![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
more stupid numbers (f1 rpf)
this is sort of a continuation of my sports/hockey rpf stats analysis that i wrote on a whim back in april. in a lot of ways f1 is more straightforward to examine for shipping than hockey is because there are only 20 drivers on the grid, which means that compared to the hundreds of active players in the nhl certain visualizations lend themselves as a given. and then of course there's the 2-driver setup for each team that naturally creates marketable dynamics. idk whether any of this will actually be interesting or revealing but let's go!
i'm lazy and don't think this will be as rambly as the last one so i apologize if many of the graphs feel arbitrary/underexplained... i'm just dicking around with my spreadsheets over here but it's fine!
0. NOTES
- â all data pertains to ao3 specifically
â all data pertains to the formula one rpf tag specifically
â timeline data is sectioned off yearly, but the "first year" actually covers all the years until 2011 to accommodate slow early growth on ao3. so it's 2004-2011, 2012, 2013, 2014, etc.
â same "date updated" vs "date posted" disclaimer applies as always
1. ALL-TIME STATS
as of june 2022, the f1 rpf ao3 tag has 15,971 works, 9,671 (60.5%) of which are public. the oldest fic crossposted to the archive is from june 2004, which means the tag spans a current total of 18 years.
before we get into timelines and the current grid, let's look at some general all-time lists. first we have the top 10 ships by work count:
then the top 10 additional tags:
these tags are pretty basic and not that different from most big fandoms, but one thing i've noticed in scrolling through the f1 tag is that it has a LOT of multiship a/b/o fic. like... enough that it feels kind of disproportionate. therefore i tested this theory by comparing f1 to the other top sports fandoms, and then ao3's top fandoms in general just to better contextualize the percentages (i feel like bts has to have the highest percentage of a/b/o fic in all fandoms with 1k+ or at least 10k+ works but i will not be the person to find out):
so hockey leads numerically but f1 easily has the highest ratio of a/b/o fic of all the big sports ficdoms. then back to the additional tags, i was also curious about the fluff/angst distributions of specific ships (note that individual fluff and angst tags are inclusive of "fluff and angst" and therefore have work overlap), so i took the top 10 ships and evaluated how "tragic" they are by this metric:
these numbers are about what i expected, but i think it's just interesting to see that the dichotomy actually has statistical backing? aka a best-friend ship like 2363 having a relatively high percentage of fluff fic compared to the ones with ex-teammates and/or rivals with well-documented tension, etc.
2. TIMELINES
i'll embed my old graph from the april sports ficdom analysis since i went through the work of getting numbers then... this shows the pretty steep jump in yearly production following the release of dts:
now here's a cumulative graph of f1 ficdom growth, with each x-value being the end of that year (2022 will probably end with upward of 18k works):
other than basic fic growth, i was also interested in seeing how the visibility percentage of the tag has shifted over time. iirc i ran this metric informally for hockey and found that the percentage was actually going down? but hockey doesn't have a mainstream netflix series driving a decent pocket of its modern fandom the way f1 does, and despite nhl teams having easily digestible c/a ship narratives i'd argue that f1 social media tries way harder at marketing and packaging their driver pairings, so i guess do things like that potentially shift fandom attitudes toward the 4th wall? i found that the percentage fell down with the ficdom's first ao3 growth spurt circa 2014, but the ratio of public fic is on the rise as of the past few years:
now moving on from graph timelines. i was also inspired to do a full timeline analysis of yearly top ships Ă my hockey ship timeline from the last post! sorry i had too much fun with this hahaha. just like before, i collected the tag's top 10 ships for each year, spanning 2011 to 2022 (this year is ongoing + once again i did 2004-2011 for the first one since the fandom was so small beforehand). then i visualized the yearly ranking changes for the top 15 ships with this table, determining total ship points by their ranking sums using the f1 points allocation system:
like with my hockey timeline, i took some liberties with the teams assigned to each driver and generally tried to reflect when the ship first became popular (also replaced toro rosso with at for consistency)... the mercedes red bull ferrari fandom chokehold is real
anyway f1 is really interesting because of how long some drivers' careers last, and that coupled with the grid being so small makes for very fun ship cycles, which are also present in hockey or any sport fandom really but perhaps i should say are more obvious in f1 because there are just sooo many more ship permutations possible in other leagues. for example just look at how many vettel ships dominate this timeline and how many different people have been popularly shipped with him over timeâthat's how you get the span of webber at the start of the timeline who is 11 years older and then leclerc at the end of the timeline who is 10 years younger.
more on this... since my data yielded 12 sets of 10 top ships, this meant i had a total set of 120 ships, with 43 of them being unique. seeing just how frequently shipped vettel isâe.g. while maxiel is the top ship in the archives, he has a higher number of unique popular shipsâi wanted to extrapolate that further for all drivers. here are the drivers with the most unique yearly top ships and the most total appearances in the yearly top ships, with their character fic count as contextual reference:
to be clear the "characters" metric isn't that interesting comparatively because obviously someone who has been around for a long time like vettel is more likely to be shipped with a higher number of unique drivers, and then character tagging also doesn't really mean much because while it measures shippability to a certain degree it also encompasses "now why is hoseok the bus driver" situations, etc. but i just thought it was interesting!
3. 2022 GRID
because the grid is fixed at 20 people you can look at "closed-loop" or complete stats like this which is pretty fun. first i pull all 20c2 ships for the drivers on this year's grid (didn't include hĂźlkenberg to make my life easier) and yielded this heat map:
and from that i made a simple table of each driver's cumulative work count to see who has the most ship fic with the other 2022 drivers, so it's like... Basically a measure of how shippable they are with the grid. but i'll go into that more later.
i was also curious about which team's current pairing is the most popular on ao3, so here are the 2022 teams sorted by driver pairing work count:
loooooooooooooool. i also did another team ranking/analysis, this time just adding both drivers' individual ship work counts (their "points" if you will) to see which team comes out on top. then i was like well i have to make the points share distribution now!!!
again the "fic count" is specifically the sum of shipfic with the 19 other drivers on the grid, which will have inevitable overlap with multiship tagging etc. but you get the general gist of it...
anyway, with the current grid i think i was also curious about like, how much does being from the same country (or speaking the same native language) and being teammates contribute to the popularity of your ships? i mean you have ships like 1016 who grew up karting together in france, and then classic marketed ex-teammates like 554, but you also have unlikely ~best friends with literally nothing in common on paper~ type of ships like 3118, etc... so basically i just wanted to see which was more prevalent atm. hence another table to evaluate each driver's current top pairing on the grid:
(unfortunately did kmag kind of dirty by excluding hĂźlkenberg because almost all his fic is with him sorry.) logically "past" teammates being a more common value makes sense since new teammates have not had as much time to be shipped together and have fic written for them. i was also going to make a reciprocity network graph to show all the losers mapping to lewis and then sewis mapping to each other but i got lazy hahaha. also i specified flag OR first language to shoehorn sharl into frenchisms (pardon) but that meant i was forced to conflate england and canada on the same basis đ
finally my last analysis based on the grid is uh... okay so i first made the second table below and called it SAx (Shippability Above Expected) as a joke, and the idea was basically to measure the rank difference between each driver's grid shipfic and their first season as an f1 driver, because i wanted to show that a rookie like guanyu probably won't have a lot of f1 fic, which means despite being last his rank remains unchanged, but then you have outliers like lando who are still fairly new to f1 but have a LOT of shipfic. however i realized that SAx isn't really accurate because it punishes older drivers who have other popular ships with inactive driversâtake alonso and his 300+ fics with webberâso it's really ultimately a test of longevity. and like relevance within the Very Current 2022 grid, which i think is still interesting because it kind of reflects that like, no one wants to ship him with the new guys but other "older" drivers like sewis don't show as much of a drastic loss LMFAO.
then in the same vein i came up with Writability Above Expected and created the table on the left, which just uses their character work counts instead. again the hoseok bus problem applies so it's more writability than shippability but i think it's still interesting to look at. and the results are truly very similar! lando and sharl kill me.
don't really know how to end this post but i hope this was even a little bit entertaining or interesting! the data was fun to extract but i can never tell whether any of the tables i post even make sense or intrigue people LOL i'm just like (regexing a billion ship counts) AO3 ANALYSIS IS MY PASSION........ i'll stop here.