6* Champion Pulls - Weirdly skewed odds
Screaminghawk
Member Posts: 63 ★
My 6* Pulls are skewed predominantly towards cosmics; is this normal? How skewed are your pulls?
Total 6* pulls: 42
Cosmics : 18
Skill : 8
Tech : 6
Mutant : 5
Science : 4
Mystic : 1
Cosmics : Airwalker x 2, Annihilus, Cpt Marvel (old), Gamora, Heimdall x 2, Knull, Kamala, Nova x 2, Phoenix, Proxima, Red Goblin, Silver Surfer, Super Skrull, Thor OG, Venom the duck
Skill: Crossbones, DD HK, Elektra, Gwenpool, Hawkeye, Jabari, Shang-chi, Taskmaster
Tech: Doc Oc, Guardian, Punisher 2099, Red skull, Silver centurion, Vision (Age of Ultron)
Mutant: Colossus, Domino, Kitty, Prof X, Sunspot
Science: Abom OG, Joe Fixit, Red Guardian, She hulk
Mystic: Doctor strange
Total 6* pulls: 42
Cosmics : 18
Skill : 8
Tech : 6
Mutant : 5
Science : 4
Mystic : 1
Cosmics : Airwalker x 2, Annihilus, Cpt Marvel (old), Gamora, Heimdall x 2, Knull, Kamala, Nova x 2, Phoenix, Proxima, Red Goblin, Silver Surfer, Super Skrull, Thor OG, Venom the duck
Skill: Crossbones, DD HK, Elektra, Gwenpool, Hawkeye, Jabari, Shang-chi, Taskmaster
Tech: Doc Oc, Guardian, Punisher 2099, Red skull, Silver centurion, Vision (Age of Ultron)
Mutant: Colossus, Domino, Kitty, Prof X, Sunspot
Science: Abom OG, Joe Fixit, Red Guardian, She hulk
Mystic: Doctor strange
10
Comments
I am heavily skewed to Mystic and Cosmic.
Skill 3
Mutant 6
Tech 3
Cosmic 6
Mystic 0
Fairly even for me
[14, 6, 1, 3, 10, 8]
However, that assumes every class will show up equally often. That's not exactly true, because there are different amounts of champions for different classes. It would be impossible to replicate the precise statistics over time without knowing exactly which crystals and exactly when they were opened, but just using the current champion distribution (which includes trophy champs that don't exist in any normal crystal, but to simplify I'm ignoring that for now, as there aren't many of those champs) and retrying the same RNG experiment, the most unbalanced set of pulls I saw out of 100 trials was:
[15, 7, 5, 3, 8, 4]
How often does this happen in random trials (one class showing up at least 18 times out of 42)? About once every 1462. So unusual, but not astronomically rare.
Planning my T5 CCs for rank-ups is a challenge, my T5CC pulls do not follow the same pattern.
I have 49 champs from 58 pulls which is heavily skewed towards Science:
Science 21 (1 from dual class)
Cosmic 11
Skill 9
Mutant 7 (1 from EOP class nexus)
Tech. 5
Mystic. 4
For the first ~20, I have no mystic and 1 tech. So it only got better as the sample size increases.
Science 4 (1 duped)
Cosmic 4
Skill 4
Mutant 4
Tech. 4
Mystic. 2
Although with my 5 stars I did go for a very long time not pulling any techs at all, even when I open duals.
It was so bad I had to use an expiring tech awakening gem on og iron man
Tech: 3
Cosmic: 3
Mutant: 3
Science: 3
Skill: 1
First 6:
Cosmic, Mystic, Tech, Mutant, Skill and Science
Next 6:
Science, Mystic, Mutant, Tech, Skill, Mystic (dupe)
Next 2:
Mutant and Mystic
Need one Cosmic now for fairly even distribution
12 science
20 skill
17 cosmic
8 mystic
12 tech
19 mutants
I am really skewed towards mutants, skills and cosmics. And really lacking in mystics
Another 6 * crystal, another cosmic. Takes the tally of cosmics to 19 out of 43
I got to 54 6* crystals before my 1st skill champ landed and almost no dupes (besides King Groot and 3x Beast naturally at the time too). I was ecstatic when it finally happened, for a small moment, until I realized it was just Squirrel Girl. RNG stinks at times. My saving grace was a Nexus pick shortly after that got me MoleMan, then I got 4 other dud skill champs (including a Squirrel Girl dupe) in a row "naturally".
Now I have 106 6* (along with many, many dupes) and 18 are skill - more than my science (16), mutant (13) and not far below the other classes: cosmic (19), tech (21) and mystic (19). It took some time, but it averaged out.
Cosmic - 14, one dupe (Champion)
Science - 10
Skill - 9
Mutant - 8
Mystic - 6 (including UC), one dupe (Maw)
Tech - 3
Fair distribution across the middle, one somewhat high outlier, one really low outlier. My stats classes from college tell me this is probably pretty normal.
Science - 23
Skill - 21
Mutant - 17
Cosmic - 26
Tech - 24
Mystic - 20
The idea of reversion to the mean is often misunderstood. Some people think that the more ones you roll on a six sided die, the more likely it is to roll anything but a one on the next roll. But that's false: if the die isn't loaded and the roller not cheating, then the rolls are independent. If you roll five ones in a row, the odds of rolling one again are still one in six. The results won't "even out" over time. If they did, that would prove the die was non-random.
What does happen is the percentages will even out over time. Suppose, just to exaggerate a bit, that someone were to pull 20 cosmics in a row from 6* basic crystals. You would *not* expect them to pull far fewer cosmics over the next few hundred rolls. You would expect that that 20 cosmic imbalance would persist over time, because the crystals don't remember. What you would expect is the drops would be random from that point forward, as if those pulls didn't happen.
Over hundreds of pulls, and then thousands, that original weird anomaly of 20 cosmics would become less and less noticeable in the statistics, until finally it would be lost in the noise when we look back at the last ten thousand crystals and see each class having about 1700 drops, plus or minus. The long term averages eventually "drown out" any statistical blips. In fact, in a very real sense that's what statistics is: the long term averages drowning out the effects of every single individual roll.
This is a difficult idea for some people to wrap their minds around, but statistically speaking two things are true when it comes to this situation of opening random crystals that contain classes of items. One, the numeric imbalances between the different classes will *grow* over time. Eventually the most common class will be dozens higher than the rest, then hundreds, then thousands of drops more, growing without limit. However, the proportional imbalances between the different classes will *shrink* over time. The most common class will be 20% higher than the rest, then 10%, then 5%, then 1%, and over larger and larger amounts of drops that value will continue to drop asymptotically towards zero.
The imbalances get bigger, and they get smaller, depending on how you measure. That's a provable property of random sequences (cf: random walk).
Suck: 20
Don't suck: 1
Science - 17
Skill - 22
Cosmic - 22
Tech -17
Mutant - 17
I think it is balanced. Total 107 6* .
Next 6* crystal, another cosmic (carnage). Takes the tally of cosmics to 20 out of 44. I suppose I need to now open only dual 6* crystals going forward to break this cosmic run though these cost 25% more shards.
Total 6* pulls: 44
Cosmics : 20
Skill : 8
Tech : 6
Mutant : 5
Science : 4
Mystic : 1
Corvus
herc dupe
Knull
Thor
Angela
Venom
Anihilus
Proxima