Lightnews — Scholar-powered news

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 7

Today's plot shows inflation adjusted U.S. Halloween spending across 4 categories from 2005 to 2025 (2025 anticipated).

#Spooktober #DataViz #R #ggplot2 #RStats #Tidyverse #Retail #Halloween

October 8, 2025 at 12:41 AM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 6

Spirit Halloween stores seem to pop up almost everywhere out of nowhere. I wanted to see just how ubiquitous these temporary shops are across North America.

#Spooktober #DataViz #R #ggplot2 #RStats #Tidyverse #SpiritHalloween #Retail #Halloween

A map of North America showing the distribution of Spirit Halloween stores using purple circles sized by store count. The highest concentrations appear along the eastern United States, particularly around the Great Lakes, the Northeast, and the southeastern states. Dense clusters also appear around major U.S. cities such as Los Angeles, Dallas, and Chicago, while western Canada has smaller clusters near Vancouver, Calgary, and Edmonton. Northern Canada, the northern U.S. plains, and remote regions show few or no stores. The map highlights how Spirit Halloween stores are most common in densely populated urban and suburban areas.

October 6, 2025 at 10:10 PM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 5

Want to know which Halloween candy is most popular? Look no further. This Halloween candy constellation plot shows which candy traits are the most popular together.

#Spooktober #DataViz #R #ggplot2 #RStats #Tidyverse #HalloweenCandy #Candy #TrickOrTreat

October 5, 2025 at 8:27 PM

Cole 🐛

@colebaril.ca

I'm not entirely sure why several states (e.g., PNW, TX) only appear in around 2015 - it is unclear as to whether data collection only started for those states then, or they did not grow pumpkins to that point.

Line chart showing U.S. pumpkin production by state from 2000 to 2024. Illinois is highlighted in orange and consistently leads all other states, producing between ~150M and 320M kilograms annually, with fluctuations and notable spikes around 2014 and 2017. Other top states include California, Michigan, Ohio, and Pennsylvania andproduce substantially less, mostly between 20M and 100M kilograms. Decorative pumpkin icons appear near the chart title.

October 4, 2025 at 5:56 PM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 4

Today, I looked into U.S. pumpkin production. Illinois dominates in the U.S. due to its ideal soil type and favourable conditions for pumpkin growing.

#Spooktober #DataViz #RStats #Tidyverse #Pumpkins #Pumpkin

October 4, 2025 at 5:56 PM

Cole 🐛

@colebaril.ca

Almost non-stop lightning—in October.

#mbwx #mbstorm

October 4, 2025 at 5:04 AM

Cole 🐛

@colebaril.ca

Interestingly, when I graphed the number of haunted places per capita, the graph completely reversed: there are more haunted places per capita in the least populous states.

Bubble scatter plot of U.S. states showing state population (x-axis) versus haunted places per capita (y-axis). Bubble size represents the total number of haunted places. Smaller-population states like Wyoming, Vermont, South Dakota, and North Dakota have the highest per-capita haunted sites, while large states like California and Texas have many haunted places overall but low per-capita density. Title includes a cartoon haunted house with a ghost.

October 3, 2025 at 4:58 AM

Cole 🐛

@colebaril.ca

Sure enough, there is a clear pattern. The more people in a state, the more haunted places. Perhaps this is due to more people to report haunted places, or more entities to do the haunting...

Scatter plot of U.S. states showing the relationship between state population (x-axis) and the number of haunted places (y-axis). Larger-population states generally have more haunted locations. California and Texas stand out with the most hauntings, while Pennsylvania, Ohio, New York, and Michigan are also high. Smaller states like Kentucky, Indiana, and Massachusetts are highlighted as well. Title includes a cartoon pumpkin with bats.

October 3, 2025 at 4:58 AM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 3

Today, I looked at the number of haunted places in each U.S. state. CA and TX have the highest number of haunted places. Does this have something to do with population...?

#Spooktober #DataViz #RStats #Tidyverse #Hauntings #HauntedHouse

Choropleth map of the United States shaded by the number of haunted places. Darker purple indicates more haunted sites. California, Texas, and Pennsylvania show the darkest shades with the most hauntings, while many western and plains states are lighter with fewer. A ghost icon is placed over the northwest and a bat hangs upside down in the southeast corner.

October 3, 2025 at 4:58 AM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 2

Continuing off yesterday's theme of horror movies, for today's graphic I made a word cloud from >32K horror movie titles using the tidytext and wordcloud2 packages.

#Spooktober #DataViz #RStats #Tidyverse #HorrorMovies

A word cloud created from >32K horror movie titles using R’s tidytext and wordcloud2 packages. Words are arranged in an oval shape on a black background, sized by frequency and colored in red, orange, and yellow. The largest words include “dead,” “night,” “house,” “blood,” “ghost,” “death,” “vampire,” and “zombie,” with smaller related words like “killer,” “curse,” “terror,” “hell,” “monster,” and “massacre.”

October 2, 2025 at 3:57 AM

Cole 🐛

@colebaril.ca

And the data confirms the rationale behind dump months. Horror releases in January indeed have the lowest popularity.

A bar chart titled Horror Movies in January are Not Popular. The x-axis shows months (January–December), and the y-axis shows mean movie popularity. Bars are orange. January has the lowest average popularity, followed by December and November. Popularity peaks in July, with high values also in August and September. Decorative Halloween icons including a ghost, bat, and witch on a broom, are placed above the plot.

October 1, 2025 at 4:34 AM

Cole 🐛

@colebaril.ca

🎃👻 Spooktober: Data After Dark - October 1 🧵

Kicking off this spooky data series by looking at when horror movies are released.

Unsurprisingly, October dominates! From the 2000s onward, October became the clear peak for horror releases.

#Spooktober #DataViz #RStats #Tidyverse #HorrorMovies

A heatmap titled Calendar of Creeps: Horror Releases by Month. The x-axis shows months (January–December), and the y-axis shows years (1980–2022). Each cell’s color represents the number of horror movies released that month, with darker orange meaning more releases. The chart shows a clear peak in October, especially after 2000, and a second, surprising cluster of releases in January.

October 1, 2025 at 4:34 AM

Cole 🐛

@colebaril.ca

Starting October 1! I have been a busy beaver curating spooky datasets!

#RStats #ggplot2 #DataViz #R #Tidyverse

Illustration for ‘Spooktober: Data After Dark’ featuring cute cartoon ghosts, flying bats, carved pumpkins, and colorful Tidyverse package logos scattered across a purple background, creating a playful Halloween and data science theme.

September 26, 2025 at 4:02 AM

Cole 🐛

@colebaril.ca

Skipped ahead to next week's (2025-09-02) #tidytuesday because I love frogs.

#r4ds
#datavisualization

August 29, 2025 at 12:38 AM

Cole 🐛

@colebaril.ca

ID’d based on telling B shape of scaling on each abdominal tergite combined with leg scales.

August 23, 2025 at 11:31 PM

Cole 🐛

@colebaril.ca

I spotted an Aedes vexans mosquito in my bathroom waving her hind legs back and forth. They do this when threatened to evaluate their environment and air currents.

August 23, 2025 at 11:29 PM

Cole 🐛

@colebaril.ca

#TidyTuesday Week 32, 2025 - Extreme Weather Attribution Studies 📊

It's almost like there's a connection between extreme weather events and climate change...

August 21, 2025 at 2:43 AM

Cole 🐛

@colebaril.ca

I made a triple layer carrot cake for my grandma’s 82nd birthday!

#bakingsky #baking

July 27, 2025 at 11:19 PM

Cole 🐛

@colebaril.ca

July 27, 2025 at 12:34 AM

Cole 🐛

@colebaril.ca

Humboldt Penguin

July 25, 2025 at 5:12 AM

Cole 🐛

@colebaril.ca

Maybe

July 20, 2025 at 4:13 PM

Cole 🐛

@colebaril.ca

Preparing an article analyzing some data from my recent job search. The x axis is a custom application progression metric measuring how far I made it. Sometimes, a lot of effort in an application and process leads nowhere. Other times, low effort processes can be relatively fruitful.