"The" Rares List Guide - Do not ask about new UR's!

[Got That Jazz?]

The August PPS lion will never go for anything on the list. Everybody was enthralled when they saw its design, and due to this fact many people have gone rather overboard with overpaying huge list pets for it. Furthest I've seen someone go was with a Pink Patchy September...
Demand may factor in a few minor parts in trading, but rarity, date, and true value are the main factors here. If people want to overpay, so be it.
It's true it has confused people in the first place, as some people argue that because of demand, it is worth list pets. That is untrue. Furthest it would be worth is probably an 09 non list pet rare.

I've seen trades offering a BA, another one offering a dark shima, and the last one offered WWA, people will do anything to get a PPS lion, and it's driving me insane xD

[Greenleaf]

It's rather sad how demand factors over everything nowadays instead of the proper way around. Until demand dies down we can't do anything to stop it, but the PPS lion is not going to be added on the list in the near future. :3

[Got That Jazz?]

It's rather sad how demand factors over everything nowadays instead of the proper way around. Until demand dies down we can't do anything to stop it, but the PPS lion is not going to be added on the list in the near future. :3

Yes, yes it is. I remembered a CS will honorable traders, who traded on value, but that was back in '09, now people expect huge overpay for their pets...

[Nitro Indigo]

Because of demand, I don't think I'd get half of the pets on my wishlist. XD
In my trade rules, it says I'm willing to swap my other two November '09 staff pets for the equimagine outcome, one each. If I was asking for Shimas instead, people would (probably) be staring at their screen awkwardly.

[DaDwarf]

Guys, please take this discussion to pm's. It's not related to the rareslist. The lion pps (As said multiples times before) will never be added to the rareslist. Not in a long while - never. There are literally 100's of pets that should come before it.

Demand is a funny thing, but you don't have to play into it. All you have to do is be patient, i got my lion pps for one of them white rare wolf pps's. Same goes for any other pet, you don't have to overpay, just be patient. Let's drop the discussion and not spam the thread any further ;3

[jinx**]

Hey everybody, since the whole list discussion is always a bit tricks and sometimes the "value" of a pet seems to be depending on a small group of people, I thought about a better solution - A solution that is absolutely objective and not depending on anyones opinions.

Let's calculate the rarity lists automatically! I would like to see some algorithms running once per month automatically on the CS server, which analysis all trades of the last X months (perhaps 3/6/9 or 12 months). This algorithm will then calculate the actual mean value of a pet for all the trades, it has been participated in. Well, actually I know this would be some work, but it sounds more complex than it is + This way is the very only solution to get all real values for each and every pet updated without any manually work, which will take even longer (including all the discussions, etc.).

Even if there were some absolutely wrong-valued trades in the last X months, this won't be bad for the algorithm, since only a mean value will be calculated, which means: If you have five unfair trades (or perhaps gifts) out of 100 in the last X months, those won't count a lot, since the other 95 trades are still setting 95% of the value (and also some of the 95 trades could be unfair to the other side, which will remove the unfairness as well).

In the result the algorithm should calculate some sort of pet-value and outputs the 100 or 200 (or perhaps all?!) highest ranked pets and their value, which means that these pets will be the most wanted pets with the highest value!

Since CS is storing each and every trade this should be implementable pretty good on the current data set.. I have already done lot's of those data mining statistic algorithms for my work

What do you think about it?

[5404]

jinx**, do you know what is the problem? Rares List is made by users. But script and trade history and all that things belong to server. Not to users. So you have to include Nick to this project.
But script like that isn't as easy as it seems. (Especially when trades would be not n pets for one pet but normal, multiple pets trades.) I'm an amateur programmer and I know what I'm saying.
Also, script would count only trading value of pets. So low demand pets would not be on list, while for example August Lion PPS would be high (and when somebody would see that list - demand would rocket even more)...

[Hekomi]

Unfortunately I don't think the users who help to run the list are knowledgeable enough with algorithms, or have enough time to do something like that. Plus Nick and Tess do not ever intend to release the actual rarities of pets.

[jinx**]

No, that is not the case.

1. I'm a professional Software Engineer and IT-Security specialist so I know what I am talking about on the technical side.

2. The point you mentioned before is exactly what I meant! The algorithm is based on pure real trades, which means there is no user input - Just the input of ALL users. The PPS lion is a pretty good example. That is one of the pets, which gets the most discussion here since this pet has been released, BUT sometimes PPS lions will get traded away for a pretty good offer, most of these trades are just fair or at least demand-based fair. But that's what all trades are. No one will offer 4 sun jewels for a PPS lions, even in case of a very high overpay. The PPS lions might get a better place, then it has currently (no list pet), but won't be better then the most list pets, since not really many people trade list pets away for a PPS lion... Also some list pets are absolutely out-dated and the value is calculated way too high for it's current real value (best example here are the Grinners)

@Hekomi: There should not be any user interaction with these algorithms - Some programs or scripts (doesn't matter in this case), will calculate the new value tables absolutely stand-alone... So no user interaction and also no interaction by Tess or Nick.

The only thing that has to be done first is: The algorithms and scripts/programs needs to get designed, implemented and integrated... Should take a while, but then it's done and no more work should be done.

EDIT: @Hekomi: Would it be possible to have Nick included into this discussion? I already talked to him about some other technical topic (PS Vita user agent integration), but I cannot contact him directly, but perhaps you could initialize the conversation with him I could then talk about feasibility and stuff..

[Azuri]

this issue with using every list trade is you would have to take a cencus of EVERY cs user, every list trade from users that quit, every trade declined, and every trade accepted, to accurately place a pet

There are multipule privacy issues with this, you would have to use coppa accounts too, so immagine catiloging all the cs users who ever sent a trade, this violates internet privacy laws

Other issues are people who overpay and people who underpay, this can greatly skew the data, for example, i have had an uncommon offered for my hollyfox, that would need to be recorded, along with the green toxic offered for a raven, if we take a cencus a raven could be main list, when it actually is quite common for a list pet

And if i may this should answer the contacting Nick

Ugh, everybody just can't decide who is rarer! How about we go ask Nick for some statistics?

I PMed nick this message.

Hi Nick.

We where wondering if you or tess would mind helping us with this list, viewtopic.php?f=20&t=112366 as no one can deside on what rarest in order. So would you or tess have that kind of data that you would share


Regards aussie09

Yes, we have that information but no, we have no plans to share it. Sorry..

ok that my reply from nick so there goes that idea