**Mastery Loadouts**
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
Options
PHC drop rates... thanks Mvinceable
Bahamut
Posts: 2,307 ★★★★
https://m.youtube.com/watch?v=KtgL2CvyIkM Thank again Mvinceable for this information
6
Comments
It was actually shard crystals. The 200, 000 unit value thing is just like an equivalent value.
He didn't use any units
Another way of stating those odds would be 4* champion would be one out of 121 or 0.8%, 3* champion would be ten out of 121 or 8.3%, and 2* champion would be 110 out of 121 or 90.9%. That appears to be roughly within the margin for error for the numbers in the video.
[If my back of the Excel calculations are correct, there would be a 46% chance for a run of 2000 openings to fall more than 3 drops outside of the predicted ~16.5 4* champions if the odds were one out of 121, which is still within the reasonable range of values.]
Incidentally, since I decided to actually run the numbers, the odds measured with 2000 drops intuitively seems like it would be highly accurate, but it is only moderately accurate for 4* champions. For example, if we assume the actual 4* drop rate is exactly as measured - 12 in 2000 or 0.6%, then what are the odds of pulling *exactly* the expected amount of 12 in a 2000 crystal opening? 11.5%. There would be about a 40% chance of pulling more than 15 or less than 9 drops.
The math is hard, but the principle is easy to explain. It is not about how many crystals you open, it is about how many 4* champions you find verses how many you expect to find. 2000 looks like a big number. But 12 is not, and it is more obvious that the margin for error when you find 12 could be high. It doesn't take a lot of random variance (aka luck) to turn 12 into 16 or vice versa. It takes a lot more random variance to turn 120 into 160 or vice versa, even though the proportions are the same.
The rule of thumb is that the margin for error with this kind of statistical measurement is SQRT(N)/N or 1/SQRT(N) where N is the number of successful draws - NOT the number of pulls. In other words, this measurement opened 2000 crystals and pulled 12 4* champs. The rough margin for error is 1/SQRT(12) = 0.289 or 28.9%. That's the margin for error for the 4* number. The test also pulled 159 3* champions. The margin for error for the 3* measurement is 1/SQRT(159) = 0.079 = 7.9%. Notice the margin for error is higher for the 4* measurement than the 3* measurement.