Will Mackey

uh oh @downround.bsky.social

oh no i just rewatched it because of this. 10/10 cannot recommend.

This triangle helps explain a titanic shift in how Australia votes This triangle is going to help us explain how Australian politics has fundamentally changed over the past five decades.

So much to love in the scrollytelling — the ternary plot, the depth of the data, the findings — but my favourite part is the explanation and the text in general. 📊

Remember, a chart doesn't have to be understood in 3 seconds. It should reward the effort you put into understanding it.

joey why do you hate geom_step

Hey #rstats! There's space in my courses this Thurs & Fri in Brisbane!

Quarto for Scientists
- The art & science of reproducibility, "getting unstuck in R"
- statsoc.org.au/event-6132307

R Package Essentials:
- Function fundamentals, write a package from scratch!
- statsoc.org.au/event-6132303

these are stunning

Grattan Institute is hiring an Associate! Are you ready to help solve some of Australia’s most pressing policy issues? 🧠 💼Join a think tank that is making a difference. For more details and to apply now, visit EthicalJobs [https://buff.ly/QoEUSaX].

Trying something new:
A 🧵 on a topic I find many students struggle with: "why do their 📊 look more professional than my 📊?"

It's *lots* of tiny decisions that aren't the defaults in many libraries, so let's break down 1 simple graph by @jburnmurdoch.bsky.social

🔗 www.ft.com/content/73a1...

in R it is day one. but python day zero. hope this helps

this is such a beautiful story (with beautiful paintings!) by @joshnicholas.com. makes me love Melbourne all over again

that’s right

oh did i say billion? i meant gazillions of rows

(doing grouped window functions on 8 billion rows of data with 16gb RAM from within an #rstats script is wild)

pretty sure i’m becoming a @duckdb.org in R guy

really like this

Hi #rstats friends!

I'm starting consulting! I've had a great journey in academia working with brilliant people, and while it's bittersweet (and a bit scary!), I'm excited for this next step. I'll still be maintaining {greta} and am hoping to be even more involved with the R community.

1/n

The guide to gradients in R and ggplot2 – James Goldie Visualising data and building things for fun and for change. #climatehealth → #datajournalism

I’ve been using the new gradients and patterns in #rstats and #ggplot2 quite a bit for recent package, so here’s a deep dive! Everything from the basics of using gradients in your plots through to sophisticated visual effects.

jamesgoldie.dev/writing/stac...

this is really cool

It is not a rigorous benchmark, but I was curious on how #duckdb duckplyr compares to other data wrangling packages in #Rstats. Turns out, it is pretty pretty fast 🚀.

blog post: blog.schochastics.net/posts/2025-0...

Toad-in-the-hole? ⛳🐸🏌️‍♂️

Cane toads in Queensland love their local golf course. ⛳

Some golfers take this as a serendipitous chance to drive 🏌️‍♂️ down their numbers.

Thanks to the {galah} package team for easy access observation data! @rowdynerd.bsky.social @daxkellie.bsky.social. #rstats

that’s coming in 2-4 months. he’ll get his revenge

love parenting

louse up? what the shit is that

you’re Keys’ good luck charm at this point

👀

I think it's good to be critical of AI hype but on here I often get the response that LLMs are completely useless. For better or worse that isn't really true. In a very limited set of applications (e.g. writing simple bash & python scripts to order) they're actually very good and getting better 1/9

incredible

all turned around for Swiatek just as i posted that

👀