Kiyakotari wrote: »

Mindcr0w wrote: »

Again, please provide said documentation.

www.crowncrates.com

In particular, check out the significant variation between the drop rates on the Radiant Apex mounts. Not how often people get a Radiant Apex - we all know that's stupidly low. But how the drop rates between the three Radiant Apex always have one (usually the cat-based, whether it's a senche or a senche-raht or a sabre cat) with a significantly lower drop rate than the other two Radiant Apex. Given that cat-based mounts tend toward being more popular than the other mounts, this is a fascinating discrepancy, no?

Yeah, except you've pointed out why it's not meaningful.

When you're collecting data you need a fairly large amount of the stuff to have accurate numbers. The larger the data sample, the more reliable the numbers will be overall.

So, I'm looking at the Nightfall Crowncrates, and, you know what? You're half right. The addon saw 103 Radiant drops, and of those, the breakdown was 41/35/27. Now, here's the problem with that. It's reasonable to take the data that we have and estimate that the Radiant Apex has a ~2% drop rate on the bonus card slot. There were 30k crates checked, and they turned out roughly 5k bonus cards. Now, 5000 cards is a pretty decent collection of data to use to estimate, and the margin of error should be somewhere around 0.6%.

So, if we look at the reported radiant drop at 1.95%, yeah, that's about right.

Now, you notice that 0.6%. What that means is that the real number could be off by up to 0.6%.

Now notice that the drop rate for each of the apex mounts is closer than the margin for error.

That's because there are so few of them.

Now, it's certainly possible that some of the radiant apex are rarer than others. The extreme example would be the flame crate's shadow senche, with only 6 appearing. Now, it's possible that senche really is rarer, but there were only 44 radiant apex mounts recorded for those crates total. The margin of error on that sample size that small would technically be over 100%.

So, it could be malicious. It could be that some of them are rarer than others. It could be that they have an equal chance of dropping, but there's so little available data that we can't confirm one way or the other.