|
|
Post by Ringel on Mar 14, 2012 18:49:25 GMT -5
Despite the title, I don't think anything is "ultra" rare. But the statistics appear to show that the distribution of rares isn't based on equal probabilities either.
Raphael's post in my other thread makes me wonder if there is an unusual distribution of rares by color. If you have the chance post your rare count divided by color and Templar/Assassin:
Templar
Gold: 28
Black: 12
Red: 26
Purple: 27
Blue: 19
Green: 27
(Total: 139)
There are 42 distinct rares, 1/6 in each color except black (3/42) and Blue (11/42). The numbers are about right except blue...
Assassin
Gold: 15
Black: 12
Red: 17
Purple: 14
Blue: 6
Green: 17
(Total: 81)
Hmm...
Statistics geek out: Running the Chi-Square test (df = 5), the evidence for Templars being distributed differently than random is statistically significant (chi-square = 11.38, p = 0.044) but not for Assassin (chi-square = 6.01, p = 0.30) but this might be a reflection of the smaller sample size.
|
|
|
|
Post by Ringel on Mar 14, 2012 21:23:13 GMT -5
Raph's (now outdated) Assassin numbers from the other thread run at Chi-Square = 21.74, p < 0.0001.
That is pretty strong evidence against a truly random distribution of rares, if the numbers are correct.
|
|
|
|
Post by Mark D. Stroyer on Mar 14, 2012 21:51:48 GMT -5
Templar
Gold: 4
Black: 4
Red: 5
Purple: 6
Blue: 1
Green: 5
(Total: 25)
Assassin
Gold:
Black:
Red:
Purple: 1
Blue:
Green: 1
(Total: 2)
Obviously, pretty new player here, with only a couple mixed packs above the usual Templar boosters.
Still, it is a very bizarre note of just how sparse Blue rares are. It's Il Carnefice, by the way.
|
|
|
|
Post by Ringel on Mar 14, 2012 22:12:05 GMT -5
Oh yeah, I forgot about the automatic Carnefice.  |
|
|
|
Post by madstryfe on Mar 14, 2012 23:27:08 GMT -5
I have absolutely no clue what this thread is talking about lol. Definitely over my head of comprehension with regards to how many rares you get over time.
|
|
|
|
Post by Ringel on Mar 14, 2012 23:30:44 GMT -5
Just count your rares by color. Since there are more blue Templar rares than any other color, you would expect your collection to contain more blue Templar rares than any other color. Likewise for black Assassin rares. The evidence is starting to show otherwise.
|
|
|
|
Post by thedude808 on Mar 15, 2012 1:20:32 GMT -5
Templar
Gold: 29
Black: 13
Red: 21
Purple: 22
Blue: 19
Green: 26
(Total: 130)
Assassin
Gold: 26
Black: 10
Red: 18
Purple: 20
Blue: 8
Green: 18
(Total: 100)
|
|
|
|
Post by Raphael Majere on Mar 15, 2012 3:50:53 GMT -5
I have absolutely no clue what this thread is talking about lol. Definitely over my head of comprehension with regards to how many rares you get over time. Basically, some of us are thinking that that might be some rares that are 'rarer' than others. The strange thing: Juno. Many people are looking for her. with 0 copies, etc. I have fricking 14 copies of her. |
|
|
|
Post by Tuism on Mar 15, 2012 3:57:41 GMT -5
It's all pretty random and unbiased on what I've seen. Everyone thinks there's a bias, based on their own experience, but when we look at the entire picture of everyone's cards, it looks a lot more normalized. Like as in what we saw in the survey thread.
|
|
|
|
Post by Ringel on Mar 15, 2012 8:50:57 GMT -5
There is increasing evidence that it is not unbiased. Blue Templar rares are rarer than other Templar rares and Black Assasin rares are rarer than other Assassin rares. The statistics tests show that it very very unlikely to be fair.
|
|
|
|
Post by Pete on Mar 15, 2012 9:18:37 GMT -5
But we are only taking a sample from a sample
ie: a sample from forum goers, which is a sample from the full game.
I have a reasonable spread of rares, including one of every crime faction (at least) since it says I have all the black cards.
|
|
|
|
Post by Ringel on Mar 15, 2012 9:34:59 GMT -5
Statistics is the means of generalizing from a sample.
How about posting your rare counts?
|
|
|
|
Post by Pete on Mar 15, 2012 9:40:03 GMT -5
Statistics is the means of generalizing from a sample. How about posting your rare counts? If I have time I will. Kind of a mission. |
|
|
|
Post by Ringel on Mar 15, 2012 11:19:04 GMT -5
Yes, it takes time which is why this post took forever.
Let me be clear, I did not start the other "ultra rare" thread under the assumption that there are ultrarares. I was trying to show the opposite: How it is we can fool ourselves, because you need a large number of rares before you can reasonably expect to get all of them. My assumption is that they are all random.
Raphael's post surprised me though-- those number are really off, if everything is random. Consequently I took a look at my rare distribution, and to my surprise it looks like things might not be as random as I thought.
|
|
|
|
Post by Tuism on Mar 15, 2012 12:13:50 GMT -5
|
|
