RunPal

RunPal is a lightweight, single-user running training plan generator that prioritises runner self-assessment over aggressive optimisation.

It generates a multi-week plan based on a target race date and length, and adapts individual runs in response to how the runner feels day-to-day — without automatically rewriting the entire plan.

The goal is simple: listen to the runner, don’t overcorrect.

---

Key ideas

• Plans are generated once and stay stable unless the user explicitly regenerates or resets
• Wellness is treated as perceived readiness for today's run, not injury or illness forecasting
• Adjustments are local and reversible, not “smart” global optimisations
• No accounts, no backend, no data collection

---

Features

• Dynamic plan generation

  • Choose race distance, race date, and runs per week (1–5)
  • Generates up to 10 weeks based on time to race, counting backwards from race date
  • Weekly runs selected using recovery / speed / endurance buckets

• Local persistence (no login)

  • Generated plans persist across page reloads using localStorage
  • The same plan is restored exactly as generated
  • Plan only changes when the user clicks Generate or Reset

• Wellness-based run adjustments

  • Each run has a daily wellness selector
  • When a run is marked “Not great”, it adjusts immediately:
    • Speed runs - downgraded to a recovery run (−20% distance)
    • Endurance runs - distance reduced by 20%
    • Recovery runs - distance reduced by 10%
  • Adjustments are visual, reversible, and scoped to that run only

• Immediate feedback

  • Adjusted runs are marked with an “Adjusted” badge instantly
  • No refresh required
  • Wellness state persists across reloads but never leaks between plans

---

What this project demonstrates

• Thoughtful state modelling (plan vs wellness)
• Deterministic persistence of randomly generated data
• Rule-based behaviour instead of opaque “smart” logic
• Incremental feature development with clean Git history
• Conscious product constraints to avoid over-engineering

---

Tech stack

• Vanilla JavaScript
• HTML & CSS
• localStorage for persistence
• No frameworks, no backend

---

How to run locally

1. Clone the repository
2. Open index.html in your browser
   (or use a simple local server such as VS Code Live Server)

That’s it — no build step required.

---

Roadmap

• Progressive long-run distance across weeks
• Minor UI copy and accessibility polish
• Optional README diagrams explaining adjustment rules

Features not planned:

• Login or accounts
• Automatic plan optimisation
• AI coaching or recommendations
• Social or gamification features

---

Project philosophy

Many training apps attempt to optimise away uncertainty.
RunPal takes the opposite approach: it assumes the runner’s perception matters and keeps adjustments lightweight, transparent, and reversible.

This project is intentionally small, focused, and explainable.