DISCLAIMER! wrote: ? Most info in this thread are backed up by official site sources, such as links to posts from mods and admins. However, some times we don't have official answers to questions. In this case, I've used my best estimates to figure out what's going on, based on what we do know. Like this box, these speculative sections are marked with a red question mark at the start!
MATHS! wrote:+- We can let loose some math to better understand how the pound works. The main text will take you through everything you need to know, but these boxes marked with a green +- contain more detail for people who are interested. You can skip these if you want.
MATHS! wrote:+- ? The reason that the distribution of pets in the pounds is likely to be similar to the "global" CS distribution is due to the law of large numbers, so long as there are no differences between how pound pets are selected and how CS pets are selected.
Specifically, the law of large numbers states that the average result obtained from a large number of independent random samples will converge to the true value, if it exists.
This means that if you repeat a process with a 20% (20 in 100, or 1 in 5) chance of one outcome and an 80% (80 in 100, or 4 in 5) chance of another outcome 100,000 times, you'd expect to see the first outcome about 10,000 times and the second outcome about 80,000 times.
You can check this by grabbing a six-sided dice and rolling it a bunch of times. Record the total number of rolls you make, and the number of sixes you roll. You have a 1/6 chance of rolling a six, so even though the dice roll itself is random, after many rolls (more than 100!) you should find that the number of sixes you roll divided by the number of rolls gets pretty close to 1/6.
Knowing this law, we can see that if an outcome makes up x% of the pets on CS, and the pound has a large number of pets, and we assume that there isn't a bias to the pets in the pound, the outcome will probably make up x% of the pets in the pound. You can think about this as the outcome being a face on a very, very large die, determined by how rare it is in the global CS distribution. The pound just rolls that die to a bunch to see what pets it has, so a pet should have the same rarity in the pound as on the global CS distribution.
There are several factors that complicate this. One is that global rarities are calculated as a function of active players (CITE), whereas the pound includes pets from inactive users. Additionally, part of the pound's stock comes from pets confiscated from cheaters. Both of these factors might over-represent the number of old or rare pets in the pound when compared to the global distribution. However, active players donating to the pound are likely to donate only common and new pets, which might over-represent these pets. We have no data on how or whether these factors balance; my instinct is that the combination of the law of large numbers and the counterbalancing nature of these forces will leave us with a distribution close enough to the global.
Happy holidays! wrote:ALWAYS put your Dec. 18 pets in a locked group immediately. NEVER trade them until you know what they're worth.
Did I Get A List Pet? • Horror's List • Pets With Confusing Values • Successful VR, ER, OMGSR Trade Thread • How Much Would You Value This Pet in C$? • Trading Advice Thread
Users browsing this forum: ACat, Google [Bot], himarry 124, Himiko., Leavay, Nicnova, oOAkaniOo and 42 guests