When a Published Plan Is the Right Tool

Abstract

A published plan — the Pete Plan ([1] The Pete Plan, Level 5), a Concept2 training plan ([2] Concept2, Level 5), a British Rowing "Go Row Indoor" series ([3] British Rowing, Level 5) — is the right tool when the goal has a known shape, the week is predictable, and the body is responding. It is the wrong tool when the assumptions break. The peer-reviewed literature on periodisation, load monitoring, and coaching effectiveness converges on this framing. The [4] Bompa & Haff 2009 Periodization 5th edition established the systematic vocabulary Western coaching inherited ([4] Bompa & Haff 2009, Level 5). The [5] Issurin 2008 block-periodisation review found that concentrated blocks beat distributed loading when the prescribed dose is reached ([5] Issurin 2008, Level 5). The [6] Kiely 2018 critical review in Sports Medicine placed the experimental base on more honest footing: periodisation's literature is thinner than the textbooks claim ([6] Kiely 2018, Level 5). The [7] Banister & Calvert 1980 fitness-fatigue TRIMP decomposition and the [8] Foster 2001 session-RPE method are the operational anchors for what the plan has to respect ([7] Banister & Calvert 1980, Level 5; [8] Foster 2001, Level 5). The [14] Mujika & Padilla 2000 detraining review established the timeline on which the plan's assumptions break ([14] Mujika & Padilla 2000, Level 5). The [18] Halson 2014 training-load monitoring review in Sports Medicine is the methodological anchor: single markers misfire; the constellation of markers is the load-bearing signal ([18] Halson 2014, Level 5). The [11] Hulin et al. 2016 acute:chronic workload ratio finding has been substantively critiqued — the [12] Impellizzeri et al. 2021 Sports Medicine paper recommends dismissing ACWR ([11] Hulin et al. 2016, Level 2b; [12] Impellizzeri et al. 2021, Level 1b; [13] Lolli et al. 2019, Level 5). The honest read: every published plan works exactly as well as the assumptions it was built on. The day those assumptions break, the rower needs a feedback channel. The AI coach is the same plan with the feedback loop.

The premise: when a static plan is right

A static plan is right when three conditions hold: the goal has a known shape, the week is predictable, and the body is responding as the plan assumes. The [1] The Pete Plan, the [2] Concept2 training plans, and the [3] British Rowing "Go Row Indoor" series are all designed for this case. They ship a calendar of prescribed splits, durations, and rate bands, with the implicit assumption that the rower will turn up on each prescribed day, hold the prescribed split, and arrive fresh.

The four conditions that make a static plan the right tool:

Condition 1 — the goal has a known shape. A 2K test, a first 5K, a season opener. The plan's structure maps cleanly to the goal's demand: a 6×500m interval session trains the aerobic capacity a 2K test requires; a long steady-state row builds the endurance a 5K test requires. When the goal is concrete and the demand is well-understood, the calendar is a defensible map ([1] The Pete Plan, Level 5; [2] Concept2, Level 5).

Condition 2 — the week is predictable. No travel, no work deadlines, no family emergencies. The [23] Leatherwood & Dragoo 2013 airline-travel review in BJSM established the basic mechanism for travel ([23] Leatherwood & Dragoo 2013, Level 5); the [24] Nieman 1994 URTI J-curve paper established the immune-system mechanism ([24] Nieman 1994, Level 5); the [25] Fulco et al. 2000 altitude review established the environmental-divergence mechanism ([25] Fulco et al. 2000, Level 5). When the week is predictable, the plan's prescribed dose is reached; when the week diverges, the prescribed dose is missed, and the plan either over-prescribes or under-delivers.

Condition 3 — the body is responding as the plan assumes. The sRPE on the most recent comparable session reads in the prescribed-load zone; the HR trend is stable; the multi-modal constellation — sleep, mood, soreness — is in the green. The [8] Foster 2001 session-RPE method is the operational anchor: load is sRPE × duration, and the chronic rolling average is what the prescribed dose has to respect ([8] Foster 2001, Level 5).

Condition 4 — the rower fits the plan's demographic. Most published plans are designed for a fresh rower who can train six days a week. The plan's weekly volume assumes the rower has time, energy, and recovery capacity for the prescribed dose. If the rower is older, returning from injury, balancing a job, or managing a family, the plan needs scaling. The [1] The Pete Plan's rotating cycle is a three-week base; the [3] British Rowing "Go Row Indoor" plans explicitly note that adaptation is the rower's responsibility ([3] British Rowing, Level 5). When the plan says "do the prescribed session today" and the rower cannot, the rower's adaptation starts at the deviation.

The boundary: when AI coaching takes over

The AI coach is the same plan with the feedback loop. The [35] Hattie & Timperley 2007 feedback meta-analysis in Review of Educational Research established the empirical anchor: feedback that is calibrated to the learner improves performance, feedback that is not, degrades it ([35] Hattie & Timperley 2007, Level 1a). The [33] Wulf 2007 and [34] Chiviacowsky & Wulf 2002 self-controlled-feedback papers added the user's argument back into the loop: learners who choose when to receive feedback learn more than learners who receive it on a fixed schedule ([33] Wulf 2007, Level 2b; [34] Chiviacowsky & Wulf 2002, Level 2b). The [32] Horn 2008 Advances in Sport Psychology chapter on coaching effectiveness placed the same shape on the operational side: quality feedback is specific, timely, and actionable; absence of any of the three degrades the athlete's learning ([32] Horn 2008, Level 5).

The boundary is the day the four conditions break. When the goal shifts mid-plan, the AI coach re-computes. When the week diverges, the AI coach re-computes. When the body stops responding, the AI coach re-computes. When the rower's demographic does not fit the plan, the AI coach re-computes. The AI coach reads every completed session, recomputes load, and picks the next session from current state. The static plan reads the calendar.

The [30] Smith & Smoll 1990 Mediated Achievement model in Journal of Sport and Exercise Psychology and the [31] Mageau & Vallerand 2003 motivational model in Journal of Sport Sciences both place the coach–athlete relationship on the perception of competence ([30] Smith & Smoll 1990, Level 5; [31] Mageau & Vallerand 2003, Level 5). The static plan's coach–athlete relationship is the calendar; the AI coach's coach–athlete relationship is the chat. The calendar is silent; the chat is live. The rower who can argue with the chat is the rower whose coach–athlete relationship is durable.

Three categories of plan failure

The peer-reviewed literature converges on three concrete failure modes for static plans. These are the failure modes that define the boundary at which AI coaching takes over.

Category 1 — wrong volume. The plan's prescribed weekly volume exceeds the rower's recovery capacity, and the [8] Foster 2001 sRPE × duration chronic rolling average climbs above the prescribed-load zone. The [18] Halson 2014 training-load monitoring review is the methodological anchor: single markers misfire; the constellation of HR trend, sRPE, sleep, mood, and soreness is the load-bearing signal ([18] Halson 2014, Level 5). The [19] Meeusen et al. 2013 ECSS+ACSM consensus in MSSE placed the overtraining continuum on the same footing: a plan that ignores the constellation is a plan that drives functional overreach into non-functional overreach ([19] Meeusen et al. 2013, Level 5).

Category 2 — wrong intensity. The plan's prescribed intensity sits above the rower's current capacity, and the rower's most recent comparable session reads above the prescribed-load zone. The [26] Hagerman 1984 Sports Medicine indoor-rowing physiology review is the rower's capacity anchor: elite male rowers hold a VO₂max of ~6.1 ± 0.6 L/min, and a 2K race draws ~70–75% aerobic and ~25–30% anaerobic ([26] Hagerman 1984, Level 5). The [27] Ingham et al. 2008 MSSE indoor-rower training study is the rower's structure anchor: low- and mixed-intensity rowing programs improved 2K time and VO₂peak similarly, with no between-group difference ([27] Ingham et al. 2008, Level 1b/2b).

Category 3 — wrong timing. The plan's prescribed session timing collides with the rower's environmental divergence — travel, illness, altitude, life stress. The [14] Mujika & Padilla 2000 detraining review in MSSE established the timeline on which the body's response drifts from the plan's assumptions ([14] Mujika & Padilla 2000, Level 5). The [23] Leatherwood & Dragoo 2013 airline-travel review, the [24] Nieman 1994 URTI J-curve paper, and the [25] Fulco et al. 2000 altitude review are the environmental-divergence references ([23] Leatherwood & Dragoo 2013, Level 5; [24] Nieman 1994, Level 5; [25] Fulco et al. 2000, Level 5).

The research on periodisation: the experimental base is mixed

The peer-reviewed periodisation literature is the academic anchor for why static plans are a scaffold rather than a verdict.

The [4] Bompa & Haff 2009 Periodization 5th edition is the canonical reference ([4] Bompa & Haff 2009, Level 5). The textbook established the systematic vocabulary that Western coaching inherited: linear, block, and undulating models share one assumption — the athlete can hit the prescribed dose on the prescribed day. The [5] Issurin 2008 block-periodisation review proposed an alternative that concentrated each adaptation in a block rather than spreading it across the year, but the model still assumes a fixed calendar ([5] Issurin 2008, Level 5).

The [6] Kiely 2018 critical review in Sports Medicine placed the experimental base on more honest footing ([6] Kiely 2018, Level 5). Periodisation's experimental base is thinner than the textbook confidence; realigning periodisation with stress theory offers a more accurate route. The honest read for the indoor rower: the plan is a scaffold, not a verdict. The scaffold performs best when the athlete is fresh and the calendar holds; it underperforms the matched-work alternative when the athlete is tired and the calendar keeps moving.

The [9] Seiler 2010 intensity-distribution review established the policy frame for which variable the plan should change: across competitive endurance sports, about 80% of training is performed at low intensity and about 20% at high intensity ([9] Seiler 2010, Level 5). The [10] Stöggl & Sperlich 2014 Frontiers in Physiology experimental comparison showed that polarized (80/20) beat threshold and high-volume distributions on VO₂peak (+11.7%) and time-to-exhaustion (+17.4%) in well-trained endurance athletes ([10] Stöggl & Sperlich 2014, Level 2b). The implication for the static plan: when something needs to give, give on volume or threshold, not on the high-intensity dose, and re-establish the polarized distribution once the rower's load returns to baseline.

The load-monitoring anchor: what the plan has to respect

The static plan's prescription has to respect the rower's load history. The [7] Banister & Calvert 1980 fitness-fatigue TRIMP decomposition in Canadian Journal of Applied Sport Sciences is the underlying math ([7] Banister & Calvert 1980, Level 5). Each training impulse contributes a positive fitness factor and a competing fatigue factor; the difference between the two is the performance state. The fatigue factor decays faster than the fitness factor, which is why a planned taper — reducing load while preserving fitness — produces a measurable performance gain.

The [8] Foster 2001 session-RPE method in Journal of Strength and Conditioning Research is the operational anchor ([8] Foster 2001, Level 5). Load is sRPE × duration, summed across a rolling window. The 7-day moving average is the acute load; the 28-day moving average is the chronic load. The operation is simple enough to do on the back of an envelope. The [15] Bosquet et al. 2007 MSSE tapering meta-analysis gave the operational numbers: a 41–49% volume taper over 7–14 days maximises performance gains; the gain is larger when the taper is timed to the rower's actual readiness, not the original race calendar ([15] Bosquet et al. 2007, Level 1a).

The [11] Hulin et al. 2016 acute:chronic workload ratio finding in BJSM proposed a safety anchor — a 7-day planned progressive overload is the safe route to a fitness gain ([11] Hulin et al. 2016, Level 2b). The [36] Gabbett 2016 "training–injury prevention paradox" paper in BJSM placed the finding on the policy frame: rapid training increases raise injury risk; planned progressive overload is the safe route ([36] Gabbett 2016, Level 5). The finding has been substantively critiqued since publication. The [12] Impellizzeri et al. 2021 Sports Medicine paper is the sharpest version of that critique ([12] Impellizzeri et al. 2021, Level 1b). The acute:chronic workload ratio is essentially a rescaling of acute load — it magnifies effect estimates and reduces variance without adding predictive value. The recommendation in the abstract is explicit: dismiss ACWR as a framework and a model. The [13] Lolli et al. 2019 BJSM editorial reached the same conclusion from a different direction: the conventional ACWR formula is mathematical coupling that produces spurious correlation, an inaccurate scaling index for an unnecessary normalisation process ([13] Lolli et al. 2019, Level 5).

The honest read for the indoor rower: the 0.8–1.3 sweet spot and the 1.5 spike zone are not load-bearing biological thresholds — they are statistical artefacts of viewing the same load data two ways. The safer rule that survives the critique is the [7] Banister & Calvert 1980 fitness-fatigue decomposition plus the [8] Foster 2001 session-RPE method: track the acute and chronic moving averages as separate signals, do not couple them into a ratio, and respect the [36] Gabbett 2016 progressive-overload policy even when the ratio says the load is still in the sweet spot.

The individualisation argument: HRV-guided beats fixed at matched total work

The peer-reviewed adaptive-prescription literature is the strongest empirical argument for AI coaching over static plans — at matched total work, adaptive beats fixed.

The [16] Vesterinen et al. 2016 MSSE paper is the closest field-trial analogue to what an AI coach does ([16] Vesterinen et al. 2016, Level 1b/2b). Endurance athletes were randomised to either a traditional periodised plan (TRAD) or to an experimental arm where each day's session intensity was chosen based on the previous morning's HRV read (EXP). EXP completed significantly fewer moderate and hard sessions (13.2 ± 6.0) than TRAD (17.7 ± 2.5). Yet 3000-m time improved significantly in EXP (+2.1% ± 2.0%) and not significantly in TRAD (+1.1% ± 2.7%). The same total work, more recovery, better outcome.

The [17] Kiviniemi et al. 2007 European Journal of Applied Physiology paper is the prior field trial ([17] Kiviniemi et al. 2007, Level 2b). Moderately fit adults trained with either daily HRV-guided prescription or a traditional plan; the HRV-guided group improved more in Load(max) than the traditional group. The honest read for the indoor rower: the principle survives the sport shift; the AI coach that reads the rower's sRPE × duration and recomputes load is the AI coach that bridges the gap between static and adaptive.

The [18] Halson 2014 training-load monitoring review in Sports Medicine is the methodological complement ([18] Halson 2014, Level 5). Single markers misfire; the constellation of HR trend, sRPE, sleep, mood, and soreness is the load-bearing signal. The AI coach that reads the constellation is the AI coach that adapts the plan to the rower's state. The static plan that ignores the constellation is the plan that over-prescribes in the second week of any intensification block.

When a static plan is the wrong tool

The static plan is the wrong tool when any one of the four conditions breaks. The peer-reviewed literature gives concrete thresholds.

When the goal shifts mid-plan. The plan's prescribed dose no longer maps cleanly to the goal's demand. The [14] Mujika & Padilla 2000 detraining review established the timeline: a one-week layoff produces measurable losses in plasma volume and glycogen ([14] Mujika & Padilla 2000, Level 5). A goal shift that requires a two-week layoff costs the rower days to weeks of fitness; a plan that resumes prescribed dose on day one of the new goal over-prescribes by design.

When the week is not predictable. Travel, work, family. The [23] Leatherwood & Dragoo 2013 airline-travel review established the basic mechanism: eastward or westward travel disrupts circadian rhythm, sleep, hydration, and nutrition; each effect alone degrades performance for several days post-flight ([23] Leatherwood & Dragoo 2013, Level 5). The [24] Nieman 1994 URTI J-curve paper established the immune-system complement: URTI risk is elevated in the 1–2 weeks following a heavy training event ([24] Nieman 1994, Level 5). The static plan that resumes prescribed dose on day two of a seven-hour time-zone shift is over-prescribing on a circadian-disrupted body.

When the body is not responding. The sRPE on the rower's most recent comparable session reads above the prescribed-load zone; the HR trend is climbing; the multi-modal constellation is in the yellow or red. The [18] Halson 2014 training-load monitoring review is the methodological anchor: single markers misfire; the constellation of HR trend, sRPE, sleep, mood, and soreness is the load-bearing signal ([18] Halson 2014, Level 5). The [19] Meeusen et al. 2013 ECSS+ACSM consensus placed the overtraining continuum on the same footing: a coach that prescribes onto unexplained underperformance is asking for non-functional overreach ([19] Meeusen et al. 2013, Level 5).

When the rower does not fit the plan's demographic. The plan's weekly volume assumes the rower has time, energy, and recovery capacity for the prescribed dose. If the rower is older, returning from injury, balancing a job, or managing a family, the plan needs scaling. The plan's coach–athlete relationship is the calendar; the rower who cannot hit the calendar's prescribed dose is the rower whose adaptation starts at the deviation. The [28] Bruinvels et al. 2017 BJSM editorial and the [29] Scott et al. 2024 IJSPP field study placed the female-athlete gap on the empirical side: sessional RPE responses to total distances ≥ 5 km were higher during the luteal phase than during menstruation ([28] Bruinvels et al. 2017, Level 5; [29] Scott et al. 2024, Level 2b). The static plan is cycle-blind by default.

The static plan as input, not answer

The honest read for the indoor rower: the static plan is the input, not the answer. The [4] Bompa & Haff 2009 Periodization textbook and the [5] Issurin 2008 review codify the structure; the [6] Kiely 2018 critical review places the experimental base on more honest footing ([4] Bompa & Haff 2009, Level 5; [5] Issurin 2008, Level 5; [6] Kiely 2018, Level 5). The [1] The Pete Plan and the [3] British Rowing "Go Row Indoor" plans are the conservative defaults for a rower whose state is on the plan's assumptions ([1] The Pete Plan, Level 5; [3] British Rowing, Level 5).

The [7] Banister & Calvert 1980 fitness-fatigue decomposition and the [8] Foster 2001 session-RPE method are the operational load anchors ([7] Banister & Calvert 1980, Level 5; [8] Foster 2001, Level 5). The [15] Bosquet 2007 tapering meta-analysis is the taper anchor ([15] Bosquet et al. 2007, Level 1a). The [16] Vesterinen 2016 and [17] Kiviniemi 2007 HRV-guided field trials are the adaptive-prescription anchors ([16] Vesterinen et al. 2016, Level 1b/2b; [17] Kiviniemi et al. 2007, Level 2b). The [26] Hagerman 1984 Sports Medicine indoor-rowing physiology review and the [27] Ingham et al. 2008 MSSE indoor-rower training study are the rowing-specific anchors ([26] Hagerman 1984, Level 5; [27] Ingham et al. 2008, Level 1b/2b). The [20] Sawka et al. 2007 ACSM position stand on fluid replacement, the [21] Burke et al. 2011 carbohydrate-intake review, and the [22] Phillips & Van Loon 2011 protein review are the substrate anchors ([20] Sawka et al. 2007, Level 5; [21] Burke et al. 2011, Level 5; [22] Phillips & Van Loon 2011, Level 5).

The right posture is to use a static plan as the input, and the rower's body as the variable, and the AI coach as the feedback channel. The plan is the structure; the body is the state; the AI coach is the recompute. The day the four conditions break, the rower needs the feedback channel. The feedback channel is the chat. The rower who can argue with the chat is the rower whose coach–athlete relationship is durable.

Three concrete scenarios

The peer-reviewed literature converges on three concrete scenarios in which a static plan is the right tool — and three in which it is the wrong one.

Scenario A (right tool) — a fresh rower training for a 2K test, six weeks out. The goal has a known shape (a 2K test). The week is predictable (the rower is not travelling, not changing jobs, not managing a family illness). The body is responding (sRPE on the most recent comparable session reads in the prescribed-load zone). The rower fits the plan's demographic (six training days a week, time and energy for the prescribed dose). The static plan is the right tool. The [1] The Pete Plan's rotating cycle is a defensible structure; the [2] Concept2 training plans are a defensible structure; the [3] British Rowing "Go Row Indoor" plans are a defensible structure.

Scenario B (right tool transitioning to wrong tool) — six weeks in, the rower lands a work deadline that consumes two days. The goal still has a known shape, the body still responds, but the week has diverged. The static plan is the wrong tool for the remaining two days of the week. The [23] Leatherwood & Dragoo 2013 airline-travel review's mechanism generalises to work-deadline stress: environmental divergence degrades capacity for several days post-event ([23] Leatherwood & Dragoo 2013, Level 5). The AI coach re-computes the prescribed dose for the rower's current window; the static plan cannot.

Scenario C (wrong tool from the start) — a 50-year-old rower returning from injury, training for a first 5K. The goal has a known shape (a first 5K), but the demographic does not fit the plan's assumptions (the [1] The Pete Plan's rotating cycle assumes a fresh rower who can train six days a week; the [3] British Rowing "Go Row Indoor" plans are designed for beginners but assume time and energy for the prescribed dose). The [14] Mujika & Padilla 2000 detraining review's timeline applies: a return from injury is a return from a layoff, and the chronic training load is depressed ([14] Mujika & Padilla 2000, Level 5). The static plan's prescribed dose over-prescribes on day one. The AI coach ramps load across three to five days before resuming prescribed dose; the static plan does not.

Practical rules for using a static plan

The peer-reviewed literature converges on a small set of operational rules for using a static plan as the input to the rower's training.

Rule 1 — verify the plan's demographic fits the rower. Most published plans assume a fresh rower who can train six days a week. If the rower is older, returning from injury, balancing a job, or managing a family, the plan needs scaling. The [3] British Rowing "Go Row Indoor" plans are explicit about adaptation ([3] British Rowing, Level 5); the [1] The Pete Plan is less explicit ([1] The Pete Plan, Level 5).

Rule 2 — track sRPE × duration every session. The [8] Foster 2001 method is cheap (one number after each session) and gives the operational load signal ([8] Foster 2001, Level 5). Two rolling windows (7-day acute, 28-day chronic) are the [7] Banister & Calvert 1980 TRIMP decomposition's policy frame ([7] Banister & Calvert 1980, Level 5). Do not collapse the two into the ACWR — the [12] Impellizzeri et al. 2021 and [13] Lolli et al. 2019 critiques are unambiguous on this point ([12] Impellizzeri et al. 2021, Level 1b; [13] Lolli et al. 2019, Level 5).

Rule 3 — plan the taper to readiness, not the calendar. The [15] Bosquet et al. 2007 MSSE meta-analysis gave the operational numbers: a 41–49% volume taper over 7–14 days maximises gains; the gain is larger when the taper is timed to the rower's actual readiness, not the original race calendar ([15] Bosquet et al. 2007, Level 1a).

Rule 4 — log environmental divergence. Travel, illness, altitude, life stress. The [23] Leatherwood & Dragoo 2013 airline-travel review, the [24] Nieman 1994 URTI J-curve paper, and the [25] Fulco et al. 2000 altitude review are the environmental-divergence references ([23] Leatherwood & Dragoo 2013, Level 5; [24] Nieman 1994, Level 5; [25] Fulco et al. 2000, Level 5). The static plan does not see the divergence; the AI coach does.

Rule 5 — when in doubt, use the static plan as the baseline. The [1] The Pete Plan, the [2] Concept2 training plans, and the [3] British Rowing "Go Row Indoor" plans are the conservative defaults for a rower whose state is on the plan's assumptions ([1] The Pete Plan, Level 5; [2] Concept2, Level 5; [3] British Rowing, Level 5). When the rower's state is ambiguous, use the plan; when the rower's state diverges, recompute.

Rule 6 — use the multi-modal signal even without HRV. The [16] Vesterinen et al. 2016 MSSE trial and the [17] Kiviniemi et al. 2007 European Journal of Applied Physiology trial give the field-trial support for HRV-guided prescription ([16] Vesterinen et al. 2016, Level 1b/2b; [17] Kiviniemi et al. 2007, Level 2b). The principle survives the signal choice: read the body, not the calendar — sRPE, sleep, mood, soreness are the cheap proxies.

Limitations and open questions

The static plan's empirical base is mixed. The [6] Kiely 2018 critical review in Sports Medicine placed the experimental base on more honest footing ([6] Kiely 2018, Level 5). The [4] Bompa & Haff 2009 textbook and the [5] Issurin 2008 review codify the structure, even when the empirical case for periodisation is weaker than the textbook confidence ([4] Bompa & Haff 2009, Level 5; [5] Issurin 2008, Level 5). The honest read for the rower: the plan is a scaffold, not a verdict; the scaffold performs best when the rower is fresh and the calendar holds, and underperforms the matched-work alternative when the rower is tired and the calendar keeps moving.

The ACWR empirical base has been substantively critiqued. The [12] Impellizzeri et al. 2021 dismissal and the [13] Lolli et al. 2019 mathematical-coupling critique are now the published consensus in the modelling literature; the [11] Hulin 2016 original is the case the field is moving past ([11] Hulin et al. 2016, Level 2b; [12] Impellizzeri et al. 2021, Level 1b; [13] Lolli et al. 2019, Level 5). The 0.8–1.3 sweet spot and 1.5 spike zone remain useful as shorthand in coaching education, but should not be the policy lever.

Transferability from elite sport to indoor rowing is by analogy, not direct measurement. The [11] Hulin 2016 finding is in elite rugby league; the [16] Vesterinen 2016 and [17] Kiviniemi 2007 trials are in mixed-discipline endurance athletes; the [26] Hagerman 1984 physiology review is over 40 years old. The indoor-rowing-specific physiology is in the [26] Hagerman reference; the indoor-rowing-specific training study is the [27] Ingham et al. 2008 trial, which found no between-group difference between low- and mixed-intensity rowing programs ([27] Ingham et al. 2008, Level 1b/2b). The indoor-rowing-specific taper evidence is borrowed from the [15] Bosquet 2007 meta-analysis and the broader tapered-on-the-rower literature.

The female-athlete evidence base is still thin. The [28] Bruinvels 2017 editorial is a research-gap call, not a body of evidence. The [29] Scott et al. 2024 IJSPP paper is the first quantitative field-study anchor for menstrual-cycle-related sRPE shifts in elite female athletes — one paper, one sport (soccer), and one population (FIFA Women's World Cup) ([29] Scott et al. 2024, Level 2b). The indoor-rowing-specific application is by analogy, not measurement. The honest read for the rower: track cycle phase as one input among several; treat the [29] Scott et al. 2024 finding as a starting hypothesis, not a calibration.

The AI-coaching literature is new. Peer-reviewed evidence for AI-driven session-by-session adaptation in indoor rowing is in early stages. The [16] Vesterinen 2016 and [17] Kiviniemi 2007 trials are the closest published analogues — both use HRV, not sRPE, but the operational principle is the same. The honest read for the rower: the AI coach's recommendation is an input; the body's response is the next input; the policy is to use both.

The summary in one paragraph

A published plan — the Pete Plan ([1] The Pete Plan, Level 5), the Concept2 training plans ([2] Concept2, Level 5), the British Rowing "Go Row Indoor" plans ([3] British Rowing, Level 5) — is the right tool when the goal has a known shape, the week is predictable, the body is responding, and the rower fits the plan's demographic. The [4] Bompa & Haff 2009 Periodization textbook and the [5] Issurin 2008 review codify the structure ([4] Bompa & Haff 2009, Level 5; [5] Issurin 2008, Level 5). The [6] Kiely 2018 critical review and the [27] Ingham 2008 indoor-rower training study frame the experimental base on more honest footing ([6] Kiely 2018, Level 5; [27] Ingham 2008, Level 1b/2b). The [7] Banister & Calvert 1980 TRIMP decomposition and the [8] Foster 2001 session-RPE method are the operational load anchors ([7] Banister & Calvert 1980, Level 5; [8] Foster 2001, Level 5). The [9] Seiler 2010 intensity-distribution review and the [10] Stöggl & Sperlich 2014 polarized-training RCT are the 80/20 distribution anchors ([9] Seiler 2010, Level 5; [10] Stöggl & Sperlich 2014, Level 2b). The [11] Hulin 2016 ACWR finding has been substantively critiqued ([12] Impellizzeri 2021, Level 1b; [13] Lolli 2019, Level 5). The [14] Mujika & Padilla 2000 detraining review, the [15] Bosquet 2007 tapering meta-analysis, the [16] Vesterinen 2016 and [17] Kiviniemi 2007 HRV-guided field trials are the adaptive-prescription anchors ([14] Mujika & Padilla 2000, Level 5; [15] Bosquet et al. 2007, Level 1a; [16] Vesterinen et al. 2016, Level 1b/2b; [17] Kiviniemi et al. 2007, Level 2b). The [18] Halson 2014 training-load monitoring review and the [19] Meeusen et al. 2013 ECSS+ACSM consensus are the overtraining anchors ([18] Halson 2014, Level 5; [19] Meeusen et al. 2013, Level 5). The [20] Sawka 2007, [21] Burke 2011, and [22] Phillips 2011 reviews are the substrate anchors. The [23] Leatherwood 2013 airline-travel review, the [24] Nieman 1994 URTI J-curve paper, and the [25] Fulco 2000 altitude review are the environmental-divergence references ([23] Leatherwood & Dragoo 2013, Level 5; [24] Nieman 1994, Level 5; [25] Fulco et al. 2000, Level 5). The [26] Hagerman 1984 indoor-rowing physiology review and the [27] Ingham 2008 indoor-rower training study are the rowing-specific anchors ([26] Hagerman 1984, Level 5; [27] Ingham et al. 2008, Level 1b/2b). The [28] Bruinvels 2017 research-gap editorial and the [29] Scott 2024 FIFA Women's World Cup RPE study are the female-athlete literature anchors ([28] Bruinvels et al. 2017, Level 5; [29] Scott et al. 2024, Level 2b). The [30] Smith & Smoll 1990 Mediated Achievement model, the [31] Mageau & Vallerand 2003 motivational model, and the [32] Horn 2008 coaching-effectiveness framework are the coach–athlete relationship anchors. The [33] Wulf 2007 and [34] Chiviacowsky & Wulf 2002 self-controlled-feedback papers, and the [35] Hattie & Timperley 2007 feedback meta-analysis, are the feedback-channel anchors ([33] Wulf 2007, Level 2b; [34] Chiviacowsky & Wulf 2002, Level 2b; [35] Hattie & Timperley 2007, Level 1a). The [36] Gabbett 2016 "training–injury prevention paradox" paper is the progressive-overload anchor ([36] Gabbett 2016, Level 5).

The right posture is to use a static plan as the input, the rower's body as the variable, and the AI coach as the feedback channel. The plan is the structure; the body is the state; the AI coach is the recompute. The day the four conditions break, the rower needs the feedback channel. The feedback channel is the chat.

For a deeper exploration of how MyNextRow's AI coach uses load governors to adapt each session, see our AI coaching load governors plain-English guide.

What to do with this article

Read the principle: a static plan is the right tool when the goal has a known shape, the week is predictable, the body is responding, and the rower fits the plan's demographic. The day any one breaks, the AI coach is the feedback channel.

Read the research: the [4] Bompa & Haff 2009 textbook and the [5] Issurin 2008 review codify the structure; the [6] Kiely 2018 critical review and the [27] Ingham 2008 indoor-rower training study frame the experimental base on more honest footing; the [7] Banister & Calvert 1980 TRIMP decomposition and the [8] Foster 2001 session-RPE method are the operational load anchors; the [9] Seiler 2010 and [10] Stöggl & Sperlich 2014 80/20 distribution anchors; the [11] Hulin 2016 ACWR critique via the [12] Impellizzeri 2021 / [13] Lolli 2019 dismissals; the [14] Mujika 2000 detraining review, the [15] Bosquet 2007 tapering meta-analysis, and the [16] Vesterinen 2016 / [17] Kiviniemi 2007 HRV-guided field trials anchor the adaptive-prescription side; the [18] Halson 2014 training-load monitoring review and the [19] Meeusen 2013 ECSS+ACSM consensus anchor the overtraining diagnosis; the [20] Sawka 2007, [21] Burke 2011, and [22] Phillips 2011 reviews anchor the substrate side; the [23] Leatherwood 2013 airline-travel review, the [24] Nieman 1994 URTI J-curve paper, and the [25] Fulco 2000 altitude review anchor the environmental-divergence failure modes; the [26] Hagerman 1984 indoor-rowing physiology review and the [27] Ingham 2008 indoor-rower training study anchor the rowing-specific bounds; the [28] Bruinvels 2017 and [29] Scott 2024 female-athlete literature; the [30] Smith & Smoll 1990 Mediated Achievement model and the [31] Mageau & Vallerand 2003 motivational model and the [32] Horn 2008 coaching-effectiveness framework anchor the coach–athlete relationship side; the [33] Wulf 2007 and [34] Chiviacowsky & Wulf 2002 self-controlled-feedback papers and the [35] Hattie & Timperley 2007 feedback meta-analysis anchor the feedback-channel side; the [36] Gabbett 2016 progressive-overload anchor.

Read the practical read: verify the demographic; track sRPE × duration; plan the taper to readiness; log environmental divergence; use the plan as baseline when state is ambiguous; use the multi-modal signal even without HRV.

When the four conditions hold, use the plan. When any one breaks, the AI coach is the feedback channel. The plan is the input. The body is the variable. The next session is the answer.