Lea / she/they / lesbian / 28 / choir adult who loves tv / yellowjackets, game of thrones, the pitt, breaking bad, dan and phil, and much more / @lealthepheal on dnp twt
"Seeing the shift in Rhaenyra's face and whole body as she sees Alicent. This breaks something in her. From this scene onward, Rhaenyra changes in a way that she doesn't even really completely understand in that moment." – Clare Kilner
HOUSE OF THE DRAGON | Season 3, Episode 2, "Queen's Landing"
Phello Everyphannie! Check out the latest statistics dump from the phan photocards survey below! It's been out for 9 days now, so here are the promised numbers. I'm not done yet, I'll be continuing work for a long while still! You can see and contribute to my list of ongoing research questions at the end of this post.
Methodology, Sample size, and Disclaimers
I'm working with a sample of 345 responses, 1196 packs, and 5987 cards in this data! (See below for why that's not divisible by 5).
If someone claimed to open Y number of packs, but their listed cards did not add up to Y*5 and no explanation was provided as to why, I assumed there was a mistake and their response was discarded. 69 responses had to be discarded for this reason :( If it's any comfort, I didn't manage to enter my own pulls correctly on the first try either :P
Please note that I'm not a statistics professional, I'm operating on the level of an AP Statistics class I took like 8 years ago, Google searches, and nerd curiosity. I'm lowkey stealing valor from the true phannies in STEM, so kind corrections are welcome!
The below percentages are not necessarily exactly the odds that were used in production, it's just based on the data I have gathered. Everything here has a margin of error and uncertainty. You can help minimize that by submitting your card pulls to the survey!
Number of Packs
The average phannie responding to my survey opened 3.47 packs. Median 2, mode 1, max 30.
Phannies who bought more packs were more likely to be plugged into the phandom enough to find this survey, probably. Special thank you to everyone meticulously logging 50+ cards here-- that can't have been quick, but I really appreciate it!
Click below to open the rest of the Day 9 report!
Card Rarity
To start us off: the Platinum Pheal accounts for 0.25% of all cards, or 1.76% of all rares in my sample. I have the odds of any random pack containing a Pheal at 1.25%.
Dan and Phil claimed that the Platinum is 1% of all rares in the most recent Patreon livestream, and I do believe that's the real odds-- less than 1% off seems like a reasonable margin of error. It's also entirely possible that people with Pheals are more likely than Pheal-less losers like me to want to brag about their cards on a survey, which would introduce some sampling bias. Keep this in mind when looking at the rest of the results :)
Dan and Phil common cards are equally likely this time around, but slighty less common than Phan common cards.
Rare cards make up 14.2% of the cards in the sample, and commons are 85.8%.
I do believe it's possible to get more than 1 rare card in a single 5 card pack (have had a couple of reports of that happening), but it's an unlikely event. If you have footage of yourself opening such a pack and are willing to share, please let me know! I'm trying to confirm the existence of such packs and make sure that my survey respondents aren't pranking me. I got exactly 1 rare in my 5 packs and secretly did all of this just to complain about my lack of card riches, surely people with such great luck are joking.
Errors
Manufacturer errors seem to be on the side of Phannies! 8 people reported getting 6 cards in a pack, while only 1 person reported getting 4 cards in a pack. In total, less than 1% of packs seem to have a quantity error.
Even better, 7 of the 8 of people that got 6 cards in a pack report that the error pack contained a duplicate holo selfie! (i.e. two wedding selfie cards or two vacation 2025 selfie cards)
Individual Card Odds
Overall, your odds of getting any particular Dan/Phil/Phan common card are about the same as any other Dan/Phil/Phan card within its group. The variance here looks pretty normal, and I don't think it represents an actual difference in likelihood of you getting, say, a Catboy Dan vs a Bunny Dan.
The only one that stands out to me here is the one with Dan bare shoulderedly clutching some kind of fabric to his chest, which I nicknamed "it's giving princess Diana". Also the Precious Baby Angels Duo, maybe? That one seems more normal than the Princess Dan card, but still a little low relative to other Phan cards.
The fact that there's no similarly less-common Phil card makes me suspect this is my fault, somehow. Is it a methodology error? Formula mistake somewhere in my spreadsheet? Entirely possible. I've checked everything I can, but the princess card count remains stubbornly 2.3 standard deviations from the mean. It's further away than any other card and, to me, within the realm of "weird outlier". If anyone better at statistics wants to take a crack at this or provide context, let me know in the replies and I'll give you raw numbers to work with!
There are much fewer rares to work with, both in terms of unique cards and number reported, so I'm less concerned about discrepancies here. It seems likely that a Phil clear is as rare as Dan clear, and the same is true within each of the other rare groups. Smaller sample sizes on these, though, remember that.
Questions still to answer:
I started collecting data on which shop people ordered their cards from a few days ago. I really should have done so from the beginning! When I have a little more data to work with, I want to see if there are any differences based on where the cards came from.
How diminishing are the returns in buying more and more packs? How many packs would you need, on average, to complete a 55 card set?
I want to give out luck superlatives and highlight some interesting pulls!
There is so much bullshit I could be doing with pivot tables here. Just you wait.
Send me your questions! I'll look into them for you :)
Please continue to submit your photocard pulls as they arrive! I'm having loads of fun with this, and with your help we can get even more accurate numbers.
You can keep track of updates and my shitposts about the project by checking out the #photocard survey tag on my blog.
theartofmadeline
ellievsbear
turnnoffyourmind
origamiaround
romaincrisis
dailybridgerton
izzy806
