test content
What is the Arc Client?
Install Arc

Support Colonization Efforts

gfreeman98gfreeman98 Member Posts: 1,200 Arc User
Whine engaged:
What is it with the crit rate on these Support Colonization Efforts missions?

Using all purple DOffs each with at least 2 or 3 crit attributes, I get a 33/67/0/0 probability. And yet I most certainly do not crit an average of 1 out of 3 tries. :mad:

I didn't count the them, but in one nebula I must have ran the mission 7 or 8 times before I finally got my purple DOff.
screenshot_2015-03-01-resize4.png
Post edited by gfreeman98 on

Comments

  • captainmal3captainmal3 Member Posts: 436 Arc User
    edited April 2013
    I've noticed this too, I just assumed it was bad luck. I do get them eventually.
  • sheppardussheppardus Member Posts: 110 Arc User
    edited April 2013
    Please be aware that % chance is per mission not overall
    you have a 33% chance each time to get a crit outcome, this does not mean running this 3 times will guarantee a crit outcome. each time you repeat the mission, the result is a statistically seperate outcome
  • auriciusauricius Member Posts: 0 Arc User
    edited April 2013
    I guess it's totally random. I remember last May I got about 12 refugees in a row, from all the support missions spread across and I just gave up doing them after that - and lately I've had more success, I only get a refugee every 1 in 5 times and sometimes it's only a 25% to crit it. :eek:
  • anazondaanazonda Member Posts: 8,399 Arc User
    edited April 2013
    sheppardus wrote: »
    Please be aware that % chance is per mission not overall
    you have a 33% chance each time to get a crit outcome, this does not mean running this 3 times will guarantee a crit outcome. each time you repeat the mission, the result is a statistically seperate outcome

    This...

    The way you read the %, is that you should get 1 refugee for every 3 assignment you run... This is not the case.

    You run the mission, stats get reset, run the mission, stats get reset.

    It's a dice-roll... Not a count-down.
    Don't look silly... Don't call it the "Z-Store/Zen Store"...
    Let me put the rumors to rest: it's definitely still the C-Store (Cryptic Store) It just takes ZEN.
    Like Duty Officers? Support effords to gather ideas
  • fraghul2000fraghul2000 Member Posts: 1,590 Arc User
    edited April 2013
    sheppardus wrote: »
    Please be aware that % chance is per mission not overall
    you have a 33% chance each time to get a crit outcome, this does not mean running this 3 times will guarantee a crit outcome. each time you repeat the mission, the result is a statistically seperate outcome

    This pretty much sums it up. Even if you had 99/1/0/0 as your chances, you'd still be able to run 1.000 supports and not get a single crit.
  • johnnojhelmsjohnnojhelms Member Posts: 8 Arc User
    edited April 2013
    This pretty much sums it up. Even if you had 99/1/0/0 as your chances, you'd still be able to run 1.000 supports and not get a single crit.

    Sorry for the ever so slight necrorevival, but as this topic has been bugging me as well and I happened on this thread from searching for something else, I felt the need to reply.

    There's no such thing as 'separate outcome' in statistics. And 'per mission chance' is silly when talking about statistics as well. Statistics is about collecting, analysing and interpreting data.

    At 33% per mission, running three missions at one time, your statistical odds to get one crit out of three is 1/3 + 1/3 + 1/3, ie about 1. That's it. Running the same mission three times in a row, the odds of NOT critting on any of them is 0.67*0.67*0.67=0.30 So there's a 30% risk that you don't crit any of your three in a row.

    That's the whole point of statistics, calculating odds of series or groups of tests or whathave you. A singular event has odds, but it's just a singular event. When you start repeating that event and doing the math then you're into statistics.

    My personal experience is that of the OP, not only regarding the support colonization assignments, but regarding most of the assignments that I repeat on a regular basis.

    I managed to not crit a single time until my 19th Epohh research assignment. At 24% crit that's a 0.7% chance. Meaning that out of 1000 players only 7 would statistically succeed with such a feat.

    And since I see this everywhere I conclude that I'm not unlucky. I just know statistics and can tell that the system is not perfect. Either there's something wrong with the RNG, or the game uses some of your personal information (be it char name, account name, amount of letters in your name) to assign you a specific success rate.

    Ultimately I think it's by design and that we're given lower odds than displayed just to further the grind.
  • fletch246xxxfletch246xxx Member Posts: 40 Arc User
    edited April 2013
    I bet the RNG upper bound is set too high

    Should be 100 if they check for a result less than or equal to the percent value
  • gfreeman98gfreeman98 Member Posts: 1,200 Arc User
    edited April 2013
    ... since I see this everywhere I conclude that I'm not unlucky. I just know statistics and can tell that the system is not perfect. Either there's something wrong with the RNG...
    That's what I'm saying. The odds displayed are most certainly not borne out by empirical evidence. I just got another non-crit in Afehirr, and I must have ran it at least 9 or 10 times by now.

    And thank you for the reasoned response. Quite refreshing compared to the others attempting to "teach" me what statistics means.

    I suspect something is affecting the algorithm to skew lower. I'd really like to hear what the Devs say about this.
    screenshot_2015-03-01-resize4.png
  • fraghul2000fraghul2000 Member Posts: 1,590 Arc User
    edited April 2013
    I managed to not crit a single time until my 19th Epohh research assignment. At 24% crit that's a 0.7% chance. Meaning that out of 1000 players only 7 would statistically succeed with such a feat.

    If you take the amount of players or characters in STO, it would mean there are still going to be a couple of thousand where this does happen and those are going to be the ones that will march to the forums and complain about it.

    Unfortunately we don't get any numbers or posts from people who do get their normal distribution and neither from those who crit 7-8 times in a row.
  • johnnojhelmsjohnnojhelms Member Posts: 8 Arc User
    edited April 2013
    If you take the amount of players or characters in STO, it would mean there are still going to be a couple of thousand where this does happen and those are going to be the ones that will march to the forums and complain about it.

    Unfortunately we don't get any numbers or posts from people who do get their normal distribution and neither from those who crit 7-8 times in a row.

    You're missing the point entirely.

    First off, would you like to publically admit that you don't know anything about statistics, since you were in your previous post supporting a 'separate outcomes' version of statistics? I mean by all means continue to reply to the thread, but at least admit you don't know what you're talking about first. Trying to defend yourself by taking stuff out of context and trying to belittle it is just ridicilous.

    Second, if I can gather empirical evidence of a statistical system that shows I'm not getting the statistical results expected you can't just brush it aside with "you're just unlucky". The only thing you can do is question the amount of evidence and at which point it would be enough to satisfy your requirements.

    And by this I'm referring to what I said earlier, I see this in all I do often enough assignmentwise that I think there's something wrong with the system. You can't seriously say someone's unlucky in all of their assignments, while others are magically lucky. If I do something enough times, in a correctly working system, it WILL balance out to said statistical numbers. Period, there's no discussion, it's not about being lucky or unlucky, maths doesn't discriminate. It's a designed system, it works as it was designed. Be it intended or uninteneded. Statistics on the other hand isn't designed, it works, period.

    Rather than trying to undermine the original poster (or myself, or other posters) how about you just answer how many of a specific assignments, or assignments as a whole, do I need to do in order to give you empirical evidence that the system is not perfect?

    That's a rethorical question by the way, since you've already proved that you don't know statistics.

    If you're refusing to accept any proof given to you of a faulty system, then there's no use even discussing it. You *believe* that the system works flawlessly, you have no evidence that it does. I have evidence of the opposite, which you can refuse all you want, but maths doesn't lie. You can say 2+2=5 all you want because you believe it, doesn't mean it's true.

    So consider once again (rethorically as it may be) how much empirical evidence do you need before you accept it as evidence?
  • johnnojhelmsjohnnojhelms Member Posts: 8 Arc User
    edited April 2013
    This pretty much sums it up. Even if you had 99/1/0/0 as your chances, you'd still be able to run 1.000 supports and not get a single crit.

    Just for fun, show me the maths on this. What would the odds be, of not critting a single one of those?

    How many people would it take, doing this side by side, for one person to manage the feat?

    Calculate it and show me the maths. Put your maths where your mouth is, so to speak.
  • fraghul2000fraghul2000 Member Posts: 1,590 Arc User
    edited April 2013
    I'm sorry if it came across that way, but I wasn't trying to belittle one of your posts, but I really don't see how you'd think that my assertions contradict each other.

    Every support you run is statistically independent of any prior or following support. The result from one support will never effect the outcome of another. Not getting a crit on Day 1 will not increase your chances of getting a crit on Day 2, 3 or 4.

    This doesn't contradict your statement of the chances of getting 7, 50 or 100 failures in a row are very, very slim. Yet they do or at least they can happen, without the numbers provided by cryptic being wrong.

    If you take the OPs numbers the chances of him not getting a crit on those 7 assignments would be around 6%. The formula would be:
    p(A and B) ≡ p(A ∩ B) ≡ p(A ∧ B) = p(A) p(B).

    so in the example provided by the OP it would be (67/100 * 67/100 * [...] *67/100).

    I think my (somewhat exaggerated) numbers would amount to a chance of 1 x (10^(-1998)).
  • johnnojhelmsjohnnojhelms Member Posts: 8 Arc User
    edited April 2013
    I think my (somewhat exaggerated) numbers would amount to a chance of 1 x (10^(-1998)).

    You're basically saying anything is possible, while I'm saying it's statistically improbable and realistically impossible.

    I'm having consistant failures, disasters and non-crits from assignments that all amount to the fact that I can tell the system is not working statistically.

    As for your somewhat exaggerated numbers, consider the following. The 1000 tries are all done with one buttonpress in one second, and at each button sits a person clicking nonstop for their entire life (say 100 years for simplicity). Use everyone on earth, say 10 billion for simplicity. How many lifetimes back to back, would the entire population of earth have to click that button once a second, for one person to achieve a perfect failure of 1% a thousand times in a row?


    How long do I have to play for the game to 'balance' itself so that I'm not 'just unlucky' ? If I write down every single assignment I do, and show you that I'm not getting statistical results, how long would you keep saying "you're just unlucky, next month maybe you get all crits" ? If I write down my results, day after day, week month and year after year, at what point will you admit that there might be a fault with the coding?
  • markhawkmanmarkhawkman Member Posts: 35,236 Arc User
    edited April 2013
    gfreeman98 wrote: »
    That's what I'm saying. The odds displayed are most certainly not borne out by empirical evidence. I just got another non-crit in Afehirr, and I must have ran it at least 9 or 10 times by now.

    And thank you for the reasoned response. Quite refreshing compared to the others attempting to "teach" me what statistics means.

    I suspect something is affecting the algorithm to skew lower. I'd really like to hear what the Devs say about this.
    I disagree because your empirical evidence conflicts with mine.

    for example: CxP turnins. I've done over a dozen in the last week and my average (listed) crit percentage is around 25%. However I had a run where I got around 8 or so in a row that did not crit. However, if I had to guess(I haven't really recorded it) I would say that my average crit % is around 25%. Why because I have gotten quite a lot of crits, just not in that specific time period.
    -=-=-=-=-=-=-=-
    My character Tsin'xing
    Costume_marhawkman_Tsin%27xing_CC_Comic_Page_Blue_488916968.jpg
  • fraghul2000fraghul2000 Member Posts: 1,590 Arc User
    edited April 2013
    You're basically saying anything is possible, while I'm saying it's statistically improbable and realistically impossible.

    Which are both statements that are true and don't contradict each other. If you take the OP's numbers, it's statistically improbable that he'll fail all 7 of his attempts, yet it's still possible, since ~6% of all people will fail them.

    He's just one of the 6% that did fail them.

    Apart from him there are statistically a large number of people who have gotten their crit within the first 3 tries and never charged to the forums and told us about it. (33%/65%/70% respectively)

    Since we have to prove or disprove the numbers provided by Cryptic in the doff system, the larger the sample, the more accurate the numbers are going to get. I'm currently convinced that if we were to collect numbers from a certain assignment, from everyone running it (not just recording those that fail or crit a lot), with the same/similar doff-setup, we'd get results close to those that Cryptic provides.
  • gfreeman98gfreeman98 Member Posts: 1,200 Arc User
    edited April 2013
    If you take the OP's numbers, it's statistically improbable that he'll fail all 7 of his attempts, yet it's still possible, since ~6% of all people will fail them.
    OK first of all, I neither "marched" nor "charged" to the forums to rant. But I freely admitted I am here to whine about it though. ;)

    As for my numbers, I did not keep an accurate count of tries, but I absolutely will in the next nebula chain I start. (BTW it has been at least 9 or 10 times in Afehirr now and counting...)

    To clarify, I am not posting about one string of "bad luck" in one nebula. This is something I've noticed for the past couple of months, in 4 or 5 separate colony assignment chains.

    Do I expect to crit on the 1st, 2nd, 3rd or even the 4th try every time? Of course not. But when I start going 7, 8, 9 times or more in every chain before getting a crit then I have to question the veracity of these probabilities displayed.
    screenshot_2015-03-01-resize4.png
  • gfreeman98gfreeman98 Member Posts: 1,200 Arc User
    edited April 2013
    I disagree because your empirical evidence conflicts with mine.

    for example: CxP turnins...
    So because CxP "turnins" work for you I can't be seeing what I'm seeing with a completely different mission? Quite the leap that one has bearing on the other. I crit on lots of assignments across 3 characters on a regular basis. It's just these colony support missions that seem to be less than advertised.
    screenshot_2015-03-01-resize4.png
  • royalsovereignroyalsovereign Member Posts: 1,344 Arc User
    edited April 2013
    Let's assume that the game is using a single RNG algorithm to generate results for every event that is 'randomised'. How many thousands of different results is that per minute?

    Even if you perform a specific operation 100,000 times, the RNG has, in that same span of time, spit out /millions/ of results.

    There is no way you can expect to get any kind of consistent statistical results, no matter how many times you repeat an operation, in that kind of environment. The only way would be to literally stop the game and force it to only perform your operation. Otherwise, the data set is far too small.

    This doesn't mean that there is not a "this person will never crit, ever' flag and you've managed to get on the devs 'naughty list' somehow. But there's no way to prove that with the amount of data you can cobble together. As long as there are computer games, there will be 'lucky' players that get the good loot every other mission or so. And as long as you have them, there will also be the poor schmucks who get a rock in their halloween bag. Every. Stinkin'. Time.
    "You Iconians just hung a vacancy sign on your asses and my foot's looking for a room!"
    --Red Annorax
  • markhawkmanmarkhawkman Member Posts: 35,236 Arc User
    edited April 2013
    gfreeman98 wrote: »
    So because CxP "turnins" work for you I can't be seeing what I'm seeing with a completely different mission? Quite the leap that one has bearing on the other. I crit on lots of assignments across 3 characters on a regular basis. It's just these colony support missions that seem to be less than advertised.
    I didn't say that, I was simply pointing out that your experiences are not adequate evidence.

    I have several of the Support Doffs. Also, I don't grind them all that often...
    -=-=-=-=-=-=-=-
    My character Tsin'xing
    Costume_marhawkman_Tsin%27xing_CC_Comic_Page_Blue_488916968.jpg
Sign In or Register to comment.