developed, so definitely have the capability. Whether what they say in a Q&A is honest or not I think is very much debatable.

    Is there any proof that they've done this exact type of things in other games they've developed?

    Diablo 3 had drop rates change based on user experience, in that case it was actually in favor of the players who experienced “dry spells”. This was admitted to and if they can do it in the players favor then they can certainly do it in their favor.

    Anthem had some highly suspect things happening to drop rates. While nothing was admitted to by the company, things definitely changed after a fair amount of complaining and players sharing their data.

    FIFA, highly suspect as well

    The bottom line is EA holds some very complex patents, which enable them to manipulate drop rates based on a vast number of individual player conditions.

    They’ve admitted to its use in Diablo 3 in favor of players, so they don’t just hold the patents, they’ve actually used changing drop rate mechanics in live games.
    That’s true about the patents, BUT the devs stated more than once that none of those systems are being used in this game. (Not searching for evidence on my phone, maybe someone will encourage my laziness).

    And that’s where the forum divides into 2 groups of people:

    - People who believe there’s nothing fishy going on, understand the nature of probability and RNG

    - People who don’t believe the devs, don’t believe the data that hundreds of users have tracked but then instantly believe posts like the first one in this very thread (which is obvious made up, embellished claptrap)
  • Ultra
    11502 posts Moderator
    Options
    Developers don’t lie to the player base, they have no reason to lying about drop rates. They can admit manipulating drop rates and we can’t do anything other than deal with it

    Skeptical players who don’t trust CG have tracked drop rates for toons during events requiring those toons and turns out the drop rate is consistently 33%

    You have developers saying they don’t manipulate drop rates

    You have players who don’t trust CG doing scientific investigation saying the drop rate is still 33% throughout the life cycle regardless of any ongoing events
  • Options
    If CG tried to manipulate drop rates people might accidentally get double drops of KAM or something
  • Monel
    2786 posts Member
    Options
    cxka9reys1te.jpg
    80% drop rate confirmed with actual data on Hyena Bomber!
  • leef
    13458 posts Member
    Options
    ZAP wrote: »
    leef wrote: »
    ZAP wrote: »
    TVF wrote: »
    TVF wrote: »
    TVF wrote: »
    TVF wrote: »
    I’ve spoken to a few people and they’ve had the same with similar events, they pull the shards from stores and try selling packs.

    This is a perfect example of "anecdotal."

    It's also wrong.

    Everyone has their own opinion and I respect that... but it’s what I’m experiencing and what others I’ve spoke to have also experienced

    This is still the perfect example of anecdotal.

    Your experience, and that of others you've spoken to, is nothing more than RNG.

    As always... you have no thought process other than “I’m right and your wrong” I assume you get updates when someone posts something on forums so you can straight away disagree with anything they say. I’ll wait for the “no your wrong” response

    The devs have stated in Q&A that they don't do things like this.

    If you think they're lying, ok. But your claim is baseless without data, of which you have none. It's 100% anecdotal. Sorry you disagree with that, but it doesn't make you right.

    And your evidence for them saying they don’t do this is them saying it in a q&a?

    I mean, also because I see shards show up in the stores during events, even have bought some myself.

    So now who's right?

    I’m not disagreeing with you on whether what hell8mysoul is wrong, but technically all you have is anecdotal evidence as well. The devs saying something in a Q&A is not independent data.

    They’ve done this exact type of thing in other games they’ve developed, so definitely have the capability. Whether what they say in a Q&A is honest or not I think is very much debatable.

    Is there any proof that they've done this exact type of things in other games they've developed?

    Diablo 3 had drop rates change based on user experience, in that case it was actually in favor of the players who experienced “dry spells”. This was admitted to and if they can do it in the players favor then they can certainly do it in their favor.

    Anthem had some highly suspect things happening to drop rates. While nothing was admitted to by the company, things definitely changed after a fair amount of complaining and players sharing their data.

    FIFA, highly suspect as well

    The bottom line is EA holds some very complex patents, which enable them to manipulate drop rates based on a vast number of individual player conditions.

    They’ve admitted to its use in Diablo 3 in favor of players, so they don’t just hold the patents, they’ve actually used changing drop rate mechanics in live games.

    following your own reasoning, this is anecdotal evidence as well. It's not independent data either.
    You're basically saying you can't take the devs' word over anecdotal evidence because there's anecdotal evidence that devs are doing this sort of stuff in other games.
    Even if there's proof that devs of another EA game have done this, which may be out there eventhough you haven't provided it to back up your claim, it's still only anecdotal evidence in this discussion. Just because they've done it in one of their games doesn't mean/prove they're doing it in more of their games. I do however agree that it's an argument in favour of not taking the devs at their word
    Also, imo you can't really use the devs of another EA(?) game claiming they've used droprate manipulation in favour of the players as an argument to cast doubts on the claims by the devs of this game that they're NOT using droprate manipulation.
    So yea, where does that leave us? We're doomed to being the victim of our own biases due to the lack of proof?
    Save water, drink champagne!
  • Options
    I'm not usually a CG defender but it really is just rng. I've farmed the same toons as my guildmates with exceptional speed in comparison to them and on others it's taken months longer than them. I mean don't get mewrong i think that the drop rates should probably be worked out a big to guarantee at least one shard per x attempts if none were gotten but I don't think that it's rigged.
  • Ultra
    11502 posts Moderator
    Options
    g19ol0c01pex.png

    Yep she’s not removed from shipments yet

    Since @Hell8MySoul isnt posting his shipment updates, I’m assuming he’s getting Shaak Ti too
  • MaruMaru
    3338 posts Member
    edited March 2020
    Options
    ZAP wrote: »

    However the fact does remain that both sides only have anecdotal evidence. At best we have the drop rate tracking project with multiple player contributions, which I actually believe, but player tracked data is not hard independent proof that no drop rate manipulation is occurring. The drop rate tracking projects were straight drop rate based on attempts, with no other data points.


    It's true that the tracking projects don't go into specific slices of time periods or any other hypothesis of drop rates not being constant. However CG never promised or published any particular drop rate. It's the player generated data that has been consistent across lengthy periods of time (i.e. 2019 vs 2018 etc), across players and across different toons, different (same variety of) gear. I also conducted my own tracking without publishing until I was confident that I converge to the same outcome.

    Anyway take any hypothesis of variable droprate and assume it's true. i.e. say that toon shard drop rates are %40 at some periods and %20 on other periods or there's some algorithm that reduces your droprates depending on your farming state and compansates back at other times...the averages you are getting will still converge to the averages over time that's gathered under the assumption of constant droprates over time. So you are still getting %32-33 toon shards and %20 purple-yellow gear over lengthy time periods. So what does really matter; what you get in short bursts or over time?

    Lastly for full disclosure we can take another hypothesis for granted; the idea that these 1000s of independent trackers only represent a portion of the playerbase and rest of the playerbase may be getting different droprates.

    - The trackers may either be some secret cult tweaking their droprates or they made up their mind beforehand how their end result should look so this is a very selective portion of players. I don't know how to address that. What I can say is I've seen many players with different presuppositions start tracking their data at some point (some even out of spite or to disprove the supposed community averages). They also ended up seeing they converge to larger pools of data that have been present for a long time and got convinced on their own.

    - Some targeted portion of the playerbase has different droprates than the rest (depending on their spending or whatnot). Maybe you are one of those. None of us can fully refute that hypothesis, but you. Track your data day to day, toon to toon and present to us. Yet to see that outlier person/people.


    Outtake: I think I've been pretty clear in how much I said is anectodal and in what context. To me it's obvious drop rate trackers vs. believers of numerous hypothesis are waging assymetrical warfare and doesn't have comparable value in their word. I'm sceptical about my own established position and welcome any further data, especially specialised data which I would try to contribute if there was a methodical project worth my time.

    Believe what you believe, let that sink in. THEN track your data.

  • ImaSmakya
    1068 posts Member
    Options
    MaruMaru wrote: »

    Anyway take any hypothesis of variable droprate and assume it's true. i.e. say that toon shard drop rates are %40 at some periods and %20 on other periods or there's some algorithm that reduces your droprates depending on your farming state and compansates back at other times...the averages you are getting will still converge to the averages over time that's gathered under the assumption of constant droprates over time. So you are still getting %32-33 toon shards and %20 purple-yellow gear over lengthy time periods. So what does really matter; what you get in short bursts or over time?

    The fact that it averages out over time is great, I have no doubt that that is happening even if there is in fact some sort of variable drop rate based on a given set of states.

    I seem to recall when that patent was uncovered about this, Carrie (think it was her) said they didn't use it in the game and didn't even have the technology to implement it. I had to chuckle a bit about the technology comment as this is something you would probably code with a simple state machine. The longest part of the process would likely be figuring out what data would be needed to define each state and transition, the code would be pretty simple.

    As for the data itself, there is the data that CG could easily define themselves like
    • how close a toon is to 330 shards
    • is that toon required for an event in the next 2 weeks
    • what state are other characters for that event in
    • etc...

    Then there is the player behavioral data, at what point along that 0-330 spectrum does the player
    • start doing hard node refreshes
    • buy packs from the store
    • buy crystals
    • etc...

    Once they define the spending patterns for that part of the player base which is more prone to spending, which I'm pretty sure they already have, it would be simple enough to build frustration mechanics based on a variable drop rate into the game to optimize player spending across the entire player base.

    I'm not saying any of this is currently in the game in any form, but CG saying they don't have the technology to do it is a bit far fetched. Its a six pack, a box of pizza pockets and a few hours to myself to get a working prototype done.
    https://swgoh.gg/p/319514721/
    DISCLAIMER: Post is subject to change.
  • MaruMaru
    3338 posts Member
    Options
    ImaSmakya wrote: »
    MaruMaru wrote: »

    Anyway take any hypothesis of variable droprate and assume it's true. i.e. say that toon shard drop rates are %40 at some periods and %20 on other periods or there's some algorithm that reduces your droprates depending on your farming state and compansates back at other times...the averages you are getting will still converge to the averages over time that's gathered under the assumption of constant droprates over time. So you are still getting %32-33 toon shards and %20 purple-yellow gear over lengthy time periods. So what does really matter; what you get in short bursts or over time?

    The fact that it averages out over time is great, I have no doubt that that is happening even if there is in fact some sort of variable drop rate based on a given set of states.

    I seem to recall when that patent was uncovered about this, Carrie (think it was her) said they didn't use it in the game and didn't even have the technology to implement it. I had to chuckle a bit about the technology comment as this is something you would probably code with a simple state machine. The longest part of the process would likely be figuring out what data would be needed to define each state and transition, the code would be pretty simple.

    As for the data itself, there is the data that CG could easily define themselves like
    • how close a toon is to 330 shards
    • is that toon required for an event in the next 2 weeks
    • what state are other characters for that event in
    • etc...

    Then there is the player behavioral data, at what point along that 0-330 spectrum does the player
    • start doing hard node refreshes
    • buy packs from the store
    • buy crystals
    • etc...

    Once they define the spending patterns for that part of the player base which is more prone to spending, which I'm pretty sure they already have, it would be simple enough to build frustration mechanics based on a variable drop rate into the game to optimize player spending across the entire player base.

    I'm not saying any of this is currently in the game in any form, but CG saying they don't have the technology to do it is a bit far fetched. Its a six pack, a box of pizza pockets and a few hours to myself to get a working prototype done.

    That patent doesn't mean all that much to me. It just gives us the ground for a hypothesis which can be tested (a difficult process), not a default state to believe in. CG not using this and just going with constant drop rates fits all my other observations,biases though. If this player harming thing is used it has been provable and they openly denied it. Meanwhile they never promised any particular droprate, just that it never changed so far and it's constant (+they'll let us know if they decide to change it). There was nothing withholding us to conduct a community project and prove time specific droprates with statistical significance (you can find some attempts at reddit conducted with small samples that lacks statistical significance and have low confidence thresholds, some attempted it). In that case it would become a pr nightmare for them that's really hard to come back from.

    What really convinces me this shouldn't be the case is the extra profit they can reap from it is miniscule. And there's simply nothing that stops them from doing&admitting it. There is neither a law boundary or anything we can do if they just came out and said we do this, I doubt it will cause people to quit in spades either. People simply love to assign bad traits into invisible parameters.

    The real ripoff mechanics in this game are right in our faces, they are in hundreds. They are easily able to blurt out new streams of revenue since the beginning (while each of them contains some risk). Some toons were for only sale at the beginning, they tried championships, they did raids where each time people went bonkers, they kept increasing special toon event needs where we finally reached out of whack GLs, new gear tiers, relics. They can keep doing this forever as long as enough of us keep participating. Our opinions are secondary, merely a sidenote compared to our actions for CG.

    Yet all of us still partake in new model of gaming and with millions that passed through this lane they managed to reap something close to 1B$. The motive for this particular "crime" is lacking.
  • Options
    Motive. Fewer and fewer krakens, whales and dolphins spending money. Revenues are sliding. Swgoh is not anymore in top 10 revenue generating games in any appstore/google play tracking daily/monthly lists.

    For me frustration based rigging mechanics are deal breaker.
  • Kyno
    32087 posts Moderator
    Options
    Motive. Fewer and fewer krakens, whales and dolphins spending money. Revenues are sliding. Swgoh is not anymore in top 10 revenue generating games in any appstore/google play tracking daily/monthly lists.

    For me frustration based rigging mechanics are deal breaker.

    @hellbournesmith sure they would love to be in the top 10, but they dont "need to be", they are still making money.

    Nothing is rigged and this has been proven time and time again, by simply tracking your data as other players have.

  • Options
    Motive. Fewer and fewer krakens, whales and dolphins spending money. Revenues are sliding. Swgoh is not anymore in top 10 revenue generating games in any appstore/google play tracking daily/monthly lists.

    For me frustration based rigging mechanics are deal breaker.

    This post is so old
    I wasn't writing haikus.
    Drop rates still aren't rigged.
  • hellbournesmith
    6 posts Member
    edited October 2020
    Options
    Everything is rigged for their pleasure. Way of slowing down progress so they don't have to create real content.

    For all playerbase it might even out at the given 33%, but for individual players it will be much higher or lower depending on hidden modifiers that are encouraging players to spending or frustrating to spending.
  • hellbournesmith
    6 posts Member
    edited October 2020
    Options
    Kyno wrote: »
    @hellbournesmith sure they would love to be in the top 10, but they dont "need to be", they are still making money.

    Like most of the humans are kind, humble and practise moderation. They are most likely working for monastery not for corporation. Somehow seems like it's more of a virtual goods mobile store rather than game.

    Only thing that in the end matters that the dole purpose for company is to generate value for owners.

    And for petty human only thing that matters is quarterly or yearly bonuses.

    Greed. Getting the short term results at the expense of long term longevity.

    So actually no one needs to get more than 50k annually, but haven't met anyone who has turned down a raise or wouldn't have used questionsble methods if the available reward is temoting enough.
  • Options
    Everything is rigged for their pleasure. Way of slowing down progress so they don't have to create real content.

    For all playerbase it might even out at the given 33%, but for individual players it will be much higher or lower depending on hidden modifiers that are encouraging players to spending or frustrating to spending.
    I’ll take “made up conspiracy nonsense” for $100
  • Options
    Everything is rigged for their pleasure. Way of slowing down progress so they don't have to create real content.

    For all playerbase it might even out at the given 33%, but for individual players it will be much higher or lower depending on hidden modifiers that are encouraging players to spending or frustrating to spending.

    Drop rate's been tested
    By individuals too,
    You're just paranoid.
  • hellbournesmith
    6 posts Member
    edited October 2020
    Options
    30 sims 2 shards of MM. If you all are convinced flavour of the month multiplier is nonexistent and my 10 day cold streak evens out.
Sign In or Register to comment.