Alright, buckle up, cuz I’m about to spill the tea on my little “taylor swift at chiefs game today” experiment. It wasn’t some groundbreaking AI thing, more like me messing around and seeing what sticks.
First off, the idea just kinda popped into my head. Everyone’s been buzzing about Taylor and Travis, right? So, I was like, “I wonder if I can track when she’s actually at the games and what the internet’s saying about it.” I wasn’t trying to be a stalker or anything, just purely for kicks and data fun.
So, I started by scouring Twitter (or X, whatever it’s called now). I basically set up some searches for “Taylor Swift,” “Chiefs game,” “Travis Kelce,” and all the combinations I could think of. I wasn’t using some fancy API at first, just straight-up manual searching and saving links to tweets that seemed relevant. Real caveman stuff, I know.
Then, I figured I needed to get a bit more organized. I threw everything into a simple Google Sheet. One column for the tweet text, one for the timestamp, one for the user, and one for my own “relevance” score (basically, how much the tweet actually talked about her being at the game, not just general chatter). I know, super basic, but it worked.
Next step: images! I noticed a lot of people were posting pics. So, I started saving those too. Again, manually. I’d right-click, save image as, and then rename the file with a timestamp so I could keep track. It got tedious FAST.
I even tried to see if I could correlate the game dates with the tweets. I pulled the Chiefs’ schedule from their official website, dropped that into another tab on my Google Sheet, and then tried to match up the dates. It was a bit clunky, and honestly, I got bored of manually doing it after a couple of games.
After that, I did some searching on Google Trends to see if searches for Taylor and the Chiefs spiked on game days. It kinda did, but nothing super conclusive. I think everyone already expects that to happen, to be honest.
Here is the part I should have done better: I did not use ANY APIs at all. I was too lazy! If I were to do this again, I would use Twitter API, and some image recognition tools, but It was a short experiment, and I did not want to put too much time in it.
What did I learn? Well, first, people are OBSESSED. The amount of content about Taylor at these games is insane. Second, manually collecting data is a HUGE pain. I definitely need to learn to use APIs better if I want to do this kind of thing more efficiently. Third, it’s hard to extract meaningful insights without some serious data analysis tools. My Google Sheet skills only took me so far.
- Could I have made it more sophisticated? Absolutely.
- Did I learn a lot from just messing around? For sure.
Will I do it again? Maybe. If I can find a way to automate more of the process. Otherwise, I’ll stick to watching the games like a normal person. Haha.
Anyway, that’s the whole shebang! It was a fun little dive into the world of fan obsession and data collection. Not exactly Nobel Prize material, but hey, I had a good time.
