Dupes

Mainloop25

Try claiming your reward at a different time of the day. 

shteev

Mainloop25 said:

Try claiming your reward at a different time of the day. 

In the last event when I opened a Consulate Crackdown, I thought to myself that I should do exactly that. But as usual, I forgot my plans today when RatC ended and I greedily opened my brand new Consulate Crackdown straight away.

Mind you, I've just checked, and I own all 18 of the AER rares now so it doesn't make a difference to my collection.

[edit] ...Is the game telling me to lose on purpose so I can get an uncommon instead?
http://forums.d3go.com/discussion/53876/

Ohboy

It would kinda be weird to be "smart" enough to make it "more random" by reseeding it with server time but deliberately removing the date from it...right?

Steeme

I'm quite certain there is more going on under the hood than people realize.

We keep asking for them to post their "drop rates".  If it were just some simple random algorithm, even with only basic logic, then it is to their advantage to actually post that and make it public.  It gives consumer confidence and trust and would likely drive more sales.  Some people would even gain confidence to start playing the law of large numbers and blow thousands at a time knowing full well that they can "expect" to proc a fair distribution of cards.

On the other hand, if they are using some "proprietary formula" behind the scenes to populate the "booster packs" and determine "rewards" then there is no way in tinykitty that they would even be allowed to publish that information.

Think about what this formula would be designed to do:

1. Control the flow of powerful cards into the economy

2. Ensure that it's nearly impossible to "complete" a collection so that there is always a demand for their products

3. Build artificial scarcity in order to manufacture "value" for certain cards and then assign a "price" in order to generate revenue

Once again, this would be insider information.  We will never hear it from them as they are not allowed to disclose the information.  We could only hear it "through rumour" since it would be unlawful to hear it otherwise.

Ohboy

//Disclaimer: The code listed below is a user-submitted program. Neither D3 Go! nor Hibernum officially support this program, and because of this we are not responsible nor liable for what may occur when running it. We advise all members to exercise caution. -Brigby

#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <stdbool.h>

#define LOOTABLE 58
#define TRIES 70


//Here we keep track of the number of times a unique rare has been generated
int rarearray[LOOTABLE];



//functions//
int RandomRare(){
    return (rand()%LOOTABLE);
}

void swap(int num1, int num2) {
   int temp = rarearray[num1];
   rarearray[num1] = rarearray[num2];
   rarearray[num2] = temp;
}

int partition(int left, int right, int pivot) {
   int leftPointer = left -1;
   int rightPointer = right;

   while(true) {
      while(rarearray[++leftPointer] < pivot) {
         //do nothing
      }

      while(rightPointer > 0 && rarearray[--rightPointer] > pivot) {
         //do nothing
      }

      if(leftPointer >= rightPointer) {
         break;
      } else {
         swap(leftPointer,rightPointer);
      }
   }

   swap(leftPointer,right);
   return leftPointer;
}

void quickSort(int left, int right) {
   if(right-left <= 0) {
      return;
   } else {
      int pivot = rarearray[right];
      int partitionPoint = partition(left, right, pivot);
      quickSort(left,partitionPoint-1);
      quickSort(partitionPoint+1,right);
   }
}

void FakeBellCurve(){
    int temparray[LOOTABLE];
    int count=0;
    int start=0,end=LOOTABLE-1;
    while(count<LOOTABLE){
        temparray[start]=rarearray[count];
        count++;
        start++;

        if(count>=LOOTABLE){
            break;
        }
        temparray[end]=rarearray[count];
        count++;
        end--;
    }
    count=0;
    while(count<LOOTABLE){
        rarearray[count]=temparray[count];
        count++;
    }
}

//end functions//



int main()
{
    //we'll seed it just once this time.
    srand(time(NULL));

    int count=0;

    while(count<TRIES){ // open TRIES packs
        //generate 70 rares
        int randomrare=RandomRare();
        rarearray[randomrare]++;
        count++;
    }


    //let's sort it for readability
    quickSort(0,LOOTABLE-1);

    //put it in a bell curve format
    FakeBellCurve();

    //print in bell curve format
    count=LOOTABLE-1;
    while(count>=0){
        int innercount=0;
        while(innercount<LOOTABLE){
            if(rarearray[innercount]>=count){
                printf("%d",rarearray[innercount]);
            }
            else{
                printf(" ");
            }
            innercount++;
        }
        printf("\n");
        count--;
    }
    /*count=0;
    while(count<LOOTABLE){
        printf("%d\n",rarearray[count]);
        count++;
    }*/
    system("pause");
    return 0;
}

I whipped up some code to generate random pack pulls(using raw rand() results) and display them in bell curve format. You can play around with it with your favourite compiler if you want. Default numbers are based on murtagan's post. I doubt anyone wants to run an exe file even if I posted it...right?

Volrak

shteev said:

I've just got my 10th Consulate Crackdown as my RatC reward, and I very much feel like something fishy is going on. Out of the other AER rares, I have 7 Release the Gremlins, and no more than 3 of anything else.

I tested how likely you are to have 10 of your most common AER rare pull while having 3 or less of everything from the third most popular rare pull onwards.

Algorithm:

Loop for each simulation run:
    Loop (rare pulls):
        Pull a uniformly random non-exclusive AER rare (one of 12 cards)
        If the third most popular pull has 4 or more copies then break out of loop
    End loop
    If the most popular pull has 10 or more copies then mark the run as shteev-compatible
End loop
Output the proportion of shteev-compatible runs to total runs

And the answer is about 0.33%, 1 in 300.  If the rare chosen for you is random each time, then this is quite unusual, but not astronomically strange: it's actually very likely to happen to at least somebody from the forum-going players.

To get enough data to move from anecdote to evidence is obviously the hard part.  If we could peek at many player's rare collections, and found results equivalent to shteev's happening significantly more than 0.33% of the time, it would demonstrate non-randomness.

Skiglass6

In advance, I am not a big statistics guy, but I think it is wrong to expect 1 system to flatten out.  Heads/Tails 50/50.  But if you flip 10 times (help me out statistics guys) you are more likely to get something other than 50% than you are exactly 50%.  Now, look at 1 system.  Say 1st ten flips are heads, it is a possibility.  Any future flips are not affected by the past and there is no code saying "hey we are off balance and need some tails here" and therefore we are at a 50% chance that we keep getting more heads.  Now yes, over time if the system does start correcting itself by being more 50% the 1st ten will start looking as a smaller deviation. But it also likely that it never does.  I think if the same code is used in multiple systems that it would flatten out the averages of the cards.

shteev

Volrak said:

shteev said:

I've just got my 10th Consulate Crackdown as my RatC reward, and I very much feel like something fishy is going on. Out of the other AER rares, I have 7 Release the Gremlins, and no more than 3 of anything else.

I tested how likely you are to have 10 of your most common AER rare pull while having 3 or less of everything from the third most popular rare pull onwards.

Algorithm:

Loop for each simulation run:
    Loop (rare pulls):
        Pull a uniformly random non-exclusive AER rare (one of 12 cards)
        If the third most popular pull has 4 or more copies then break out of loop
    End loop
    If the most popular pull has 10 or more copies then mark the run as shteev-compatible
End loop
Output the proportion of shteev-compatible runs to total runs

And the answer is about 0.33%, 1 in 300.  If the rare chosen for you is random each time, then this is quite unusual, but not astronomically strange: it's actually very likely to happen to at least somebody from the forum-going players.

Well, this is one reason I've decide to stick my neck out and predict my next few prizes. Science is not just about looking at an existing dataset, trying to spot a pattern in it, and drawing conclusions; the next step is to take those intuitions, form hypotheses, perform experiments, and observe the results.

Turns out I was paying more attention in science classes at school than statistics

wereotter

Possibly confirmation bias rearing its head here, but on a related note, Aethershere Harvester was the last holdout rare from Aether Revolt I had no copies of. I finally got a copy of it, and ever since, every event reward has been that same card.

wink

There do seem to be a lot of reports of people getting the same rare event rewards over and over and over again in a row. A lot of reports. Enough that confirmation bias doesn't seem like a sufficient explanation.

Ohboy

Maybe someone can go over the history to see if it's the same people complaining over and over again, or a huge chunk of the population too. I can't remember which one i it is.

FARTFACTORY

I can't understand why anybody would assume drop rates are random. It isn't stated anywhere in the game and the devs never chime in on the matter. Here's a thread where a player is speculating on whether or not gems falling onto the board is random. One of the devs pop on to clear up the confusion.
http://forums.d3go.com/discussion/53273/biased-ai-cascades-and-matchmaking

Hibernum_JC:
"There is no such thing. We do not use rubberbanding or anything of the sort. If cascades happen more often in certain circumstances, it's pure RNG. We do not, under any circumstance, trick the RNG so that any side in a match gets more cascades than the other.Like I said, all of this is confirmation bias - when something detrimental to you happens, you tend to notice it a lot more than when it is beneficial."

Whenever a thread about drop rates shows up around here, D3 stays away from it like the plague. There silence speaks volumes.

shteev

Well, look, as far as the drop rates being random is concerned, I think there's an easy solution. We just tag @Brigby , and ask him for a formal statement about the randomness of drops in boosters. If they don't make one, then I think we can safely assume that the drops are manipulated rather than random.

Meanwhile... the end of this week's NoP is coming up! Looks like I'll be getting 2 AER rares from it. Here's my collection just beforehand:

shteev

Release the Gremlins, and a Solemn Recruit. It makes me a little suspicious, but perhaps that's well within the bounds of reasonableness.

madwren

madwren said:

What's always odd to me is the streaks. I had a run wehre I got 4 Glint-Sleeve Siphoner in a row from events, followed by 3 Solemn Recruits, followed by 3 Greenbelt Rampagers and a pair of Quicksmith Spies. Today I got a Crackdown, I wonder if I'll get 2 more?

Nope. I did get two more Quicksmith Spies, though. Don't know how that Crackdown snuck in there. 

Houdin

Gee imagine that. Two more ballista as my reward for nodes personal and team. Wow good thing I'm the only statistical anomaly, I might really start thinking we are getting ripped off if this was happening to anyone else

shteev

Houdin said:

Gee imagine that. Two more ballista as my reward for nodes personal and team. Wow good thing I'm the only statistical anomaly, I might really start thinking we are getting ripped off if this was happening to anyone else

Now this is important: If we're going to do science here, you have to design your experiment and state your predictions in advance. By all means, do some experiments here in the future.

Your two Ballistas, declared after the fact, are too easy to write off as Confirmation Bias.

Chavez303

So I've had a theory about the drops here and let me know what you guys think.  

We're all operating under the assumption that a random number generator picks the cards.  Well, what if it's not so random.  We all have a UID.  My theory is that there is some algorithmic hocus pocus that all starts with your UID.  I've been chasing several rares and mythics for well over a year now and just can't seem to get them....however I have plenty of others that I seem to get quite often like the rest of you.  

If we all had our own "bell curve" based on our UID number, then that would make sense what shteev is getting Consulates and Houdin is getting Ballistas, and so forth and so forth.

Just a theory.  

madwren

Chavez303 said:

So I've had a theory about the drops here and let me know what you guys think.  

We're all operating under the assumption that a random number generator picks the cards.  Well, what if it's not so random.  We all have a UID.  My theory is that there is some algorithmic hocus pocus that all starts with your UID.  I've been chasing several rares and mythics for well over a year now and just can't seem to get them....however I have plenty of others that I seem to get quite often like the rest of you.  

If we all had our own "bell curve" based on our UID number, then that would make sense what shteev is getting Consulates and Houdin is getting Ballistas, and so forth and so forth.

Just a theory.  

Maybe it's the same one that randomly determines our opponents in events. =p

shteev

madwren said:

Maybe it's the same one that randomly determines our opponents in events. =p

I see your smiley face there, but you could be right. They always talk about 'fixing' the matchmaking algorithm, as if it was actually broken, and not merely an misguided attempt at acheiving something.

What else is broken in the game?