Okay, so I was messing around with some sports data, and the idea popped into my head: “What if I compared Serena Williams and, like, a top tennis player from a different era?” I landed on Guglielmo Vicario, not because they’re directly comparable (they aren’t!), but because it was an interesting data exercise. More of a “how can I use this to learn?” thing than a serious sports analysis.

Getting Started
First, I hunted down data. This was surprisingly tricky. Finding comprehensive, easily accessible historical tennis data isn’t as simple as you’d think. I ended up scraping some stuff from various websites, which was a bit of a pain. It wasn’t pretty, lots of manual copy-pasting and cleaning up messy tables.
The Data Mess
- Serena’s Data: Easier to find, thankfully. Lots of resources track her stats, Grand Slam wins, match records, etc. Still, it was scattered across different sites.
- Vicario’s data: Way harder. Since He’s name is not very famouse. I relied more on historical records, old articles, and whatever I could find. It felt like digging through an old attic!
Wrangling the Numbers
I decided to focus on just basic.
I pulled the data into a spreadsheet.
I spent a good chunk of time just cleaning things up. Standardizing dates, making sure names were consistent (you wouldn’t believe how many variations there are!), and dealing with missing data points.
I calculated some simple win percentages and stuff, just to have something to compare.

The (Slightly Silly) Comparison
Obviously, directly comparing a modern player to someone much earlier isn’t really fair. The game has changed! But, it was a fun way to see how the data looked side-by-side. I played around with some basic visualizations, just to see the numbers in a different way.
What I Learned
More than anything, this was a lesson in data wrangling. I spent way more time cleaning and organizing data than actually analyzing it. It also showed me how much sports data is out there, but how scattered and inconsistent it can be.
I won’t pretend this was some groundbreaking sports analysis. But it was a fun little project that helped me brush up on my data skills. And hey, I learned a bit more about tennis history along the way!