Announcement

Collapse
No announcement yet.

statistical

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • statistical

    I've had previous posts about collecting certain pieces and how many boxes and time it would take and it was pretty accurate but I also wanted to see one done in real time.
    I decided to test how long it would take to collect ninja cue pieces in ROME. I started exactly 1 month ago with a level 8 ninja at 0-80 pieces and wrote down each box opened.
    So far in a month period I've collected 16 out of 80 and opened the following boxes: 3h 85, 8h 23, 12h 11. So over a month with 16 out of 80 collected that means I have 64 left to go.
    The type of box doesn't really matter much but I just noted them down to see as comparison how rare the 8h and 12h are compared to 3h.

    Presuming the same ratio applies 64 divided by 16 is 4 so it would mean that I would have to play another 4 months give or take to max out the ninja cue.
    3x85+8x23+12x11=571 hours worth of opening boxes on a span of 30 days that equals 720 hours which means I had a downtime of 149 hours.
    I was pretty active but even so you can't always time every box to make sure you don't waste anything but that's probably a good average to use as comparison.

    Upgrading an expert cue goes like this in number of pieces 2-6-14-30-40-60-70-80. Which means that you need 302 pieces in total.
    If we take the 16 pieces per month as average that equals to 18.8 months to max the ninja cue which is a rough estimate but a fairly good one.
    Depending on how many boxes you open it can go faster or slower or if you use $$ to open boxes even faster.
    I also only played in ROME so in theory this could go faster if you play higher tables as boxes reward more cue pieces.

    Let's also keep in mind that the ninja cue is the most common expert cue there is but maxed out it will have pretty nice stats 10-9-10-7
    Collecting BERLIN pieces is a wholenother story but the good part of that is the cues already have high stats to begin with so upgrading them only a few times is enough.
    If you want to go for 10-10-10-10 expert cues that could be a worthy challenge if you have the patience and skill and time and dedication to grind them out

    I can't remember where I posted when I got my platinum cue to level 2 (I also got shark level 4) and both have same stats 8-9-9-8 which is more than enough for me.
    I could go over the numbers again based on the number of berlin wins as I quit playing there but maybe somebody can help me find my older post?

    found it, seems the ninja cue would take about 2.3k rome boxes (and i think it took about 1500 berlin boxes to get platinum to level 2 and shark to level 4)
    you can do the math individually depending how many boxes per day you open and what kind of cue pieces you are hunting
    i did not apply anything to other cues as i didn't really care much to do so but if any other collector wants to contribute it would be nice

    edit: found it https://miniclipforum.com/forum/mult...061#post363061
    Last edited by johndoe; 24-07-20, 02:39 AM.

  • #2
    johndoe WOW!!! I really enjoyed reading your post! That was quite an experiment and your data collection was superb.

    However, I still think there's more variables at play. Yes you determined your average of 16 pieces a month based off YOUR wins/losses of victory boxes in Rome. Yes you qualified that it would be different if you played another table like Berlin. It would also be different based on a player's skill level. So it's simply hard to pin down an exact number like 16 or 20 or 10 or 30 pieces a month. I wonder if I played Rome for a month, how many pieces of Ninja would I get? Of if you did this exact experiment for another month... I wonder how close to 16 would you be?

    Although it was a great exercise, I really believe that Miniclip leaves it all up in the air. That is, they use their algorithms for their so called "randomness", but is it really random?!?!?!? I say no. I believe their code uses algorithms based partly on random variables, but also factors in things like if you've spent any real money, what current level you're on, how many other cues you have, what other cues you have, etc etc. Maybe I'm way off. But if it's really just "up to the universe random" variability, I would hit much better "spaces" on the spinning wheel and scratch offs! So just the wheel and scratch offs tell me that your "chances" are not as "random" as if you were to play Black Jack at a Black Jack table in Vegas.

    All in all, I still LOVE that you did this experiment. Your results DO tell us that if you consider an average of 16 pieces a month, how much time it would take... 18.8 months to max. Mind blowing! Unfortunately, my average will probably be different than yours.

    Also you pointed out that with Expert cues at least with the Ninja, there's 8 levels and 302 total pieces. That may be true for Expert cues, but for Legendary cues, it's different. Yes Leg cues are in their own league because you can "purchase" leg boxes. Cues are different so with Ninja you had 8 levels. With most if not all of the leg cues there's only 5 levels with the level 5 being 20 pieces. I noticed this because I have my Plasma at Level 5 with 4/20 and one more space left in spin so it's currently (10-10-9-10).

    Again, loved your post!

    Comment


    • #3
      thanks d4ve_tv and i do want to clarify some things based on what you said.

      1- accuracy is determined by scaling a test over time. the more time you spend, the more data you get and the better the end result is
      so if i were to continue this test over 10 months, 100 months and so on, the more i would do it the more accurate my math will be

      2- obviously it's not random, each cue piece has a chance as i said above, that's why you can see that some cues drop a lot and some are super rare
      by generalizing this test and running it for a very long period of time i could even calculate the odds of getting any cue piece from any box

      3- wins and losses do not matter. only thing that is important is how many boxes you open per any given time period and that's how you can make your statistic
      wins matter in a sense that IF you only play for boxes then you can count each win as a box because you only play after opening a box in order to get another box
      or you can simply write down in a document each time you open a box just in case you play and you cannot be awarded another box because then the math based on wins alone would be wrong
      anybody could do a test per month or per whatever time they choose and see the results and in order to get a certain formula you have to account how many boxes you open and what downtime you have

      based on my test alone, there are small variations but you get a very good idea of an expected result and I will highlight the most important part again
      3x85+8x23+12x11=571 hours worth of opening boxes on a span of 30 days that equals 720 hours which means I had a downtime of 149 hours. (less downtime would be better as you get to open more boxes and thus collect cue pieces faster)
      another thing you can do to speed up the process is to open more boxes with $$ but this would add another variable to the equation but as long as you can keep track of these as well it's not an issue

      the bold region is the only math that matters and using this model, you can make a formula for any number of boxes over a month (or over whatever period you decide to do the test for and again - longer testing period=better result)
      Upgrading an expert cue goes like this in number of pieces 2-6-14-30-40-60-70-80. Which means that you need 302 pieces in total to go from zero to max level for the ninja cue.
      If we take the 16 pieces per month as average that equals to 18.8 months to max the ninja cue which is a rough estimate but a fairly good one.

      we can make another formula based on my case:
      time span: 30 days
      number of boxes opened: 119 (85x3, 23x8, 12x11) (will vary on a case by case basis)
      30 days converted to hours=720 (fixed number)
      number of boxes opened in hours=571 (will vary on a case by case basis)
      for an expert cue you need 302 pieces total (fixed number)

      you can use this formula for any expert cue that is available in rome and all you have to do is divide 302 to whatever number of pieces you get in this 30 day period
      just remember that the end result is strictly based on presuming that each month you respect the same numbers of boxes opened
      actually going even further, you can use this test for any other cue that you may collect as long as you keep track of the above variables

      i will post an update in another month with the same numbers and we can compare the result but it will vary because i will probably open a different amount of boxes
      that doesn't mean the first month will change anything because the formula for that specific month will still be the same and if i could continue at the same rate the overall result would still be 18.8 months

      i can also calculate the odds of getting 1 ninja cue piece like this, 16 cue pieces with 119 boxes opened means 16/119=0.1344
      so that means the chance of getting 1 ninja cue piece from rome box is 0.1344

      edit: i can also calculate the odds of getting a platinum cue piece by playing in berlin which is a lot more accurate as i opened 1700+ boxes
      i got 8 cue pieces with roughly 1700 boxes opened which means 8/1700=0.0047 chance of getting 1 platinum cue piece from opening a berlin box
      this process probably took around 15 months (with roughly same amount of downtime i had for the ninja cue but time is not important for this formula)
      of course you could get super lucky and get 8 pieces (or any other number) from a 12h or even an 8h box but the odds of that are very very small
      these are pretty accurate numbers but the real chance can be pinpointed down even more by extending the test to more boxes which i cannot do since i do not play berlin anymore
      but a player that has say 10k boxes opened in berlin can apply the same thing, check out a cue that is berlin exclusive and simply divide the number of pieces he has to the number of boxes opened

      Comment


      • #4
        Hey johndoe I'm not sure if I'm disagreeing with you or it's just because I can't quite wrap my head around your statistical mind I tried to follow your posts as carefully as I could but I still don't understand.

        You said in a month you opened:
        the following boxes: 3h 85, 8h 23, 12h 11. So that means you won 119 Rome games in a month right? Right there that's 3 different variables that make it "random" because you can't control whether you get a 3hr or 8hr or 12hr box. But let's just say for simplicity, we group all 3 box types into same category and just consider them win boxes. You didn't say how many total Rome games you played in a month to win 119 win boxes, but no matter. So we both do an experiment where we both play Rome until we both win 119 win boxes. It might take me 90 days (because I suck), but it only took you 30 days. Nevertheless, we both opened 119 Rome win boxes (all a mixture of 3h, 8h and 12hrs). Then out of those 119 boxes, you got 16 Ninja cue pieces, but say I got 32 pieces. How do you explain that? You can't reasonably *predict" how many Ninja pieces I would get out of those 119 boxes.

        Now I agree, if we repeated this test over and over and did multiple tests where we both opened 119 Rome boxes at a time and took the average of how many pieces we both got each time, then I may be persuaded. But with your one single test of just getting 16 pieces out of 119 boxes and then saying "If we take the 16 pieces per month as average that equals to 18.8 months to max the ninja cue which is a rough estimate but a fairly good one." I can argue that there are players who have played Rome for longer than 18.8 months and never got close to maxing out their Ninja cue and there are some that maxed it playing Rome nonstop in 6 months.

        Again, there's the 3 different hourly boxes that you also get "randomly" because you can't control which box type you get and some will argue that the 12hr boxes give you better cue pieces. I dunno.

        All I'm saying is that you can't reasonably predict how, when, which pieces you get. In other words, if someone asked you "Johndoe, how long does it take me to max the Ninja cue?" If you answer in 18.8 months. I would respectfully disagree. My answer would be, it depends on such things as how many Rome boxes you opened. The obvious answer is that the MORE Rome boxes you open, the more chances you have of getting Ninja cue pieces. 119 boxes a month x 18.8 is 2238. So you're saying if you win 2238 Rome games and opened 2238 Rome boxes, you will get the 302 Ninja cue pieces needed to max it? Well if you opened 2238 Rome boxes, I would hope that the Pool Gawds would be kind and give you the 302 pieces, but some will get the 302 pieces opening 1100 Rome boxes and some may have to open 4000 Rome boxes to get the 302 pieces. Yes I know this is all based on your 16 pieces a month average. All I'm saying is that if I did what you did, my average may be 10 or it may be 30.

        Am I missing something? Not trying to disagree with you John. Just sincere questions.

        Comment


        • #5
          well first off, nothing is random same with the odds of getting a 3h 8h or 12h box as you can see and you can also calculate the odds of getting those as well
          secondly, the reason i say that the more you test something over time or quantity the more accurate the result is and i will give you the simplest example for this:
          -flip a coin, you have 50-50 to get either heads or tails right? now if you flip it say 4 times and you get heads does that mean you have 100% chance of getting heads? ofc not
          but flip it hundreds or thousands and etc.... of times the odds will even out eventually the more you repeat it and they will gravitate towards the 50-50 ratio

          to continue on the matter, my test only shows the average time it would take to get 302 ninja cue pieces in rome with an average of 119 boxes in a month
          the number of 3h 8h and 12h will not vary that much for any given player because as i said above they are based on a chance which evens out in the end
          another thing to remember is that you only open a rome box at a time and you wait the timer out which can be 3h 8h or 12h and each time you can open a box you do so
          that means you don't have keep playing because you cannot open multiple boxes at a time only if you use real $$
          your question: how many rome games i played does not matter, because the test is done only by the number of boxes that i opened since as i said above you cannot open more than 1 box at a time

          the type of box you get evens out in the end as well, but one thing to remember is that 8h and 12h give out more cue pieces than a 3h but you are a lot less likely to get those
          so based on my case, if we say the odds of getting a 3h box is X, then the odds of getting an 8h box is roughly x/4 and getting a 12h box is x/8 (this is a rough estimate but no need to go into decimals and such)
          going by this mat you go like this X+X/4+X/8=100% 11x/8=100 which means X=72.72 so odds of getting a 3h box is 72.72, odds of getting 8h box is 18.18 and odds of getting 12h box is 9.09 and to confirm 72.72+18.18+9.09=99.99

          if somebody asked me how long does it take to max the ninja cue i could give them a very rough but reasonably accurate time if they tell me how many rome boxes they open on average per month
          mind you, this average should be more or less the same, obviously, since if you are not constantly opening the same amount give or take, the end result will vary a lot more
          my average of 18.8 is simply based on opening 119 boxes per month and only fits my math by my example but if they do open, on average, 119 boxes per month it is highly likely they will get it in 18.8 months

          you are correct in saying that it would take roughly 2238 rome boxes to max out the ninja cue (correct in a sense of based on my math and my test) but it's a very good ballpark
          the average margin will vary but it will not vary as much as half or double the boxes (since you roughly used that as comparison with the 1000 and 4000)

          again, the key to being very accurate is to extend the number of boxes to as high as you can and then you will very close to the real number
          the game developers could easily give out all the exact numbers they used or perhaps even a computer programmer with the ability to do reverse engineering
          they know the exact number for everything including:
          -odds of getting a 3h-8h-12h box (which i already calculated roughly) 3h=72.72, 8h=18.18% 12h=9.09%
          -odds of getting any and all cue pieces (i estimated ninja and platinum so far)
          chance of getting 1 ninja cue piece from a rome box is 0.1344
          chance of getting 1 platinum cue piece from a berlin box 0.0047

          i enjoy this conversation as I always had a math based mind when it comes to statistics so if you have any uncertainties or questions i will gladly keep answering

          Comment


          • #6
            Originally posted by d4ve_tv View Post
            Hey johndoe

            Now I agree, if we repeated this test over and over and did multiple tests where we both opened 119 Rome boxes at a time and took the average of how many pieces we both got each time, then I may be persuaded. But with your one single test of just getting 16 pieces out of 119 boxes and then saying "If we take the 16 pieces per month as average that equals to 18.8 months to max the ninja cue which is a rough estimate but a fairly good one." I can argue that there are players who have played Rome for longer than 18.8 months and never got close to maxing out their Ninja cue and there are some that maxed it playing Rome nonstop in 6 months.
            also to clarify this even more, if somebody played rome for longer than 18-19 months and haven't even gotten close, that means they opened a lot fewer boxes because this strictly depends on how many you open on average in a month
            if somebody managed to do it in 6 months they opened a lot more (i don't even think somebody can open that many boxes in 6 months even with zero downtime of opening boxes) so they would probably have to spend real $$ to open more at that point

            the best way to estimate this number for any person is know the average number of boxes needed which in our case by my example and my math over a month period is as you said 2238 (this is a rough number but it's a good one to refer to)
            so if we can agree that this number, give or take, is the number of boxes needed to max out the ninja cue by playing rome, then anybody can apply it to their specific case
            -somebody opens 10 rome boxes per month? 223.8 months to max
            -somebody opens 100 rome boxes per month? 22.38 months to max
            -somebody opens 200 rome boxes per month? 11.19 months to max
            so on an so forth, you get the idea, just divide our estimated number of 2238 boxes by how many boxes per month and you get your specific time

            Comment


            • #7
              Are you by chance a math professor?!?

              Originally posted by d4ve_tv View Post
              You said in a month you opened:
              the following boxes: 3h 85, 8h 23, 12h 11. So that means you won 119 Rome games in a month right? Right there that's 3 different variables that make it "random" because you can't control whether you get a 3hr or 8hr or 12hr box. But let's just say for simplicity, we group all 3 box types into same category and just consider them win boxes. You didn't say how many total Rome games you played in a month to win 119 win boxes, but no matter. So we both do an experiment where we both play Rome until we both win 119 win boxes. It might take me 90 days (because I suck), but it only took you 30 days. Nevertheless, we both opened 119 Rome win boxes (all a mixture of 3h, 8h and 12hrs). Then out of those 119 boxes, you got 16 Ninja cue pieces, but say I got 32 pieces. How do you explain that? You can't reasonably *predict" how many Ninja pieces I would get out of those 119 boxes.
              Again, the fact remains that "for you", you got 16 Ninja pieces after winning and opening 119 Rome boxes in 30 days. For me and my usual bad luck, I would probably only get 10 Ninja pieces if I opened 119 Rome boxes and it would take me probably 90 days to win 119 Rome games. So how can I use your average of 16 a month when my average is would be 3.3 pieces a month? Yes I'm exaggerating of course because my avg would prob be closer to 16, but there are players of many different skill levels. Yes by the time they get to point where they're playing Rome and trying to max out the Ninja cue, they should be an experienced player so their average may be closer to 16. OR they could be an exceptional player and have won way more Rome boxes in 30 days and their avg could be say 35 pieces.

              So I guess I am not as mathematically convinced as you. I believe there's too much randomness/luck and unpredictability to be able to comfortably predict (in this instance).

              Yes I totally agree that the more Rome boxes you open the more chances of getting Ninja pieces. That can be statistically proven. But there's too much variability to reasonably predict how long it would take a player. It would have to be based on that own player's skill level for ex. Sorry I didn't have time to study your complete response above. I glossed thru the main parts... I think.

              Comment


              • #8
                i stand by my numbers and i am continuing the test until i max out my ninja so i can be even more accurate
                now i know the following (and i will calculate this further for better results, as i said in time it evens out)
                average chance of getting the boxes 3h 8h 12h 3h=72.72, 8h=18.18% 12h=9.09%
                average chance of getting a ninja cue piece in a rome box 0.1344
                average chance of getting a platinum cue piece in a berlin box 0.0047

                math is not wrong and luck doesn't matter (as much as you think)
                the only variable that matters is the average number of boxes a player opens per month
                it's easy to understand as well and this is the last thing that makes the game interesting for me now
                i don't need coins and i got the cues i wanted so after i'm done with this test i will only play as the internet kids say, for the lols

                Comment

                Working...
                X