Add y at the beginning of your ign and see what happens~~
It is time to change my ign to CloudY_Guwop_xxxxx
1 Like
6-36 (2 today, SE dupe)
40%-60% its a joke
Set of 40, still can’t reject null hypothesis.
Sample size too small.
Actually isnt it more likely that you would not roll a legend than that you would roll a legend?
Im garbage at math, but each hatch is an individual event wherein you are less likely to roll a legend than you are to roll a legend.
Each time you roll the odds are still against you.
Its not saying 40% of your hatches will be legends, just that each hatch has a 40% chance to be a legend.
So technically you could reasonable roll 35 straight SEs and there could still be a 40% chance of hatching a legend in any individual hatch event.
Right @Killerdog ?
Lol. Sample size 30 is enough to publish papers.
1 Like
It’s right for single roll but we are talking about past rolls. If you look only on the new roll that never happens, what is the purpose?
To compute rate of multiple rolls, it is another story.
Each instance is 40-60.
Just like every time you flip a coin is 50-50.
So you could reasonably and easily never roll a legend.
That is very true.
You can also compute how low the rate is if all past 10 tosses of your coins are tails, let alone past 70 tosses. If this is common for many players, it means the rate of coin toss is not 50/50. Similar to rare eggs.
Yeah, each roll is not conditional. Also, it’s possible to get any kind of result (e.g. 0 legendaries in 100 rolls) because of that. That’s pretty much what you said.
As for the sample size of 40… I swear sample size is the most irritating thing in all of statistics. I HATE statistics, it winds me up how overcomplicated some things are made by it. Experts can’t even decide on what a good sample size is. I tried to come up with a much better way of calculating a good sample size but I can’t find my notes on it (bit sad that I actually looked).
There are two big problems with using any small sample size for rare eggs:
- The 40%/60% is literally what it’s been programmed to give. It’s not just someone’s hypothesis which could be inaccurate in the first place.
- The variance is so high [0.24*n] (n = sample size). This is caused by both results being close to 50% and it means it’s harder to get a significant result as well as being unlucky seeming a lot more significant than it is (getting 30% legendaries is far more common than people might expect).
So basically you need a huge sample size to really prove anything. However, assuming that 6 in 40 eggs is the truth then it does technically reject the null hypothesis (0.5% chance of occurring). The thing is though, take the 15% drop rate of legendaries as a new null hypothesis and it will almost immediately be broken by anyone else’s results because it’s such a big outlier.
I think what makes this a topic in the community more than anything else is the high variance (speaking mathematically). It’s so easy to get a little bit unlucky and because of that you can keep getting a bit unlucky and get some staggering results. People expect their luck to average out but it’s harder to do that with high variance.
Thanks!
Thats pretty much what i was thinking lol but your version is mathier and therefore more authoritative.
Got gearcroc… Feel like rage quitting already lol
Gear is aweosme
1 Like
Now now, let’s not judge a monster by its looks alone, as a professional you need to test it before you pass on any judgements 
1 Like
True . I jumped into conclusions early cos it looked bad n tiny… But within 3 hours of using it, it’s in my ultra evolve list in first spot…
That knockback has given me lots n lots of wins in PvP 
It feels sooo good when they do all sorts of tricky stuff to charge up survivor n blood moves , then croc knocks them back 
Funny af
Rare gold egg
4 legends: 9 SE
Bane, Terror, Ultima and Bazilo
18/88. two se dupes.
16/78, 2 legend dupes, 3 new se’s (cryogolem, volcaiga, shinobi) and 4 se dupes
Now he has light lion, maggot, turtle, auro, atra, chroma, null, ankou, black penguin, nova and another 10+ normal legends in his one-month-old account.