Maintained conclusion register¶
Reference contents · Glossary · Evidence method · Sources
This is a question-routing index, not a second set of judgments. The linked chapter owns each conclusion and its supporting, conflicting and limited evidence. Qualitative grades are editorial shorthand, not numerical scores or formal GRADE ratings. Checked 2026-10-02.
Aerobic fitness, volume and easy intensity¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| PHY-C01 — Sustainable cycling power is the interaction of VO2max, the fraction that can be held, and efficiency, not an FTP trait. | Strong evidence | PHY01, PHY02, PHY09, PHY14 |
| PHY-C02 — Short-term VO2max gains are mostly oxygen delivery; extraction is already high in well-trained athletes. | Moderate evidence | PHY02, PHY03, PHY04, PHY05, PHY06, PHY07, PHY08 |
| PHY-C03 — Critical power, MLSS, second thresholds and FTP are related and not interchangeable. | Strong evidence | PHY10, PHY11, PHY12, PHY13, PHY40, PHY41, PHY42 |
| PHY-C04 — Mitochondrial and capillary gains follow training load, not a unique Zone 2 pathway. | Moderate evidence | PHY23, PHY24, PHY25, PHY26, PHY27, PHY56 |
| PHY-C05 — Lower gains at higher baseline fitness and a null study-level volume regression are not a randomized weekly-hour dose-response trial. | Moderate evidence | PHY23, PHY28, PHY29, PHY30, PHY34, PHY60, PHY61 |
| PHY-C06 — Progress volume by recoverable steps; a universal 10% rule is not validated for cyclists. | Mostly coaching practice | PHY46, PHY47, PHY48, PHY49, PHY50, PHY51 |
| PHY-C07 — One long ride and two shorter rides of equal time are not known to be equal training stimuli. | Plausible but uncertain | PHY30, PHY57, PHY21, PHY19 |
| PHY-C08 — No retrieved direct longitudinal comparison ranks 55-60% versus 65-75% FTP. Whether those bands differ in performance adaptation is uncertain. Fixed percentages of maximum misclassify intensity domains on separate evidence. | Plausible but uncertain | PHY55, PHY45, PHY38, PHY39, PHY40 |
| PHY-C09 — Research Zone 1, cycling Zone 2 and popular Zone 2 are different constructs. | Strong evidence | PHY31, PHY32, PHY56, PHY13 |
| PHY-C10 — Riding easier as weekly hours rise is a recovery-oriented coaching choice, not a proven workload rule. | Mostly coaching practice | PHY33, PHY30, PHY22 |
Distribution, periodization and recovery¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| TID-C01 — 80/20 describes a common high-volume session pattern; it is not an experimentally identified optimum for every volume, phase, or event. | Moderate evidence | TID01, TID02, TID15, TID36, TID53 |
| TID-C02 — Polarized training has not been shown superior to pyramidal or other non-polarized distributions for trained cyclists' time-trial performance; a small VO2peak edge and older threshold-heavy contrasts are narrower claims. | Moderate evidence | TID04, TID05, TID06, TID07, TID08, TID09, TID10, TID11, TID12, TID13 |
| TID-C03 — Session-goal, time-in-zone, and 3/5/7-zone labels are different measurements and must not be treated as interchangeable prescriptions. | Strong evidence | TID01, TID03, TID15 |
| TID-C04 — When load is matched, ordering 4-, 8-, and 16-minute mesocycles as longer/lower-intensity first versus shorter/higher-intensity first has not been shown to matter over 12 weeks. Neither Sylta order is a pure VO2-before-threshold or threshold-before-VO2 sequence. Elaborate base-build-peak or reverse sequencing is not a demonstrated requirement. | Moderate evidence | TID17, TID41, TID42, TID43 |
| TID-C05 — Short HIT or moderate-intensity blocks can improve well-trained cyclists, but superiority over even spacing is conflicting and not established. | Plausible but uncertain | TID18, TID19, TID20, TID21, TID22, TID44, TID45, TID46 |
| TID-C06 — About two hard endurance sessions per week is a workable and commonly studied dose, not an experimentally proven optimum versus one or three. | Mostly coaching practice | TID23, TID37, TID38, TID39, TID47, TID48 |
| TID-C07 — About 48 hours is a reasonable default gap after severe work based on autonomic recovery summaries, not a performance law; multi-hour easy rides are training, not free recovery. | Moderate evidence | TID24, TID25, TID26, TID49 |
| TID-C08 — A universal maintenance minimum and an obligatory mathematical deload cannot be prescribed; pre-event tapers have better evidence than routine 3:1 unload weeks. | Moderate evidence | TID27, TID28, TID29, TID30, TID31, TID32, TID33, TID34, TID50, TID56 |
| TID-C09 — Distribution should change with volume, days, phase, and event, but specific zone percentages for amateur cyclists are not fixed by factorial trials. | Plausible but uncertain | TID15, TID16, TID35, TID36, TID53, TID54 |
VO₂-focused intervals¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| INT-C01 — In one well-trained cyclist cohort, higher mean %VO2max during 5x8 min intervals at PO40min was associated with larger gains in VO2max, Wmax, power at 4 mmol, and a composite index. This is not proof that raising the fraction causes the gain. | Moderate evidence | INT01, INT24, INT46 |
| INT-C02 — Time ≥90% VO2max is a noisy acute marker, not a validated surrogate for subsequent adaptation. | Moderate evidence | INT01, INT21, INT22, INT25, INT33, INT40, INT46 |
| INT-C03 — Effort-matched 30/15 series can outperform 4x5 min intervals on several outcomes in trained male cyclists, but this is not a universal short-versus-long rule, and the 2015 trial did not measure the proposed time≥90% mechanism. | Moderate evidence | INT03, INT04, INT05, INT06, INT21, INT26, INT41, INT54 |
| INT-C04 — Ordering 4x16, 4x8, and 4x4 over 12 weeks does not detectably change outcomes when load is matched. | Moderate evidence | INT09, INT10, INT48 |
| INT-C05 — A unique 4-8 minute sweet spot is not established for well-trained cyclists. It is supported only as an intensity-by-duration interaction in one recreational cohort. | Plausible but uncertain | INT12, INT13, INT09, INT10 |
| INT-C06 — Exact 5x5 and isolated 1-minute VO2 formats still lack a trained-cyclist performance head-to-head. 40/20 has one 4-week randomized comparison against 4x8 that did not separate VO2peak; calling any of these optimal remains coaching practice. | Mostly coaching practice | INT14, INT17, INT42, INT46, INT13, INT12, INT53 |
| INT-C07 — Fixed %FTP, %MAP, or %VO2max anchors are weak. Self-paced maximal effort raises acute time near VO2max, especially for longer intervals. | Moderate evidence | INT06, INT11, INT34, INT44 |
| INT-C08 — Within-interval power variation sometimes raises acute time near VO2max and sometimes does not. No retrieved trial shows this changes adaptation. | Plausible but uncertain | INT07, INT08, INT02, INT21 |
| INT-C09 — Near-maximal intervals are not uniquely required for endurance gains in well-trained cyclists. RPE and breathlessness are not measurements of %VO2max. | Moderate evidence | INT37, INT38, INT39, INT01, INT17 |
| INT-C10 — Maximal aerobic power is test-protocol dependent, so an unnamed %MAP target is not portable across ramp designs. | Moderate evidence | INT35, INT36 |
Threshold, durability and long rides¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| DUR-C01 — FTP, CP, MLSS and RCP correlate at group level but are not interchangeable for an individual. | Strong evidence | DUR01, DUR02, DUR03, DUR04, DUR05, DUR06, DUR07, DUR08, DUR09, DUR10 |
| DUR-C02 — Sustainable power is raised by more than one intensity domain. A threshold-heavy week is a viable stimulus, not a proven best or exclusive system. Cove 2025's null polarised versus non-polarised contrast is not a measured 80/20 or pyramidal comparison. | Moderate evidence | DUR11, DUR12, DUR13, DUR14, DUR15, DUR16, DUR17, DUR18, DUR19, DUR20, DUR21 |
| DUR-C03 — Named FTP recipes and a uniquely efficient sweet-spot band are coaching practice, not trial-established optima. | Mostly coaching practice | DUR11, DUR15, DUR16, DUR22 |
| DUR-C04 — Threshold and short time-trial power fall after prior work; absolute 1500/2000/3000 kJ cut-points are not universal. | Strong evidence | DUR27, DUR28, DUR29, DUR31, DUR32, DUR34, DUR36, DUR40, DUR51 |
| DUR-C05 — Fresh FTP or VO2max does not fully identify who retains power after prior work. | Moderate evidence | DUR32, DUR33, DUR34, DUR35, DUR39, DUR41, DUR30 |
| DUR-C06 — Durability can improve. Heavy strength twice weekly for 12 weeks improved 5-min power after 185 min in one controlled cyclist study; that does not prove an effect independent of fresh fitness. The optimal endurance-session type is uncertain. No matched fresh-versus-late interval trial was retrieved. | Plausible but uncertain | DUR43, DUR44, DUR45, DUR46, DUR47, DUR42, DUR30, DUR33 |
| DUR-C07 — No controlled evidence shows fatigue-state intervals improve late-race performance more than the same intervals done fresh. A higher recovery cost is a plausible risk, not a matched early-versus-late result. | Mostly coaching practice | DUR32, DUR34, DUR30, DUR50, DUR40 |
| DUR-C08 — No trial establishes an optimal long-ride duration or weekly share. A 25-40% share of cycling time is an untested coaching span, not a ceiling. One example that already includes strength inside 7/9/12 h is 120/390, 180/480 and 210/660 cycling minutes (~31%, ~38%, ~32%). | Mostly coaching practice | DUR57, DUR58, DUR23, DUR30, DUR39 |
| DUR-C09 — Carbohydrate availability can change the moderate-to-heavy transition and estimated CP, but that is not structural durability. Main 2026 is depletion plus 24 h of ≤1 vs ≥9 g/kg CHO, not a skipped breakfast. Norte 2026 used 3 h at 95% GET, not FTP; 120 g/h beat water for preserving CP and raised some GI scores versus water, while 120 vs 60 g/h was not significant and 90 g/h was not tested. Neither universal tolerance nor a fasted long ride as a durability method is supported. | Moderate evidence | DUR51, DUR52, DUR27, DUR28, DUR35, DUR53, DUR54, DUR55, DUR56, DUR49, DUR50 |
Fueling and energy availability¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| FUEL-C01 — Carbohydrate versus none during prolonged exercise is strong evidence. The hourly target is moderate evidence. Superiority of 100-120 g/h over 60-90 g/h is plausible but uncertain; high dose is optional, not mandatory. The claim grade is Moderate because the practical conclusion is the hourly/high-dose claim, not the placebo contrast. | Moderate evidence | FUEL01, FUEL02, FUEL03, FUEL04, FUEL05, FUEL06, FUEL07 |
| FUEL-C02 — Glucose or maltodextrin plus fructose, via SGLT1 and GLUT5, raises exogenous oxidation above a single-source ceiling; 2:1 and about 1:0.8 both work, and ratio differences are smaller than mixed versus glucose alone. | Strong evidence | FUEL03, FUEL08, FUEL09, FUEL10, FUEL11, FUEL12, FUEL13, FUEL14, FUEL15 |
| FUEL-C03 — Glucose plus sucrose supplying the same overall 2:1 glucose:fructose as free glucose plus fructose oxidised similarly in Trommelen 2017 (10 trained men, 108 g/h; 0.6 g/min glucose + 1.2 g/min sucrose vs 1.2 glucose + 0.6 fructose). Sucrose alone was not tested in that comparison and is not claimed to match a free mix. Separately, sucrose versus glucose protects liver glycogen during moderate cycling and speeds liver, not muscle, repletion. Table sugar is a cheap legitimate carbohydrate, not a proven sports-drink clone. | Moderate evidence | FUEL16, FUEL17, FUEL18, FUEL19, FUEL20, FUEL21 |
| FUEL-C04 — Oxidation, absorption, and tolerance are separable. Short gut training can reduce symptoms in some runners but does not reliably unlock 120 g/h as a performance requirement. | Moderate evidence | FUEL09, FUEL15, FUEL22, FUEL23, FUEL24, FUEL25, FUEL26, FUEL27 |
| FUEL-C05 — In-ride and around-session carbohydrate mainly protects session quality and next-day work; it is not evidence of a separate structural adaptation from extra carbohydrate. | Strong evidence | FUEL26, FUEL28, FUEL29, FUEL31, FUEL32, FUEL33, FUEL34, FUEL35, FUEL05 |
| FUEL-C06 — Periodised carbohydrate restriction has not reliably improved endurance performance in trained athletes. Gejl 2021 SMD 0.17 (95% CI -0.15 to 0.49) is an imprecise null, not proof of zero or of universal no benefit. Not fasting quality sessions is a risk-oriented practical choice, not proof of a null. A benefit of train-low remains plausible but uncertain. | Moderate evidence | FUEL28, FUEL29, FUEL30, FUEL36, FUEL37, FUEL38, FUEL39, FUEL40, FUEL41, FUEL42, FUEL43, FUEL44, FUEL45, FUEL46, FUEL47, FUEL48, FUEL49, FUEL58 |
| FUEL-C07 — When rapid glycogen restoration is the goal, about 1.2 g/kg/h early is the rate at which muscle resynthesis is described as maximised. It is not a minimum after every ride. Sucrose or glucose-fructose favours liver glycogen. Even 10 g/kg over 12 h may not fully restore muscle glycogen. Sollie 2023 is design-only here; the title is not an outcome. | Moderate evidence | FUEL21, FUEL50, FUEL51, FUEL52, FUEL53, FUEL77, FUEL78 |
| FUEL-C08 — Endurance protein needs are credibly about 1.4-2.1 g/kg/day, with point estimates near 1.6-1.8 on hard days; timing is secondary to the daily total unless carbohydrate is inadequate. | Moderate evidence | FUEL53, FUEL54, FUEL55, FUEL56, FUEL57, FUEL58, FUEL59 |
| FUEL-C09 — Fluid and sodium replacement must be individualised. Overdrinking, not a missed universal sodium dose, is the dominant cause of exercise-associated hyponatremia. | Strong evidence | FUEL15, FUEL60, FUEL61, FUEL62, FUEL63, FUEL64 |
| FUEL-C10 — Low energy availability and REDs can limit health and training, but screening is not diagnosis and there is no universal caloric threshold. The share of recovery limited by intake versus programming is not a known percentage. | Strong evidence | FUEL69, FUEL70, FUEL71, FUEL72, FUEL73, FUEL74, FUEL75, FUEL35, FUEL44 |
Sleep, heat and iron¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| REC-C01 — Acute sleep loss impairs physical performance, including endurance; total deprivation and late restriction are clearer harms than every partial short night, and afternoon tests are more vulnerable. | Strong evidence | REC02, REC03, REC04, REC06, REC01 |
| REC-C02 — Nine hours plus a nap is not a proven requirement. Elite athletes often sleep about 6.5-7 h and report needing about 8.3 h; short sleep is strongly tied to schedule, not a demonstrated extra biological need from training hours. | Moderate evidence | REC10, REC11, REC12, REC13, REC01 |
| REC-C03 — A few nights of sleep extension can maintain endurance time-trial performance better than normal or restricted sleep in a small cyclist crossover; basketball and tennis extension trials do not establish a cycling adaptation dose. | Moderate evidence | REC05, REC08, REC07, REC09, REC01 |
| REC-C04 — Daytime naps can improve acute physical performance and perceived fatigue, especially 30 to <60 min with a buffer before testing; 20-30 min naps do not reliably avoid sleep inertia, and cycling time-trial evidence is thin. | Moderate evidence | REC14, REC15, REC16, REC17, REC20, REC21, REC18, REC19 |
| REC-C05 — Severe sleep loss lowers muscle protein synthesis in young adults, and endurance overreaching can reduce objective sleep efficiency, but this is not evidence that ordinary short nights block VO2max or FTP adaptation. | Moderate evidence | REC22, REC23, REC24, REC25 |
| REC-C06 — Heat acclimation improves performance and tolerance in the heat more consistently than in cool conditions. Post-exercise sauna or hot-water immersion has uncertain performance effects in meta-analysis. | Strong evidence | REC28, REC29, REC30, REC31, REC33, REC37, REC41, REC42, REC57 |
| REC-C07 — Plasma-volume expansion within days is not the same as a rise in haemoglobin mass over weeks. A lower haemoglobin concentration after heat or training is often dilution. Heat-driven erythropoiesis is not established as the same mechanism as hypoxia. | Strong evidence | REC36, REC29, REC43, REC35, REC44, REC45, REC47, REC48, REC49, REC53, REC54 |
| REC-C08 — Validated heat doses must be stated exactly, and sauna is not hot-water immersion. No retrieved trial specifically tested 3x45 min/week sauna, so that dose should not be declared effective or ineffective. Jenkins 2025 tested 45 min hot-water immersion at 5 sessions/week, a different modality and frequency. Daily sauna has not been shown head-to-head to beat three sauna sessions for the same outcome. | Moderate evidence | REC39, REC35, REC52, REC56, REC45, REC51, REC49, REC41 |
| REC-C09 — Heat acclimation reduces heat strain but remains a stressor. Dehydration is not an effective haemoglobin-mass strategy. A studied hot-water block did not worsen sleep as an overreaching marker. Maximal heat is not a default amateur prescription, and adaptations decay within days to two weeks. | Strong evidence | REC33, REC34, REC56, REC58, REC31, REC49, REC54 |
| REC-C10 — No indiscriminate supplementation in an iron-replete athlete. Treat documented deficiency under professional assessment, not a ferritin-alone rule. Acute-phase ferritin can mask deficiency; the clinician uses haemoglobin, ferritin, transferrin saturation or other iron markers, inflammation, and context. A normal or high ferritin alone does not exclude deficiency or make treatment categorically contraindicated. Cycling causes less foot-strike haemolysis than running, and one temperate bout loses little sweat iron, but hepcidin versus sweat versus menstrual loss is sex- and context-dependent, not a universal ranking. | Strong evidence | REC59, REC60, REC62, REC63, REC64, REC66, REC68, REC69, REC70, REC71, REC72, REC76 |
Strength, warm-ups and mobility¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| STR-C01 — Heavy lower-body strength can improve cycling time-trial or time-to-exhaustion performance without a reliable VO2max increase. | Moderate evidence | STR01, STR02, STR03, STR07, STR09, STR11, STR16, STR17, STR06 |
| STR-C02 — Economy and anaerobic power are the more consistent companions of that performance effect; threshold and VO2max are not guaranteed. | Moderate evidence | STR01, STR05, STR07, STR09, STR10 |
| STR-C03 — Sprint and plyometric work are task-specific and are not a proven replacement for heavy strength. | Moderate evidence | STR14, STR15, STR19, STR20, STR18, STR55 |
| STR-C04 — Low-load core and activation routines are not a required performance method. There is no adequate evidence that they correct a claimed imbalance or prevent cycling injury; that gap is not proof that every such routine is ineffective. | Moderate evidence | STR11, STR37, STR38, STR39 |
| STR-C05 — Concurrent interference is real for explosive strength and next-day peak power, but a 6-hour cyclist rule is not established. | Moderate evidence | STR22, STR23, STR24, STR25, STR26, STR27, STR28, STR29, STR30 |
| STR-C06 — Two weekly sessions can build the strength effect and one session often maintains it. Those minutes and their recovery cost count inside a 7/9/12 h total budget by replacing easy cycling (examples 390+30, 480+60, 660+60 min), or add about 1-1.5 h if the hours were cycling-only. This is a supported implementation, not a 30-minute optimum. | Moderate evidence | STR02, STR04, STR12, STR13, STR21, STR52 |
| STR-C07 — A short cycling warm-up changes output; prolonged static stretching can impair isolated force; band activation is not required. | Strong evidence | STR40, STR41, STR42, STR45, STR46, STR47, STR48 |
| STR-C08 — Strength, mobility, and activation have not been shown in adequate cyclist trials to prevent cycling injury or correct imbalances. | Mostly coaching practice | STR31, STR32, STR33, STR34, STR35, STR36, STR60, STR61 |
Load and adaptation monitoring¶
| Conclusion | Grade | Original-source records |
|---|---|---|
| MON-C01 — TSS, NP, IF, CTL, ATL, and TSB are deterministic bookkeeping models. One hour at the entered FTP is 100 TSS by construction. CTL is not measured fitness. Default 42- and 7-day constants are vendor conventions, not fitted biological time constants. | Strong evidence | MON01, MON02, MON03, MON04, MON05, MON17, MON18 |
| MON-C02 — Equal TSS is not equal physiological stress. Steady-state TSS scales with IF squared, so the same TSS can represent very different kilojoules. A high FTP setting systematically understates TSS and inflates prescribed zones. | Strong evidence | MON01, MON02, MON04, MON19, MON20, MON24 |
| MON-C03 — Hours, kilojoules, heart-rate TRIMP, and session-RPE answer different questions. They correlate, especially when duration is shared, but are not interchangeable and should not be converted by a universal equation. | Moderate evidence | MON08, MON09, MON10, MON12, MON16, MON18, MON19, MON22, MON23, MON49, MON50, MON51, MON53 |
| MON-C04 — TSS and related impulses show dose-response associations with fitness changes in some observational blocks. That does not validate CTL targets, ramp-rate rules, or the default 42/7 constants as decision thresholds. | Moderate evidence | MON03, MON04, MON10, MON14, MON15, MON16, MON17, MON18 |
| MON-C05 — Acute:chronic workload ratios are not established causal injury models. TSB is a description of recent versus chronic load timing, not a measurement of readiness or injury risk. | Strong evidence | MON03, MON05, MON13, MON59 |
| MON-C06 — Whether a program is working is judged from standardized performance beyond test noise, not from a rising CTL. A 20-minute FTP test can miss a real maximal-lactate-steady-state change. | Strong evidence | MON03, MON16, MON27 |
| MON-C07 — A 20-minute test is reliable for tracking threshold-domain performance. Ninety-five percent of that power, critical power, and 3-minute end-test power are related but not interchangeable, and none measures VO2max. W' and 3-minute work above end-test power are too noisy or biased for week-to-week decisions. | Strong evidence | MON24, MON25, MON26, MON27, MON28, MON29, MON30, MON31, MON32, MON33, MON34, MON35, MON36, MON37, MON38, MON52 |
| MON-C08 — Heart rate at fixed power, decoupling, resting heart rate, and HRV can show multi-week trends but are too noisy and confounded for single-day pass/fail decisions. The 5% decoupling rule is coaching practice, not a trial endpoint. | Moderate evidence | MON05, MON06, MON40, MON41, MON48, MON54, MON55, MON56 |
| MON-C09 — Subjective wellness is often more sensitive to acute load than common objective markers. HRV-guided training has small, often non-significant group effects on performance. A composite of HRV, resting heart rate, and wellness is more defensible than HRV alone and is not proven superior to a well-built fixed plan. | Moderate evidence | MON39, MON42, MON43, MON44, MON45, MON46, MON47, MON57, MON58 |
| MON-C10 — When metrics conflict, the repeated performance test and the internal response outrank the smoothed external-load score. The exact weekly checklist is a practical synthesis, not a tested package. | Moderate evidence | MON13, MON19, MON27, MON39, MON40 |