As the obsessed data nerd I am 🤓, I wanted to be able to reflect on my 2024 running journey so that I can better plan for 2025. This lets me answer questions like:
- How well did I train in 2024?
- Did I leave anything on the table? How do I improve?
- How can I actually improve?
If you use a fitness watch like an Apple Watch, then all of the data is recorded for us to use! Downloading your data from Apple Fitness is as easy as clicking a few buttons (and, ok, maybe an obligatory swear at the XML data format) – go to your Health app, click on your profile in the top right, and finally click on “Export all Health Data” in blue on the bottom. This exports a file in a criminal XML format, but from there we can process this data with Python and use Plotly to create all plots and interactive visualizations.
In this series, we’ll be demonstrating how to use this data to draw health and fitness insights – I will be using this to plan for a 3 hour Boston Marathon Qualifier race in 2025!
What can we do with this data?
The raw data unlocks a level of customization that Apple’s native tools just don’t touch. Think of it as your DIY toolkit for becoming a fitter, faster, and smarter athlete. Some cool experiments I’ve tried include spotting training plateaus with change-point detection, and optimizing routes to get a difficult hill workout in.
This data allows us to investigate loads of other areas as well including:
- 🦾 Biomechanics – we can understand our running form using measurements like stride length, ground contact time, and vertical oscillation.
- 🫀Heart rate analytics – is your heart thriving or barely surviving after that hill workout? Metrics like cardiac strength and VO2 max are example indicators well prepared your heart is for race day.
- 🏃🏼 Workout analytics – from splits to weekly mileage, this data will help you optimize your training plan—and maybe earn bragging rights on your Strava profile.
- 🌦️ Environment impact – how does your performance change with different weather conditions? For me, humidity is such a killer.
- 🛏️ Recovery – we know we need to rest to build muscle, but are we resting sufficiently between workouts?
Who cares?
I assume you’re at least partially a data nerd if you train with a tracker like an Apple Watch – while Apple’s rings are a great nudge to exercise, I think we can go so much deeper to use this data to understand what’s working and what’s not. For myself personally, I found my weekly training volume was far too low and slow for the types of races I want. If you’re the kind of person who loves tracking your progress—or just wonders why your training sometimes feels more like survival than progress—you’re in the right place.
In this series, we’ll use the fitness data export as a case study to cover topics like:
- Setting up your data pipeline and data models: How to turn that XML file into something you can actually analyze.
- Creating your first visualizations: Spot your strengths, weaknesses, and everything in between with Python and Plotly.
- Discovering patterns in your training: Learn what’s holding you back—and what’s pushing you forward.
- Building a personal dashboard for ongoing tracking: Keep all your key metrics in one place, updated automatically.
Whether you’re a data enthusiast or someone striving to train smarter, this series will combine technical insights with practical strategies to help you achieve your fitness goals. For more projects and insights, connect with me on LinkedIn and YouTube.
Home 
Leave a comment