I’m a runner based in Seoul, and I also work in the IT/data field.
After running for about 3 years, I started wondering…
Since Strava already has all my data, shouldn’t it be possible to understand what kind of runner I actually am, not just look at numbers?
So I began digging into my own data.
I broke the GPX into segments, combined pace/elevation/heart rate, looked for points where my pace dropped…
And since the analysis started getting complicated, I asked AI to interpret the patterns.
Surprisingly, it gave me some meaningful insights.
For example:
-
HR spikes right after an uphill → pace collapse shortly after
-
Starting too fast → inevitable crash in the second half
-
Wind/temperature affect performance more than expected
-
My average pace is okay, but my stability is terrible (lol)
It turned into a full side project, and now I’m analyzing my friends’ GPX files as well while refining the algorithm.
If you're curious, I put together a small Notion page where you can try the analysis yourself. Here’s the link!