Key takeaways

  • Weekly mileage adjusts based on recovery data from WHOOP or Oura. If your HRV tanks after a long run, the following week backs off.
  • Race pace predictions come from your actual tempo and long run data, not a calculator. The AI updates your predicted finish time weekly as fitness changes.
  • A 2015 study in the Journal of Sports Science and Medicine found that runners who followed periodized plans with individualized recovery had 23% fewer injuries than those on fixed plans.
  • Taper optimization uses your personal recovery patterns, not a generic 'cut volume 40% for 3 weeks' approach.

AI Marathon Coach

Train for your marathon with an AI coach that adjusts your mileage based on how your body is actually recovering, not a static 16-week spreadsheet.

Adaptive mileage planning

Weekly volume adjusts based on your training load, recovery scores from WHOOP or Oura, and how your body responded to recent long runs. No rigid spreadsheets. If you had a great recovery week, you might get an extra few miles. If recovery lagged, the plan backs off.

Race pace prediction

Using your GPS pace data, heart rate drift during long runs, and elevation profiles from Strava, the AI calculates a realistic goal pace. This updates weekly as your fitness changes rather than staying fixed from day one.

Taper optimization

The AI watches how your body responds to reduced volume by tracking HRV, resting heart rate, and performance in key workouts during the taper. It customizes taper length and intensity rather than using a one-size-fits-all approach.

Long run fueling feedback

By looking at your heart rate drift and pace decay during long runs, the AI can identify when you're likely under-fueling. If you fade consistently after 90 minutes, it'll suggest adjusting your in-run nutrition strategy.

Frequently asked questions

How does AI help with marathon training?

It takes your run data from Strava, recovery metrics from WHOOP or Oura, and body composition from Withings to build a plan tailored to your actual fitness level and goal time. It continuously adjusts weekly mileage, workout intensity, and rest days based on how your body responds.

What data does it use?

Pace, distance, cadence, heart rate zones, and elevation from your runs. HRV, resting heart rate, sleep quality, and strain from your wearable. Body weight from Withings if connected. All of this feeds into mileage decisions, pacing targets, and taper timing.

Can it help me qualify for Boston?

Yes. Set your BQ time as your goal and the AI reverse-engineers a training plan with the right mix of long runs, tempo work, and speed sessions. It tracks your projected finish time weekly and adjusts the plan if you're ahead or behind schedule.

How long until I see improvement?

Most runners notice improved pacing consistency and lower heart rate at the same pace within 4-6 weeks. The AI needs about 2-3 weeks of data to calibrate your plan. After that, recommendations get increasingly specific as it learns your training response.

Does it work with Strava?

Yes, Strava is the primary data source. Every run syncs automatically and the AI uses your GPS data, heart rate, splits, and elevation to evaluate each workout. It pulls your entire activity history for context.

Related

Start training smarter

Connect your apps, meet your AI coach. No credit card required.