Bronze! wrote:Obviously what this whole thread needs is more math theory... I don't think there's a way to replace the 2:1 rule with a new rolling rule, is the problem, and that just comes down to how rarity works. I mean, outside of URs and wonky demand, the basic rarity calculation is a function of two things: the relatively rarity of outcomes in a given litter, and the number of players who were around to adopt it, right? So the real calculation we're trying to figure out is the relative number of active players from each year. Obviously plenty of people active today have taken hiatuses at some point between when they joined and now, but for simplicity's sake, we're assuming everyone who joined in 2008 or 2009 adopted the 2009 pets, and the next year, 2008-2010 people adopted the 2010 pets, and in 2011 it was 2008-2011, etc. So the 2:1 ratio between years has historically assumed that every year, there are about twice as many of the new pets being adopted. Except we can pretty easily figure out that this isn't the case.
With every year that goes by, we have fewer and fewer extras from those earlier years, which is the thing the new rarity changes can't accurately account for (even if they are factoring in inactive accounts), and is the thing a lot of people here have already mentioned. The 2010 people picked up most of the 2009 extras for their collections, so there weren't as many left for the future members and thus everything gets rarer. But at this point, yeah, there's a negligible difference between the number of 2010 and 2009 rares left over and 2009/2010 members still active. And the pets from 2009 that are still rare and not VR were super common when I started. Like, I picked up most of them from free adoptions, there were so many. Nobody was making a big deal at that point about trading them for 2010s because we were talking what would now probably be in the omgsc-vc range. So I would still argue for a 2009 rare = a 2010 rare (probably + an uncommon, in practice, because old habits die hard) and then 2 '09 or '10 rares for an '09 VR, which would more or less maintain the values we had pre-update, really.
But also, a lot more people joined in 2011-2013. Like, a LOT more. Right away, about 50% more people joined in 2011 than in 2010. In terms of overall membership, for reference, we're currently at about member #1,090,000. Roughly 400,000 of those accounts were made from 2011-2013. Pets in that range still get rarer with time, sure, but I think it's worth factoring in these numbers if you want to build a rarity math system that can last. Since then, membership has fallen off but fluctuates a bit year to year, and I think the within 3 years rule starts making more sense.
So basically, my adjustment to the tiers suggestion above:
1 2009 rare = 2010 rare (or platinum, if you like)
[2:1 gap]
1 2011 rare = 1 2012 = 1 2013 (gold)
[2:1 gap]
1 2014 = 1 2015 = 1 2016 (silver)
[2:1 gap]
1 2017 = 1 2018 = 1 2019 (bronze)
[+1 gap - neater than 1.5 imo]
1 2020 = 1 2021 = 1 2022 (copper)
[+1 gap]
<current year here>
This at least gives new members the chance to trade up, but exactly how it converts to moving up to MA+ is a trickier question. This would give us 1 '09 VR = 2 '10 rares = 4 '11s = 8 '14s = 16 '17s = 17 '20s = 18 '23s, which, I dunno. I think we might be doing this part backwards. We can all agree that north of a thousand 2023 rares for 1 '09 is ridiculous, but 10 feels a little low. Maybe we should be asking people's minimum current year rare : '09 rare exchange rate, and then adjust the gaps between the new tiers accordingly.
I agree with this! Really well broken down to, and easier to understand without the 0.5 breaks