Roast My Fitness Tracker — What Your Strava and Apple Watch Say About You

What Fitness Trackers Actually Know About You

The data these apps collect is embarrassingly complete. Your Apple Watch knows you closed your rings twice in the first week of January and then quietly stopped trying. Strava knows your longest run of the year was in March and nothing has come close since. Fitbit has a graph of your sleep quality that looks like a heart monitor with something wrong.

The metrics that hit hardest are the ones that seem motivational until they're read back to you by something with no feelings:

• →Step count vs. goal. You set 10,000 steps a day. Your average is 4,200. You've been 58% of the way to your own goal for six months without adjusting either the goal or the behaviour.
• →Sleep patterns. The app calls it “restless.” The AI calls it what it is. You went to bed at 1:47am on a Tuesday and woke up at 6:15am and wondered why you felt awful.
• →Workout consistency. Or the lack of it. The three-week streak you had in October was impressive. Its abrupt end on November 4th has never been explained.
• →The January → March cliff. It's in everyone's data. Activity peaks in the first two weeks of January and then drops off like a graph of someone who had a new year's resolution.

How to Get the Roast

You don't need to export a CSV or connect any accounts. Just screenshot whatever dashboard makes you feel the worst:

• →Apple Health: The weekly summary screen in the Health app — the one with steps, active energy, and the ring completion percentage — is perfect. Scroll down to include sleep if you're feeling brave.
• →Strava: Your activity log, especially if it shows months with zero activity. The “This Week” summary screen works well. Bonus points if your last activity was titled something aspirational like “Back at it 💪.”
• →Fitbit / Garmin: The weekly report screen. Sleep score, steps, active minutes. All of it is readable by the AI and none of it is flattering.

Then upload to RoastCard — the Strava roast type is on the main roast page.

What the AI Roasts

The AI is particularly good at spotting the gap between intention and reality in fitness data. It notices things like:

• →You hit your move goal on two days last week. Both were days you had plans in the evening.
• →Your resting heart rate is 82bpm. Not dangerous. Just slightly elevated in a way that suggests you need to either exercise more or relax more. Probably both.
• →The three-week workout streak ended the same week you got a new TV subscription. The AI may or may not mention this. It depends on how much is visible in the screenshot.

Example Roasts

Here's what the output looks like:

🤖 Fitness · Savage · Apple Health weekly summary

“Your 10,000-step goal has been achieved exactly twice this week. Both on days where you had a commute. Left to your own devices — working from home, no obligations — you managed 2,340 steps on Tuesday, which is roughly the distance between your desk and the kitchen. Your resting heart rate is 79bpm. Not a red flag. Just an amber one, gently waving.”

🤖 Fitness · Nuclear · Strava activity log

“January: 8 runs. February: 4 runs. March: 1 run, titled ‘Getting back into it 🔥’. April through October: nothing. November: 2 runs. The second one is titled ‘Back at it for real this time.’ You have said this before. Strava remembers everything. Strava has seen your arc and it is not an upward one.”

If you want a dedicated Strava experience, we also have a dedicated Strava roast page at /roast/strava — it's built specifically for Strava screenshots and pulls out the details that make Strava roasts particularly cutting: segment times, kudos counts, and the trajectory of your training over time.