Data-Driven Coaching: Theory to Athlete Action

IRSRI Coach Development — 14 Weeks

Good coaching is not a contest to collect the most numbers. It is the disciplined practice of asking better questions, gathering useful evidence, listening to athletes, and turning what we learn into the next sensible action. This course connects the human theory of coaching with practical data from watches, GPS, heart rate, video, performance history, and athlete conversations.

The course is designed for mixed-experience coaches working across sports. Running is the primary worked example because many IRSRI athletes are runners, while every module includes a multi-sport translation using movement analysis, timing, practice observations, and sport-specific performance targets.

What coaches will learn

• Define a durable coaching purpose and a healthy coaching identity.

• Build motivation, confidence, culture, and athlete ownership.

• Evaluate the quality, limits, and uncertainty of performance information.

• Design practices and feedback that produce useful learning.

• Combine athlete self-report with wearable and video evidence.

• Turn longitudinal data into individualized goals, training decisions, and competition plans.

• Communicate conclusions honestly without turning coaching tools into medical diagnoses.

Weekly rhythm

Part 1 — Instruction, 5–7 minutes. A ready-to-deliver coaching script introduces one memorable principle, a practical IRSRI example, and one challenge to carry into the lab.

Part 2 — Application, 30–45 minutes. Coaches work with their own data during Weeks 1–6. Beginning in Week 7, each coach works with an assigned group of 10–15 athletes and progressively turns their data into useful observations, conversations, and plans.

Participation assignment. Each week produces a practical work product. Assignments are for participation and coaching development; they are not graded.

Discussion preparation. Each coach finds and verifies one credible research source, uses Generative AI to create a structured summary, and arrives ready to discuss one finding, one limitation, and one coaching question.

The data-to-decision pathway

ASK → COLLECT → CHECK → INTERPRET → COACH → REASSESS

1. Ask: Begin with a decision or coaching question, not a dashboard.

2. Collect: Gather only the information that may help answer the question.

3. Check: Inspect missing data, context, sensor quality, and uncertainty.

4. Interpret: Compare like with like and connect numbers to athlete experience.

5. Coach: Offer one clear, proportionate action rather than an automatic plan.

6. Reassess: Observe the response and revise the decision when new evidence appears.

The four phases

1. Foundations & Motivation — Weeks 1–6

Clarify why you coach, how you define success, and how data can support culture, goals, confidence, individual attention, and responsible decision-making. Coaches practice on their own data and complete an app feedback brief in Week 3.

2. Practice & Feedback — Weeks 7–9

Receive an assigned group of 10–15 athletes. Progress from athlete and data onboarding to measured practice design and concise, actionable feedback.

3. Athlete State & Coaching Systems — Weeks 10–12

Interpret stress, arousal, recovery, communication, fueling, and workload as connected coaching problems. Separate useful field signals from medical conclusions.

4. Training Design & Athlete Achievement — Weeks 13–14

Use trends, uncertainty, and athlete context to create target-performance guidance, sample training, race or competition plans, and a final coaching clinic for the full athlete group.

Hands-on platforms

rslts Athlete Performance Platform connects watch, GPS, heart-rate, training, racing, development, and team information. Use its summaries to ask specific coaching questions and its development trends to compare an athlete with their own earlier data.

APOPT Watch Analysis reviews activity files, weekly trends, workout detail, predicted performances, training targets, and AI-ready exports. Data stays in the browser.

APOPT Form Analysis uses sport-specific video workflows for running, gait, starts, strength, throws, ball and racquet sports, aquatics, and other movements. Treat pixel-derived results as coaching information that should be checked against observation.

APOPT Heart Analysis reviews compatible ECG and RR-interval files for data quality, heart-rate response, recovery, and screening flags. It is not a diagnosis and does not replace clinical review.

Protect athlete data

Use only authorized athlete information. Remove names and direct identifiers before placing any athlete information into an external Generative AI tool. Do not put athlete records on public pages. Share the minimum information needed for the decision, and describe uncertainty instead of presenting a precise-looking estimate as fact.

Evidence + Generative AI workflow

1. Find one peer-reviewed article, systematic review, or professional consensus statement.

2. Ask AI for a 150-word summary of the research question, participants or context, method, key result, limitation, and coaching implication.

3. Open the original source and verify the title, authors, year, DOI or stable link, and the claim you plan to use.

4. Bring one evidence-based point, one limitation, and one discussion question.

Never cite a source you have not opened. Do not ask AI to “analyze this” without a specific decision question.

Open the tools

• rslts: https://rslts.run/

• APOPT Watch Analysis: https://apopt.com/watch/

• APOPT Running Form Analysis: https://apopt.com/form/form.php

• APOPT Heart Analysis: https://apopt.com/heart/

• rslts AI Summaries guide: https://rslts.run/docs/performance/data-analysis/ai-summaries/

• rslts Development Metrics guide: https://rslts.run/docs/performance/data-analysis/development-metrics/

Start here

Begin with 1. Foundations & Motivation, then move through the weekly modules in order. The course is cumulative: each phase adds a new layer to the same central skill—turning evidence into humane, useful coaching action.