1. Central hypothesis
This is a question to investigate, not a claim to prove. Health findings will be reported accurately regardless of direction — an improvement, a decline, and a null result are all treated as valid, publishable outcomes.
2. Why a marathon, and why four months
A marathon forces a real, sustained training load that's hard to fake. Four months is long enough for meaningful physiological adaptation — and for a plan to collide with sleep, work, injury risk, and ordinary life — without turning into an indefinite, never-concluding project.
3. Athletic target and project success criteria
The athletic target is fixed: 42.2 km in 5:00 at the BYD Singapore Marathon 2026. The project succeeds if that target is met and the process is documented well enough to reconstruct what was planned, what happened, what changed, and why. If the five-hour target is missed, the project is still a success as long as the result is reported in full, without spin — athletic outcome and documentation quality are judged separately.
4. Health markers observed
| Category | Markers | Core? |
|---|---|---|
| Cardiorespiratory | VO₂ max (Garmin-estimated; lab test if scheduled) | Yes |
| Recovery | Resting heart rate, sleep | Yes |
| Body composition | Weight, body fat %, muscle mass | Yes |
| Strength | A fixed benchmark set (protocol on the strength page) | Yes |
| Cardiometabolic | Blood pressure, glucose | Yes |
| Laboratory | Selected blood tests, extracted values only | Yes |
| Supplemental | HRV, if a source that's already in use later provides it | No |
HRV is deliberately not a core metric. Adding it as a requirement would mean asking readers to trust a device dependency (commonly a paid subscription service) this project doesn't need. Resting heart rate remains the core practical recovery marker instead.
5. Baseline and follow-up schedule
Baseline measurements are being scheduled and are not yet public — see the plainly labeled "Awaiting Data" state on the Dashboard rather than a number standing in for one. The intended cadence, once baselines exist: body composition and strength benchmarks at roughly monthly checkpoints; resting heart rate and sleep as a rolling weekly average from Garmin; blood pressure and glucose at baseline, mid-project, and post-race; a full blood panel at baseline and again shortly after the marathon.
6. Garmin and other data sources
- Garmin device + app — training data (distance, duration, pace, heart rate, elevation), and algorithm-based estimates (VO₂ max, resting heart rate, sleep staging).
- Home readings — a validated blood pressure monitor, a consistent bathroom scale, and a consumer-grade glucose reading where relevant.
- Clinical / laboratory — blood panels and any formal fitness testing (e.g. a lab VO₂ max test), if scheduled.
- Subjective self-report — perceived effort, pain, fatigue, motivation, logged alongside the numbers, not instead of them.
7. Estimates vs. measurements
These are not interchangeable, and the site tries never to imply otherwise:
- Device estimate — e.g. Garmin's VO₂ max, derived from an algorithm (Firstbeat) using pace, heart rate, and run data. Useful for trend, not a diagnostic number.
- Home reading — e.g. a blood pressure cuff at home. Real measurement, but with more variable technique and conditions than a clinical setting.
- Laboratory / clinical measurement — e.g. a blood panel or a gas-exchange VO₂ max test. The highest-confidence category, and the rarest.
- Subjective report — e.g. perceived effort or sleep quality. Real data about lived experience, not a physiological measurement.
Every reading published on this site states which of these four categories it belongs to. See Measurement methodology for the full detail.
8. Confounders and limitations
This is one 40-ish-year-old person, not a controlled trial. Training isn't the only thing changing over these four months — sleep, travel, work stress, diet, and Singapore's heat and humidity all move at the same time, and none of them can be held constant. Where a health marker changes, the site will name plausible contributing factors rather than crediting or blaming training alone. A single before/after comparison is never treated as proof of causation, and short-term noise (a few days of an elevated resting heart rate, a single low-glucose reading) is not reported as a trend.
9. Dr. Reddy's role and evidence standard
Dr. Varun Reddy is a sports-medicine and research collaborator on this project — not a treating clinician for readers. His background is confirmed and published on his profile; his review of the training plan is still in progress, and no commentary is published under his name until it's actually his, reviewed and dated. Where he does comment, the site distinguishes an evidence summary, his clinical/expert judgment, and a project-specific suggestion — three different kinds of claims — and links primary research or strong evidence reviews where that's appropriate.
10. How conclusions will be framed
- "After" is not treated as "because of." Sequence is described as sequence; causal language is reserved for when the evidence actually supports it.
- Individual observations are reported as observations about one person, not general recommendations for readers.
- Missing, inconsistent, or non-comparable readings are labeled as such rather than smoothed over.
- Changes in measurement method (a new scale, a different cuff) are flagged, and series measured differently are not silently merged into one line.
11. Project timeline
-
Aug 2026Foundation phase begins (Weeks 1–4)Plan v2 in effect. Baseline health measurements being scheduled.
-
Sep 2026Build phase (Weeks 5–9)Quality sessions and mileage increase; first monthly checkpoint.
-
Oct 2026Peak phase (Weeks 10–13)Peak long run (32 km, Week 13); second checkpoint.
-
Nov 2026Taper (Weeks 14–16)Mileage tapers; final pre-race checks.
-
4 Dec 2026Race week & BYD Singapore Marathon 2026Race day, inside Week 17 of the plan (30 Nov–6 Dec).
-
Within weeks of race dayImmediate post-race retrospectiveResult, post-race health comparison where available, and a full accounting of plan vs. actual.