SYNTH-30 · Guide

Low readiness

What the athlete says beats what the strap measures, and, by the base's inference, autoregulation helps the athlete who overreaches while excusing the one who trains inconsistently

Blimited evidenceMostly from recreational to elite athletesWritten before October 2026, being re-checked

The decision this framework makes

Whether today's session changes, by how much, and on what evidence — and, before any of that, whether this athlete should be autoregulating at all.

1. The subjective data is the better data

REF-368 (A, systematic review of 56 studies) is the entry that should govern this whole area: subjective self-reported measures reflected acute and chronic training load with superior sensitivity and consistency compared with the commonly used objective ones (resting heart rate, blood markers, exercise HR and oxygen-uptake responses) — and the two kinds of measure generally did not correlate with each other.

Not "are a useful supplement" — the review says they may stand alone. Self-reported well-being dipped after an acute load increase and during chronic hard training, and recovered when load was cut.

Not from a study; the base's own practice: a short morning check on sleep, soreness, energy, mood and motivation — the review names mood and perceived stress as example measures and gives no item list or duration.

This is why Blocks' readiness instrument is subjective-first with HRV/RHR optional, and that ordering follows the review's finding rather than a UI convenience. REF-367 (B+, narrative review) adds the framing: no single definitive marker of fatigue exists; what carries information is the dissociation between external load (power, pace) and internal load (RPE, heart rate) in an individually monitored athlete. Not from a study; the base's own practice: the minimum stack we run — morning HRV + the athlete's RPE of yesterday's session + a short wellness check — is our operational default built on that principle; Halson prescribes no stack.

Note the middle item — retrospective RPE of the last session is monitoring, and it is nearly free.

Caveat worth stating to athletes: self-report only works if it is honest. REF-368 lists social-desirability bias among the method's limitations; "always 5 out of 5" is a known failure mode, and coaching athletes toward honest reporting is house practice built on that caveat.

2. Autoregulation is not for everyone — this is the profile rule

The HRV-guided trials are consistent in direction: in REF-362 (Javaloyes 2019, 17 well-trained cyclists, 8 weeks) the HRV-guided group improved peak power output +5.1%, power at VT2 +13.9% and 40-min time trial +7.3% while the traditional-periodization group did not improve significantly (no significant between-group difference; magnitude-based inference "likely beneficial" for the TT); in REF-438 (Vesterinen 2016, two-arm controlled trial, allocation method not stated; 40 recreational runners, 8 weeks) the HRV-guided group did significantly fewer moderate/hard sessions (13.2 vs 17.7) and still improved 3000-m time (+2.1% vs +1.1%, small between-group effect), with VO2max up in both groups.

Inference, not a finding of either trial — the base's own reasoning about who this transfers to:

The motivated athlete prone to overreaching — HRV-guided autoregulation *preserves* adaptation by holding hard days back when the system says no. That is the closest thing the trials tested: volunteers who measured every morning for 8 weeks and, in REF-438, did the hard session only when HRV sat within their band (REF-362's entry gives no rule; "consistent" is our reading of who completes such a trial).
The undertrained or inconsistent athlete — structure-first beats HRV-first, because HRV-guided training can enable "excuse avoidance". A soft number becomes permission to skip, and the athlete who most needs consistency gets least of it. No trial in this base tested HRV-guidance in inconsistent athletes; this is reasoning from who the trial participants were.

So before offering to move a session on a readiness signal, ask which athlete this is. House practice, not a tested rule: for someone who trains 3 days out of 5 planned, the answer to a mediocre readiness score is usually "do the session, easier if you must" — not "take the day".

3. What a signal actually is

A direction over ~3 days, not a morning. One poor night is context. REF-361 (B+, narrative review + elite case data) is explicit that single-day HRV values are noisy and must be read as an averaged trend against resting HR and the training phase, per athlete — and that in elites HRV can fall with fitness (saturation) or with freshness in a taper.Blocks' aggregates exist for exactly this — compare today against the 28-day baseline and the 3-day rolling value, and check n28 before trusting the baseline at all; a baseline built from four entries is not a baseline (house practice; the window lengths are ours).
A late hard session goes with a lower HRV that night. In a year of wearable data from 14,689 active adults, a maximal-strain session ending 2 h before habitual sleep onset went with a nocturnal heart rate 3.9 beats/min higher and HRV 8.3 units (14.1%) lower than a light one, and every strain level went with a slightly lower night HRV than rest days even when it ended 6-10 h before sleep (about 1.2-1.6 units after light sessions, 4.5-5.8 after maximal ones) (REF-1193, B+, observational, wrist device, not athletes).So a low overnight number after a hard evening may be largely the session's footprint: read it against nights after similar sessions before treating it as low readiness (inference); the authors read it as delayed parasympathetic recovery.
RPE drifting up at the same power or pace is the cheapest marker to collect (house judgement) — it is the external-vs-internal dissociation REF-367 describes as a possible sign of fatigue; in the horse overload model REF-139 (C, 7 horses, mechanism only) the early signs were behavioural (irritability, inability to complete the training, weight loss) while test performance held, and the authors concluded no single parameter detects early overtraining.
HRV that has not recovered means today's intensity comes down — not that the block is over.Not from a study; the base's own practice (that is also the logic of the HRV-guided protocol in REF-438: a morning value outside the band swaps the hard session for low-intensity work, it does not cancel the block — the hard session ran only when HRV sat within the individual SWC band, so an unusually high value also meant an easy day; REF-362's entry says only that prescription was based daily on HRV and does not give its decision rule).
Faster heart-rate recovery is not automatically freshness. In functional overreaching HRR gets FASTER — +8 ± 5 bpm in the 10 of 21 overloaded triathletes who developed f-OR (31 in the study, 10 of them controls), alongside a −2.1% performance drop and high perceived fatigue (REF-208, B+); read it together with RPE at the same power before calling the athlete fresh.
Sustained easy work below LT1 does not add autonomic cost — HRV was back to baseline within 5–10 min after 60 or 120 min below VT1 in highly trained runners, versus ~30 min after threshold or interval work (REF-161, B) — which is why "easy day" and "rest day" are different prescriptions; that the easy day usually wins on a low-readiness day is the base's inference — the study measured acute HRV after single sessions in fresh, highly trained runners, not an easy-vs-rest comparison.

4. The response ladder

Change the smallest thing that resolves the signal, and say what would reverse it (Not from a study; the base's own practice):

1

Keep the session, drop the targets — same structure, lower intensity band. Preserves the habit and most of the stimulus.

2

Keep the duration, remove the intensity — convert to easy volume (REF-161: nearly free autonomically, tested up to 120 min).

3

Swap the day — move the quality session later in the week rather than deleting it, if the week still has room (SYNTH-26 for how much room that is).

4

Cut the week — only on a sustained direction plus the markers in SYNTH-29, not on a number.

5

Stop — illness is not a readiness decision. Fever or its company (chills, malaise, aches), a chesty cough, breathlessness or vomiting: no training, no exceptions (REF-609, C, for fever; the rest is the base's own practice, as in SYNTH-36). Symptoms above the neck only and no fever: easy and short, reassess tomorrow (base practice, SYNTH-36).

A sustained shift over weeks in several nightly signals away from the athlete's own baseline, together with unexplained fatigue — especially in an older athlete or one with known heart disease — is a reason for a medical check, not a lighter week (REF-1057, C, a single case report analysed in hindsight: a 76-year-old with advanced structural heart disease).

Practical Application

Ask before you adjust. The check-in numbers are a prompt for a question, not a verdict; the athlete's sentence usually contains the answer (REF-368: the subjective layer is the more sensitive one).
Check the consistency pattern first — autoregulate the athlete who tends to overreach, hold structure for the one who trains inconsistently, and say which pattern you think you see so they can correct you. This is a hypothesis about recent behaviour, not an athlete type: types are retired (SYNTH-13) and the goal and the stimulus it needs come first (SYNTH-42). Inference from who REF-362 and REF-438 studied, not a tested split.
Read trend, not today — 3-day direction against a 28-day baseline, and check n28 first (REF-361 for averaging; the windows are ours).
Prefer lowering targets over cancelling. A completed easy session beats a skipped hard one on every timescale that matters (house practice).
Never moralise about a missed day, and never use a readiness number to argue with an athlete who says they feel fine (house practice).
No wearable? Nothing is lost. The best-supported instrument here is a short set of subjective questions (REF-368; REF-367 for the load-dissociation logic) — an athlete with no strap is not flying blind.

What NOT to do

Do not treat one bad morning as a signal (REF-361: single-day HRV is noise until averaged).
Do not offer HRV-guided autoregulation as the fix for an athlete whose problem is consistency — the trials were run in consistent athletes, and the "excuse avoidance" risk is our inference (REF-362, REF-438).
Do not trust a baseline without checking how many entries built it (house practice).
Do not read fast HRR as recovery on its own (REF-208).
Do not let objective numbers overrule what the athlete tells you; the review says the athlete is the more sensitive instrument (REF-368).
Do not silently rewrite the plan because a number moved. Propose, explain, and let the athlete choose (house practice).

Limitations

The HRV-guided trial literature is small and its participants were consistent volunteers — 17 well-trained cyclists (REF-362) and 40 recreational runners (REF-438) who measured every morning for 8 weeks, with no significant between-group difference in the cycling trial and a small effect in the running one.The split in §2 (overreacher vs inconsistent athlete) is the most decision-relevant thing here and is inference from that population, not a finding; it is the part most worth re-checking as evidence accumulates.
[REF-500] carries no DOI and is a composite of foundational scales rather than a single study; treat it as the methodological background, not as a citable finding.
[REF-132] was a duplicate of [REF-361] (the stub was emptied on 2026-09-27); cite REF-361.
[REF-139] is a horse model (n=7); its behavioural markers are instructive, not human evidence.
The response ladder in §4 is reasoning, not a finding. No entry in this base compares "lower the targets" against "swap the day" against "take the day off" for outcome. It is a defensible ordering built on the monitoring evidence — present it that way.
Nothing here is validated against Blocks' own readiness scale specifically; the instrument's item list and the 3-day / 28-day windows are house choices and have not been checked against outcomes in this population.
How this entry was checked and changed
Why this level

(the subjective-first rule rests on an A-level systematic review, REF-368; the HRV evidence is B+ (REF-438, REF-361) and B (REF-362); the overreaching-versus-inconsistent split and the response ladder are house practice)

Content correction 2026-09-27
"REF-367 names the minimum stack: morning HRV + session RPE + weekly 5-item wellness" -> Halson prescribes no stack; the stack is the base's own heuristic. REF-367's finding is "no single definitive marker; external–internal load dissociation".
"REF-362 (RCT + SR) draws the line: overreacher vs inconsistent athlete / excuse avoidance" -> not a trial result; REF-362 is Javaloyes 2019 (17 well-trained cyclists: PPO +5.1%, WVT2 +13.9%, 40-min TT +7.3%, no significant between-group difference). The profile split is relabelled inference; REF-438 (Vesterinen 2016, 40 recreational runners, 13.2 vs 17.7 hard sessions, 3000 m +2.1% vs +1.1%) added.
"REF-368 flags 'teach athletes not to answer 5/5'" and the 30-second five-item check -> not in the paper; relabelled house practice; REF-368's actual results added (56 studies, subjective and objective did not correlate, subjective more sensitive).
"RPE at the same power is the most informative marker (REF-139)" -> REF-139 (horses) found behavioural signs with test performance NOT decreased and no single detecting parameter; RPE re-attributed to REF-367's dissociation logic.
"HRV not recovered → intensity down (REF-161)", "one bad morning (REF-161)" -> REF-161 is acute post-exercise HRV in runners (now B); relabelled practice / re-sourced to REF-361 averaging; REF-161's 5–10 min vs ~30 min numbers added.
"HRV-guided literature is Tier 4-5 heavy" -> the trials are well-trained cyclists and recreational runners; REF-208 population corrected to 31 triathletes.
REF-438 called an RCT ("HRV-guided RCTs") -> the corrected entry is a two-arm controlled trial with allocation method not stated (B+); wording changed to "trials". Added REF-438's actual decision rule: hard session only when HRV was within the SWC band, so a high value also meant an easy day.
Cross-references to duplicate stubs repointed to their canonical entries (REF-132->REF-361); the stubs were emptied on 2026-09-27.
Level parentheticals synced to the cited entries' current Level lines: REF-361 A -> B+.
Level parenthetical "over A-level reviews and RCTs" -> one A-level consensus plus B+ trials and reviews.
Resync 2026-09-28: Level parenthetical "an A-level consensus, REF-368, and B+ HRV trials" -> REF-368 is a systematic review (A), not a consensus; REF-362 is now B (parallel RCT n=17, no significant between-group difference); named.
Resync 2026-09-28: "the logic of the HRV-guided protocols in REF-362 and REF-438: a depressed morning value swaps the hard session for easy work" -> only REF-438 states a decision rule (hard session only when HRV within the SWC band); REF-362's entry gives none. Scoped to REF-438.
Resync 2026-09-28: "REF-368 lists social-desirability bias as the method's main limitation" -> the entry lists it among its limitations, not as the main one.
Resync 2026-09-28: Stub-repointing artefacts "REF-361 (= REF-361)" and "REF-361 are the same paper filed twice" -> the duplicate was REF-132 (emptied 2026-09-27); corrected.
Verification 2026-10-01: all numbers attributed to REF-368, REF-362, REF-438, REF-361, REF-161 and REF-208 (apart from the 10-of-31 denominator) match the entries; level/design/n parentheticals match the head blocks.
Verification 2026-10-01: the title carries the same unlabelled profile claim as the lead ('autoregulation helps one kind of athlete while excusing another'); not rewritten here — owner's call whether to add 'our inference'.
Verification 2026-10-01: REF-500 (C+, composite, no DOI) and REF-132 (emptied duplicate stub) appear only in Cross-References/Limitations, where the card already says so; no body claim rests on them.
Verification 2026-10-01: rows counted = §3 signal bullets (5), §4 ladder (5), §2 profile split (2), §1 subjective-first and the minimum stack (2). Evidence: subjective-first, trend-not-morning, fast HRR, easy work autonomically cheap. Practice: ladder x5, profile split x2. Mixed: minimum stack, RPE-drift marker, HRV-not-recovered.
Verification 2026-10-01: no contradiction found with SYNTH-26, SYNTH-29 or SYNTH-35 on the lines that point here.
2026-10-01: cross-framework seams. §4 step 5 said 'illness or fever … No training, no exceptions', which contradicted the neck check in SYNTH-36 and the assistant's illness rule; it now reads: fever or its company, chesty cough, breathlessness, vomiting = stop; above the neck only and no fever = easy and short, reassess tomorrow. 'Decide by athlete type first' reworded to a consistency pattern (a hypothesis, not a type) with pointers to SYNTH-13 and SYNTH-42, because SYNTH-13 retired athlete types; the title and Level line now say the split is the base's inference.
2026-10-02: aligned with the rewritten SYNTH-42 (missing stimulus, one lever per block, level C+): "the goal and its limiter come first" -> "the goal and the stimulus it needs come first".
Revision 2026-10-03: added one labelled sentence to §4 step 5 (Stop): a sustained multi-signal shift from the athlete's own baseline over weeks plus unexplained fatigue, especially in an older athlete or with known heart disease, means a medical check, not a lighter week (REF-1057, Lautman 2026, C, single case report, hindsight analysis). No rule, default or level changed.
Revision 2026-10-26: folded REF-1193 (Leota 2025, B+, 14,689 active adults, WHOOP): a late hard session goes with a lower HRV and higher heart rate that night (observational), so §3 now asks to read a low overnight number after a hard evening against nights after similar sessions before treating it as low readiness (inference). Default, ladder and level unchanged.