Most runners I talk to are using HRV wrong. They see a red score, skip workouts, and three weeks later wonder why race day felt so hard.
The Protocol
| Parameter | What I Do |
|---|---|
| Timing | Morning, immediately on waking |
| Position | Sit up, same position every day |
| Duration | About 1 minute recording |
| Device | Same device always |
| App | HRV4Training (shows raw numbers, no black-box scores) |
I measure HRV seated in the morning because it’s more sensitive to training stress. Overnight HRV can look great while adaptation is quietly falling behind.
Inconsistent measurement = useless data.
HRV only becomes useful after several weeks of consistent measurement establish your baseline, from there I look at short trends around that range.
The Daily Check I Run
- How do I feel right now? (This is #1. Always.)
- How hard was yesterday's session?
- What does HRV say relative to my normal range?
Perception leads. HRV confirms.
When all three align, the signal is strong. When they conflict, I trust how I feel over what the app says.
The question: Am I absorbing training stress, or just accumulating it?
What I Pay Attention To
- 7 day HRV trends relative to my own baseline, not single days
- HRV during hard blocks
- Hard training + HRV stays normal = I’m absorbing load, continue
- Hard training, I feel off, + HRV trending down = stress accumulating, adjust intensity
- HRV paired with resting HR
- Both moving the same direction = real signal
- HRV during deloads
- Should rebound. If it doesn’t, something else is driving stress
- Mismatch between HRV and how I feel
- HRV low + feel great = train
- HRV low + feel terrible = investigate and adjust
- Life stress confirmed by HRV
- Poor sleep, travel, illness, work stress → HRV down → training is not the problem
- Life stress confirmed by HRV trends (sick, deadline, travel → feel off → HRV down → adjust)
If HRV says "recovered" but I feel terrible, something's off. If HRV says "stressed" but I feel great, I train.
What I Actively Ignore
- One-day drops
- HRV race week
- HRV that contradicts how I feel and how I'm performing
A quick credit where it’s due: much of how I think about HRV is shaped by the work of Marco Altini.
His app, HRV4Training, is what I use to view raw HRV without black-box scores, and it’s where you can find more detail on the exact morning measurement approach I follow.
Practical Takeaways
- HRV is a stress barometer, not a training commandment
- Trends matter. Single days don’t. Look at 7-14 day patterns.
- Suppressed HRV during hard training is expected, not a warning.
- Trust yourself over the algorithm. Feeling > app.
- Consistency beats precision. Same time, same position, same device.
- Consistency beats precision. Same time, same position, same device.
Bottom line: Use HRV to ask better questions, not to cancel workouts.
48-Hour Action Checklist
[ ] Stop reacting to single-day HRV changes [ ] Screenshot your last 7 days of HRV data [ ] Compare trends to sleep, fueling, and training load [ ] Decide how HRV will (and won’t) influence your next training block [ ] If HRV creates more anxiety than clarity, delete the app
Headlines (Ranked by Predicted Open Rate)
-
Your Red HRV Score Might Be Lying to You - Type: Curiosity - Strongest open loop. Readers will click to find out what their app is getting wrong. Direct challenge to a daily decision they're making.
-
Stop Letting Your Watch Cancel Your Workouts - Type: Contrarian - Pattern interrupt. Most runners assume they should obey the app. This challenges that assumption directly.
-
The HRV Mistake That's Making You Undertrained - Type: Benefit (negative framing) - Stakes are clear. Targets the pain point of unexpectedly poor race performance.
-
HRV During Hard Training Should Be Low (Here's Why) - Type: Specific/Educational - Counterintuitive fact as headline. Appeals to science-minded readers who want to understand mechanisms.
-
When to Ignore Your HRV (A Decision Framework) - Type: Benefit (utility) - Promises a practical tool. Slightly less emotional pull but high relevance for daily decision-making.