#with data from the full 2023 season... for the everlasting p/entry correlation

[39oa]

mini summer break update... new entries to the f1 rpf centrality graph (PIA, SAR) + oscar has had by far the largest relative increase in ship fic since i first pulled data back in april 👨‍🍳

f1 rpf graphing & archive insights

intro & prior work

hello! if you're reading this, you may already be familiar with my previous post about graphing hockey rpf ships and visualizing some overarching archive insights (feel free to check it out if you aren't, or alternatively just stick around for this intro). i've been meaning to make an f1 version of that post for a while, especially since i've already done a decent amount of f1 rpf analysis in the past (i have a very rough post i wrote a year ago that can be read here, though fair warning that it really does not make any sense; while i've redone a few viz from it for this post i just figured i'd link it solely because there are other things i didn't bother to recalculate!)

f1 is quite different from many team sports because a large part of my process for hockey was discovering which ships exist in the first place—when there are thousands and thousands of players who have encountered one another at different phases of their careers, it's interesting to see how people are connected and it's what was personally interesting to me about making my hockey graphs. however, with f1's relative pursuit of "exclusivity," barriers to feeder success and a slower-to-change, restrictive grid of 20 drivers, it becomes generally expected that everyone has already interacted with one another in some fashion, or at least exists at most 2 degrees of separation from another driver. because of this, i was less interested in "what relationships between a large set of characters exist?" (as per my hockey post) and more so in "what do the relationships between a small set of characters look like?"

process

my methodology for collecting "ship fic" tries to answer the question: what does shippability really look like on ao3? (the following explanation is adapted from my hockey post:) a perceived limitation i have with character tagging numbers on ao3 is that they don’t exactly reflect holistic ship fic; that is, if lando is tagged as a character in a max/daniel fic, it gets attributed to his character tag but doesn’t actually say anything about how many Relationship Fics exist for him on a whole. my best solution for this was essentially uncovering most of a driver's relationships and summing their individual fic counts to create an approximate # of “relationship fics” for each player. so any kind of shippability graph going forward will use that metric.

i used ao3’s relationship tag search and filtered by canonical in the formula 1 rpf fandom and only pulled relationship* fics (“/” instead of “&”) with a min. of 5 works. ao3’s counts are… Not the most accurate, so my filtering may have fudged some things around or missed a few pairings on the cusp, which again is why all the visuals here are not meant to show everything in the most exact manner but function more so as a “general overview” of ficdom. although i did doublecheck the ship counts so the numbers themselves are accurate as of time of collection.

(*i excluded wag ships, reader ships, threesomes to make my life easier—although i know this affects numbers for certain drivers, team principal/trainer/engineer ships, and any otherwise non-driver ship. i left in a few ships with f2, fe, etc. drivers given that that one character was/is an f1 driver, but non-f1 drivers were obviously excluded from any viz about f1 driver details specifically. this filtering affected some big ships like felipe massa/rob smedley, ot3 combinations of twitch quartet and so on, which i recognize may lower the… accuracy? reliability??? of certain graphs, but i guess the real way to think of the "shippability metric" is as pertaining solely to ship fic with other drivers. although doing more analysis with engineers and principals later down the line could be cool)

also note that since i grouped and summed all fics for every single ship a driver has, and since one fic can be tagged as multiple ships, there will be inevitable overlap/inflation that also lessens the accuracy of the overall number. however, because there's no easy way to discern the presence and overlap of multiship fic for every single driver and every single ship they have, and attempting to do so for a stupid tumblr post would make this an even larger waste of time… just take everything here with a grain of salt!

data for archive overview viz was collected haphazardly over the past few days because i may have procrastinated finishing this post haha. but all ship data for section 2 was specifically collected april 22, 2023.

PART I. f1 rpf archive overview

before i get to ship graphing, here are a few overviews of f1 ficdom growth and where it measures relative to other sports fandoms, since i find the recent american marketability of f1 and its online fandom quite interesting.

first off, here's a graph that shows the cumulative growth of the top 8 sports rpf fandoms from 2011 until now (2023 is obviously incomplete since we're only in may). i've annotated it with some other details, but we can see that f1 experienced major growth after 2019, which is when the first episode of dts was released.

something that fascinated me when making this graph was the recent resurgence of men's football rpf in 2023; while the fandom has remained fairly consistent over the years, i had noticed that its yearly output was on the decline in my old post, and i was especially surprised to see it eclipse even f1 for 2023. turns out that a large driver behind these numbers is its c-fandom, and it reminded me that out of all the sports rpf fandoms, hockey rpf is fairly unpopular amongst chinese sports fans! i wanted to delve into this a little more and look at yearly output trends for the top sports fandoms since 2018, only this time filtered to exclusively english works (a poor approximation for "western" fandom, i know, but a majority of sports fandom on tumblr does create content in english).

another thing i've long been curious about with f1 specifically is—because of how accessible dts and f1 driver marketing are to fans online, does f1 rpf and shipping culture skew a bit more "public" than other fandoms? i'd initially graphed the ratio of public fic on ao3 for hockey because i also wanted to see whether it was on the rise (again, apologies for how many callbacks and references there are in this post to hockey rpf... it's just easy for me to contextualize two familiar sports ficdoms together *__*), but i was surprised to see that it's actually been steadily trending downward for many years now. f1 fic, on the other hand, has steadily been becoming more public since 2016.

another note is that c-ficdom follows different fic-posting etiquette on ao3, and thus chinese-heavy sports rpf fandoms (think table tennis and speed skating) will feature a majority public fic—here's another old graph. since f1 fandom has a relatively larger representation of chinese writers than hockey does, its public ratio falls a little bit if you filter to english-only works, but as of 2023 it remains significantly higher than hockey's!

anyway, onto the actual ship graphing.

my ship collection process yielded 164 ships with 57 drivers, 46 of which have been in f1. all 20 current active f1 drivers have at least one ship with min. 5 fics, though not all of them had a ship that connected them to the 2023 grid. specifically, nyck de vries' only ship at time of collection was with stoffel vandoorne at 56 works.

once again because f1 is so strongly connected, i initially struggled a lot with how i wanted to graph all the ships i'd aggregated—visualizing all of them was just a mess of a million different overlapping edges, not the sprawling tree that branched out more smoothly from players like in hockey. this made me wonder whether it even made sense to graph anything at all... and tbh the jury is still out on whether these are interesting, but regardless here's a visualization of how the current grid is connected (color-coded by team)! i graphed a circular layout and then a "grid-like" layout just for variety lol.

of course, i still wanted to explore how ships with ex-f1 drivers have branched out and show where they connect to drivers on the current grid, especially because not too long ago seb was very much the center of the ficdom ecosystem, and the (based purely on the numbers) segue to today's max/charles split didn't really come to fruition until the dts days. so here's a network of f1 ships with a minimum of 75 works on ao3:

before i go into ship breakdowns, i also have a quick overview of the most "shippable" drivers, aka the drivers with the highest sum of fic from all their respective ships. the second bar chart is color-coded by the count of their unique ships to encapsulate who is more prone to being multi-shipped.

PART II. ship insights

first let's take a look at the most popular f1 ships on ao3, again filtered to driver-only ships.

here's another graph filtered to the current grid only, and then one that shows the 15 ships where one driver isn't and has never been an f1 driver:

for this section, i ended up combining my ship data with a big f1 driver dataset that gave me information on each driver's birth year, points, wins, seasons in f1, nationality... etc., so that's what i'll be using in the rest of the post. disclaimer that i did have to tweak a few things and the data doesn't reflect the most recent races, so please note there might be some slight discrepancies in my visualizations.

anyway—in my hockey post i did a lot of set analysis because i was interested in figuring out what made the players who were part of the ship network different from the general population. with f1, since almost Every Driver has at least one ship and it's a much more representative group, doing a lot of set distributions wasn't that interesting and so i stuck more to pure ship analysis. still, the set isn't completely representative, which i noted by checking the ratios of driver nationalities in my dataset and then in the large database of f1 drivers i merged with (though filtered to debut year >= 2000 to maintain i guess the same "dimensions").

while british and german drivers have been the most common nationalities in f1 since 2000, both in general and in my ship data, it seems that ficdom slightly overrepresents/overships them and then underrepresents brazilian drivers. i was also curious to see the distribution of ships by nationality combination (which is actually quite diverse), and though it once again wasn't surprising that uk/germany was the most common combination given that we've just established the commonality of their driver groups, i found it somewhat interesting to realize just how many ships fall under this umbrella.

i then once again wanted to see what the distribution of age differences looked across ships. the ships i graphed yielded a range of 25 years, with the oldest age difference being 25 years between piastri and webber. tbh, something that's interesting to me about f1 ships is not just how connected current drivers are but also how there is a very strong aspect of cyclicality, wherein long careers in combination with well-established celebrity culture and post-retirement pivots to punditry & mentorship position drivers perfectly to still be easily shipped with any variety of upcoming drivers, hence why we encounter a relatively significant variety of age differences.

of the ships with two f1 drivers, 38% were within 2 years of each other, while 44% had an age difference of 5 years or more.

more experimentally (basically i wanted to use these performance metrics for something!), i tried graphing driver metrics against "shippability" to see whether i could uncover any trends, normalizing to percentile to make it more visually comprehensible.

one thing that was interesting to me is that there is a strong correlation between a driver's points per entry and their number of ship fic; really, this isn't surprising at all because it's basically a reflection of whether they've driven for a big 3 team, and we know that the most popular drivers are from big 3 teams, but then i guess it does become a bit of a chicken and egg question... which is something i'm continuously fascinated by when discussing success and talent in sports fandom, especially in a sport like f1 where there is so little parity and thus "points" do not always quantifiably translate to "talent," making it difficult to gauge why and when a driver's skill becomes consciously appealing to an audience. i don't know but here's that scatterplot.

similarly, i also wanted to look at years active vs. fic to gauge which drivers have a High Number Of Ship Fic relative to how long they've actually been in f1, basically a rough rework of the "shippability above expected" metric i'd tried exploring in my old f1 post haha. because the set i merged with attributed 1 "year active" to a driver just like, filling in as reserve for a single race, and it also included drivers who maybe raced one season and then never raced again, but then i still wanted to include current rookies in their first season to show where their Potential lies... i settled on filtering to drivers who were or have been active for at least 5 seasons OR who debuted recently and thus have a bit of rookie leeway. there's a decent amount of correlation here, which is again... in f1, the underlying argument for remaining active for many years is that you have to be good enough to keep your seat, so it's expected that if drivers stay on the grid for a long time they will eventually accrue more fandom interest and thus ship fic. still, we can see some drivers who underperform a little relative to their establishedness—bot and per, interestingly also below the trend line in the points/entry graph–and then those who overperform a decent amount, like nor and lec.

this is somewhat interesting to me because i'd tried to make a similar scatterplot with my hockey set and found that there was... basically nooo correlation at all, but i also had to make do with draft year and not gp which i think might move the needle a little bit. regardless, it's just interesting to think about these things in the context of league/grid exclusivity and then other further nuances like the possibilities of making your niche in, for example, the nhl as a 4th line grinder or f1 as a de facto but reliable #2 driver for years down the stretch, and then how all of that impacts or shapes your fandom stock and shippability.

moving on, here's a look at the current top 20 f1 ships and how much of their fic is tagged as fluff or angst! out of all their fic, kimi/seb have the highest fluff ratio at 38.44%, while lewis/nico hold the throne for angst at 34.74%.

lewis/nico are also the most "holistically" tragic ship when you subtract their fluff and angst percentages (by a large margin as well), while jenson/seb are the fluffiest with a difference of 17.38%. really makes you think.

and finally this is a dumb iteration from my old f1 post but i thought this was kind of funny haha so: basically what if teammate point share h2h but the points are their shippability on ao3.

closing thoughts

that's really all i have! again, i don't know whether any of these graphs make sense or are interesting to anyone, but i had fun trying to adapt some of my hockey methodology to f1 and also revisiting the old f1 graphs i'd made last year and getting to recalculate/design them. i know there's a lot more i could have done in examining drivers' old teams since many ships are based on drivers being ex-teammates and not the current grid matchups, but it would have been too much of a headache to figure out so... this is the best i've got. thanks for reading :)