Strength Circuits, Lower Impact, Higher Purpose

Abstract

A rowing piece is limited by a specific set of physiological variables, and resistance training is the only thing in most plans that moves them. "Lower impact" is what the circuit does not cost you; the force profile is what it buys.

The honest justification

The usual arguments for strength work in a rowing plan are injury prevention and muscle mass. Both are real. Neither is what the ergometer is missing, and both are hard to measure, which makes them poor reasons to defend a session in a programme meeting.

The measurable reason is narrower. Bourdin's study of 54 highly trained rowers tested whether peak power output from incremental testing predicts ergometer performance over 2,000 m, and found it to be an overall index of it ([2] Bourdin 2004). That is a significant statement: a variable derived from a different kind of test — maximal, incremental, anaerobic — carries information about the seven-minute piece that the piece itself does not fully express.

This is the entire argument for the circuit. It is not that strength training is good in general. It is that peak power is a limiting variable for the event, and it is the one variable your plan is otherwise not touching.

The aerobic ceiling is not in dispute: metabolism supplies 70-75% of the energy over the standard 2000 m ([1] Hagerman 1984). A circuit cannot raise that ceiling and should not be sold as if it does. It works underneath it.

What a rowing-specific circuit should actually train

Toigo's framing is the most useful one available, because it argues that muscle does not respond to "resistance training" as a category but to specific loading conditions that trigger distinct molecular events ([3] Toigo 2006). Under that framing, a generic circuit of eight exercises is a collection of stimuli chosen at random, and the question worth asking of each movement is which one it produces.

Force production across the range that matters. The drive phase of the stroke is where power is made, and it happens at a joint angle and a velocity that no machine-based isolation movement replicates. Movements that train the legs through a range, under load, at a controlled velocity address this directly.

Rate of force development rather than maximum force. The distinction matters because maximum force and the speed at which force can be raised are different adaptations with different mechanisms. A circuit built entirely from slow, heavy, grinding repetitions trains the first and largely ignores the second.

Trunk and handle-side rigidity. This is the least glamorous and most rowing-specific item. The force couple that transmits leg drive through the trunk to the handle is the mechanical link, and it is trainable.

Unilateral capacity. A rowing stroke is one-sided by definition. Bilateral-only strength work leaves an asymmetry that the ergometer will then reproduce in every piece.

Kraemer's ordering principles apply directly to the first of these: large muscle groups before small, multi-joint before single-joint, higher intensity before lower, because the quality of the exercise degrades as fatigue accumulates ([8] Kraemer 2002). The 2009 ACSM position stand restates the progression model with the loading and repetition guidance attached ([9] American College 2009).

The dose-response, and the myth about failure

Wernbom's review is the quantitative anchor, computing the magnitudes and rates of cross-sectional area increase across frequency, intensity and volume rather than describing them qualitatively ([5] Wernbom 2007). The practical finding from that literature is that the volume–response relationship flattens: adding sets past a point buys progressively less, and the ordering of the variables matters less than continuing at all.

The specific myth worth dismantling is proximity to failure. It is the most repeated piece of strength advice in popular coaching, and the meta-analysis does not support the strong version of it. Refalo's synthesis of 15 studies found the effect of training to failure on hypertrophy to be small ([6] Refalo 2023). Set failure — which includes anything short of momentary failure — behaves similarly.

Schoenfeld's review is worth reading alongside it, because it explains why the advice persists: bodybuilders and powerlifters both display impressive muscularity while training on opposite loading strategies, which makes the question look empirically open when the mechanistic accounting is not ([4] Schoenfeld 2010). The variable that distinguishes the two approaches is not the one most coaches argue about.

The practical reading for a rowing circuit is that you do not need to grind the last repetition. That matters twice over, because a circuit that trains to failure leaves you unavailable for the ergometer the next morning, which is the interference problem discussed below arriving early.

Concurrent training: a sequencing problem before an interference problem

Leveritt's review is the standard reference, and its framing is more useful than the usual summary. Concurrent strength and endurance training appears to inhibit strength development compared with strength training alone, but the underlying evidence is hard to interpret because studies differ in mode, frequency, duration, intensity, training history, session scheduling, and which outcome they measure ([7] Leveritt 1999).

That list is the point. Most of what gets called concurrent-training interference is really a scheduling artefact — two demanding sessions too close together, competing for the same recovery capacity, on a calendar that was not designed for two modalities.

The recovery cost is the thing being competed for, and it is a real account. Banister's impulse model framed performance as the balance between fitness and fatigue that training engenders rather than as a function of load alone ([17] Banister 1980), which treats every added session as a withdrawal from that balance. Halson's survey is blunt about the limit: very few available load markers have strong scientific evidence behind them ([14] Halson 2014).

Three practical consequences follow, none of which require a laboratory.

Separate the demanding sessions by at least a day. The most common error is pairing the heaviest strength session with the hardest aerobic session. Nothing about that pairing is required, and it is the version most likely to produce the interference everyone warns about.

Keep the strength session short and the intensity high. This follows from the proximity-to-failure finding: you can get the stimulus without the fatigue cost, and the fatigue cost is what interferes.

Reduce strength volume inside a heavy aerobic block rather than deleting it. The periodization framework and its critique both apply here ([23] Issurin 2010; [24] Kiely 2018). The aerobic work is what the plan is for; the circuit is a supplement to it, and a supplement can be dosed down without being removed.

The competing consideration is recovery. Banister's impulse model framed performance as the balance between fitness and fatigue that training engenders rather than as a function of load alone ([17] Banister 1980), which is a framing that treats every added session as a withdrawal from some account. Halson's survey is blunt about the limit: very few available load markers have strong scientific evidence behind them ([14] Halson 2014).

Knowing when to stop the circuit

A circuit that continues through a taper is a circuit that is costing you race performance. Bosquet's taper meta-analysis gives the reference for how much load to remove and for how long ([19] Bosquet 2007), and the practical rule is simpler than it sounds: the circuit stops when the aerobic work becomes the priority, because that is the priority.

The other stopping condition is a schedule. Mujika's detraining review shows that trained athletes lose adaptations faster than most people expect — VO2max and blood volume decline rapidly within four weeks of insufficient stimulus ([18] Mujika 2000). A circuit is an adaptation too, and it will not wait patiently for the racing season.

It also has a recovery bill that the plan is already paying in part. Nieman's exercise-immune relationship is the reason a week with an extra demanding session carries an infection cost that a week without one does not ([27] Nieman 1994), and the carbohydrate and protein frameworks together describe what the added session is spent on — resynthesis, not adaptation ([25] Burke 2011; [26] Phillips 2011).

Meeusen's distinction between functional and non-functional overreaching is the right lens for the whole question ([13] Meeusen 2013). Functional overreaching is a short-term performance dip that resolves into a gain. Non-functional overreaching is the same presentation without the gain, and the two are indistinguishable without monitoring — which is why the monitoring matters more for a second modality than for one.

Monitoring a circuit without a laboratory

You do not need a research setup to know whether the circuit is working. You need three things, and all three are cheap.

Bar velocity on the compound lifts. Foster's velocity-based monitoring work is what makes resistance progression measurable rather than assumed ([15] Foster 2001). If the same load is moving faster in week six than in week one, something is adapting. If it is moving slower at the same load, you are accumulating fatigue, not fitness.

Effort, not load, as the primary variable. The Borg scale is the original work on perceptual effort and, importantly, the reason category scales work for comparing people ([10] Borg 1982); Scherr's review covers how well it tracks physiological variables and where the correspondence weakens ([11] Scherr 2013). Reed's talk test is the cheapest credible proxy, correlating well with ventilatory and lactate measures ([12] Reed 2014). The practical version: if the circuit leaves you unable to produce a hard word during it, the prescription is wrong regardless of what the bar says.

Heart rate and recovery, interpreted carefully. Buchheit's argument is that most contradictory findings in the heart-rate literature are measurement inconsistency rather than physiology, and are substantially resolvable by standardising how the measures are taken ([16] Buchheit 2014). Standardise before concluding anything.

The six-week check

Six weeks is long enough to tell whether a circuit is producing an adaptation, and short enough that a bad one is not a season-long mistake. Three things to look at.

The ergometer, not the gym. Ingham's determinant work is the reason to look here: power at VO2max, maximal power and maximal force correlate with 2,000 m time at r = 0.95 ([20] Ingham 2008). If the circuit is doing what Bourdin's work implies it should ([2] Bourdin 2004), that correlation is where it shows up first.

Consistency under load. A circuit that leaves you weaker week to week at the same weight is not producing an adaptation. This is the cheapest and most reliable signal available, and it is the one most often ignored because it requires no equipment and no resting heart rate.

The aerobic side did not drop. Ingham's 12-week study is the relevant caution: low-intensity and mixed programs improved rowing performance and VO2peak by the same magnitude ([20] Ingham 2008). Adding a modality should not subtract from the aerobic work. If VO2max or your ergometer times move backwards while the circuit is running, the separation is not working, and the fix is scheduling rather than more effort.

The distribution frame matters here too. Seiler's convergence on roughly 80% of sessions at low intensity is the standard the circuit has to fit inside, not compete with ([21] Seiler 2010). A week with three hard ergometer pieces and two circuits is not a polarized week, whatever the plan says.

Safety and the junior caveat

One thing has to be said plainly. Lloyd's International Consensus is the reference for youth resistance training, and its central requirement is proper technique ([22] Lloyd 2014). A circuit is a place where poor technique is most expensive and least visible, because the athlete is often working at a load that hides a movement fault until it does not.

For a junior, supervised and technique-first is not a cautious compromise. It is the entire content of the recommendation.

What to do with this article

Build the circuit around the force profile of the stroke rather than around a list of exercises, keep it short and intense, separate it from the hard aerobic work by a day, and check it against the ergometer in week six rather than against the mirror in week one.

The circuit is worth its time because peak power limits the piece. Measure it there.