I'm still not convinced ...

Prev1
Stenun
328 posts Member
edited December 2018
There is a claim I have seen repeated several times on this forum that the percentage chance of a node dropping its primary reward is 30%.
I remain unconvinced.
Firstly, if this was the average then I would have roughly as many days above that % as I do below. That's how an average works - it's in the middle. If I see 5 days in a month dropping less than that then surely I should see at least a few days per month dropping more? But I never do.

Also, it seems obvious to me that certain nodes have different drop rates from certain other nodes. Is the 30% figure supposed to be the average for each individual node or is it the average across ALL nodes? Because I'd be more willing to accept the latter but some people seem to insist it's the former.

Anyway, I'm curious as to who else agrees with me that the 30% claim is wrong.

Replies

  • I feel a lot of the time its a confirmation bias, somedays i get more yet others i get bad but like dveryone else i can be quick to forget the good days.

    Bassically the moral of the story is just get on with the farm and deap with what you got without moaning :-)
  • I'm not moaning about what drops I get.
    I'm challenging the forum figure of 30%.

    There's a difference. :-)
  • TVF
    9388 posts Member
    It's not an average. It's an individual rate on each attempt. You haven't tracked anything, let alone a statistically relevant sample size.
  • robjay04
    82 posts Member
    edited December 2018
    I tracked zeta drops for a month and it was right at 30 percent.

    Sure you will go 2-3 attempts straight getting zero but then you’ll get multiples a few days to even it out.

    Some nodes do seem tougher but I am certain 30 percent is accurate.

    Some nodes do seem better than others though but the worst you can get is 30%
  • I feel a lot of the time its a confirmation bias, somedays i get more yet others i get bad but like dveryone else i can be quick to forget the good days.

    Bassically the moral of the story is just get on with the farm and deap with what you got without moaning :-)

    Possibly but there are absurdly difficult farms that are known, such as blaze, so that leads me to believe there likely are slightly different drops on "premium" characters
  • Ill answer that question as good as i can to give you a satisfying answer. First of all - 30% is overall - there were some players in the good old days that actually tracked theyre droprrates over a year - and most likely 30% is the average droprate. There are some notes that are below or above 30% - a good example is mission vao - from my perspektive, and from many of my guildmates responses the farm was really hard. T3M4 on the other hand - feels a lot smoother - i get this feedback from my guild members aswell. So basicly mission will be around 25-27% and T3M4 at 33-35%. This is just and example - i dont recorded my numbers. But overall from my experience ...30% drop chance on Hard nodes pretty much hit the nail.....pls try to remember that your mind will always remember the ,,bad'' drop days - and will forget the ,,good'' drop days a lot sooner ;) thats how humans work xD we stick to the bad things - not to the good ahahah :D
  • Stenun wrote: »
    There is a claim I have seen repeated several times on this forum that the percentage chance of a node dropping its primary reward is 30%.
    I remain unconvinced.
    Firstly, if this was the average then I would have roughly as many days above that % as I do below. That's how an average works - it's in the middle. If I see 5 days in a month dropping less than that then surely I should see at least a few days per month dropping more? But I never do.

    Also, it seems obvious to me that certain nodes have different drop rates from certain other nodes. Is the 30% figure supposed to be the average for each individual node or is it the average across ALL nodes? Because I'd be more willing to accept the latter but some people seem to insist it's the former.

    Anyway, I'm curious as to who else agrees with me that the 30% claim is wrong.

    My experience is that purple gear from LS/DS nodes drops at a rate of ~25%. Character shards drop consistently at roughly 33%. But I never did a comprehensive analysis because knowing any real numbers does not change anything.
  • Darknesswon
    548 posts Member
    edited December 2018
    Ill answer that question as good as i can to give you a satisfying answer. First of all - 30% is overall - there were some players in the good old days that actually tracked theyre droprrates over a year - and most likely 30% is the average droprate. There are some notes that are below or above 30% - a good example is mission vao - from my perspektive, and from many of my guildmates responses the farm was really hard. T3M4 on the other hand - feels a lot smoother - i get this feedback from my guild members aswell. So basicly mission will be around 25-27% and T3M4 at 33-35%. This is just and example - i dont recorded my numbers. But overall from my experience ...30% drop chance on Hard nodes pretty much hit the nail.....pls try to remember that your mind will always remember the ,,bad'' drop days - and will forget the ,,good'' drop days a lot sooner ;) thats how humans work xD we stick to the bad things - not to the good ahahah :D

    Mission seems harder cause the node costs more so you get less attemps per refresh, otherwise its the same drop rate.
  • I've tracked a 10-energy Cantina node, a 16-energy Cantina node, and now a 12-energy Cantina node. The first two were 37.90 and 38.17% respectively, the current one is sitting at 33.00% but I'm only about a third complete.

    That's a total of 1,688 tracked attempts so far.
  • Nauros
    2120 posts Member
    Just to avoid confusion, you talk about "primary reward", but the 30% probability pertains only to shards. Gear has different probabilities, depending I think on the type of node where it appears first.
    Revive the stores, the game needs it.
  • Nauros wrote: »
    Just to avoid confusion, you talk about "primary reward", but the 30% probability pertains only to shards. Gear has different probabilities, depending I think on the type of node where it appears first.

    Not sure it's the node as much as the gear piece itself. Is it not: gold rarer than purple, which is rarer than blue?
  • Nauros
    2120 posts Member
    Nauros wrote: »
    Just to avoid confusion, you talk about "primary reward", but the 30% probability pertains only to shards. Gear has different probabilities, depending I think on the type of node where it appears first.

    Not sure it's the node as much as the gear piece itself. Is it not: gold rarer than purple, which is rarer than blue?

    I think I read somewhere that it's the node, but the two mostly overlap anyway. You get green gear on easiest nodes and gold on hardest.
    Revive the stores, the game needs it.
  • Waqui
    4823 posts Member
    Stenun wrote: »
    There is a claim I have seen repeated several times on this forum that the percentage chance of a node dropping its primary reward is 30%.
    I remain unconvinced.
    Firstly, if this was the average then I would have roughly as many days above that % as I do below. That's how an average works - it's in the middle. If I see 5 days in a month dropping less than that then surely I should see at least a few days per month dropping more? But I never do.

    No, that's not how it works.
    The approximate 30% drop rate is measured in shards dropped per battles won/simmed. How many days you're above the average 30% and how often you're below doesn't matter at all.

  • Liath
    3065 posts Member
    People don’t just say this because it feels right or something. There are people that have actually tracked the numbers across large sample sizes, and that’s where the numbers come from. The people that insist the numbers are wrong have never tracked any significant number of drops.
  • TVF
    9388 posts Member
    I feel a lot of the time its a confirmation bias, somedays i get more yet others i get bad but like dveryone else i can be quick to forget the good days.

    Bassically the moral of the story is just get on with the farm and deap with what you got without moaning :-)

    Possibly but there are absurdly difficult farms that are known, such as blaze, so that leads me to believe there likely are slightly different drops on "premium" characters

    This post has at least two problems.
  • Kyno
    19851 posts Moderator
    No one will be here to convince you.

    Record the data yourself and watch the magic happen.
  • Kyno wrote: »
    No one will be here to convince you.

    Record the data yourself and watch the magic happen.

    I didn't ask anyone to convince me. I asked who agreed with me. :-)
  • Stenun wrote: »
    Kyno wrote: »
    No one will be here to convince you.

    Record the data yourself and watch the magic happen.

    I didn't ask anyone to convince me. I asked who agreed with me. :-)

    No one agrees with you either, this topic has been raised and beaten into submission many Times. Track your drops, and you will see that it’s around that 30% mark,

    Or, use the search feature and look at the threads on this topic that have been around for years at this point. Either way. The answer for shards is approx 30% over the course of an appropriate number of attempts.
  • TVF
    9388 posts Member
    Gravity makes things fly towards outer space. Who agrees with me?
  • For all the constant reassurances that this is just RNG, I'm not convinced. I've been tracking my Mission farm for the last week and the rate has been about 20-25% over some 150 sims -- 34 shards vs 50 expected. If that's a bad run of RNG, then RNG kernel is absurdly prone to bad streaks.
  • TVF
    9388 posts Member
    Hal_10000 wrote: »
    For all the constant reassurances that this is just RNG, I'm not convinced. I've been tracking my Mission farm for the last week and the rate has been about 20-25% over some 150 sims -- 34 shards vs 50 expected. If that's a bad run of RNG, then RNG kernel is absurdly prone to bad streaks.

    150 is not a terribly large sample size, especially when the 20-25% rate isn't that far off from the expected 30% rate.
  • Hal_10000 wrote: »
    For all the constant reassurances that this is just RNG, I'm not convinced. I've been tracking my Mission farm for the last week and the rate has been about 20-25% over some 150 sims -- 34 shards vs 50 expected. If that's a bad run of RNG, then RNG kernel is absurdly prone to bad streaks.

    Technically 30% of 150 is 45, . .
  • Yeah ... the science part of my brain says that. Doesn't feel like it, tho.
  • TVF
    9388 posts Member
    Hal_10000 wrote: »
    Doesn't feel like it, tho.

    This is the point - your feelings fool you. Numbers don't.
  • I have recorded 9017 attempts (and counting) across about 45 different Dark/Light/Cantina/Fleet nodes with 3065 total drops giving me a 34.0% overall drop rate. Having said that, I've had multiple streaks where I've had low drops. It all comes down to the probabilistic nature of RNG. For a truly random set of events, it's easy to calculate what the likelihood of having a bad streak is using binomial distribution functions. Getting 20-25% drops over 150 attempts is definitely on the low side, but there is still a 0.2% probability of that happening in any given sample that you come across. In other words, the odds of that happening are low (500 to 1), but not negligible when you are doing this many times.

    From my statistics, individual nodes have varied from 26.9% up to 43.8%, but the statistical probabilities of these outcomes for given sample sizes are only below 2% for a couple of outliers. At one point, I had an extremely bad run on Xanadu Blood (35 of 162 or 21.6% [odds of this happening are almost 1400 to 1 or 0.07%]), but it has since popped back up to 27.1% [odds of this are only 43 to 1 or 2.2%--still low, but much more probable than the earlier result].

    Last thing: RNGs are never truly random. They are generated by one of various computer algorithms, all of which have some bias. However, the typical RNG biases are so low, that you might only barely notice it in samples numbering in the trillions.
  • ^ this guy gets it. Make this a sticky in its own thread.
  • AntiFunn
    179 posts Member
    edited February 2
    TVF wrote: »
    It's not an average. It's an individual rate on each attempt. You haven't tracked anything, let alone a statistically relevant sample size.

    This.

    That means each turn it battles out as: 30% vs 70%. Most of the time the 70% will win definitely. Just tests how lucky you are in those attempts, just like casino.

    If you are unconvinced, try fighting against PA in Dota with maxed out Evasion lol you will feel the odds lol. Even at 50/50 its a double edged knife. just to give an idea no promotions whatsoever
  • Starkomat
    28 posts Member
    edited February 7
    Farming Mission Vao for 3 months now in cantina.
    Ratio: today 0 of 8. Yesterday 0 of 9. Days before maybe 2 of 9. Very short time of 3-4 days with better than 1 of 3. First month with good will 1of 5 or 6. count every day if above or under 1of3. For cantina and especially mission i can not find the “magical“ the math is right.
    But in 100 years there sure will come a winning strike with 1of1 for a whole year and ....in sum it will be still under the 33% ratio.
  • Liath
    3065 posts Member
    Starkomat wrote: »
    Farming Mission Vao for 3 months now in cantina.
    Ratio: today 0 of 8. Yesterday 0 of 9. Days before maybe 2 of 9. Very short time of 3-4 days with better than 1 of 3. First month with good will 1of 5 or 6. count every day if above or under 1of3. For cantina and especially mission i can not find the “magical“ the math is right.
    But in 100 years there sure will come a winning strike with 1of1 for a whole year and ....in sum it will be still under the 33% ratio.

    The fact that you are using the word maybe and only counting “above” or “below” is sufficient evidence that you are not properly counting drops in any meaningful way.
  • Everybody says around 30% is the average but nobody said that every player will be in this range.

    If in one week one player get one shard for every attempt and two other players get nothing the whole week it's still 33,3% in average, game just isn't fair to everyone.
    Never make the mistake of believing forbearance equates to acceptance, or that all positions are equally valid.
Sign In or Register to comment.