Posture on the Rower: Tall, Supported, and Mobile

If someone films you from the side and shows you a single frame from the middle of the drive, they can tell you that you are sitting well. That frame is uninformative, and it is uninformative in a specific way: the trunk passes through a materially different position at the catch, at the end of the drive and through the recovery, and it passes through a different set of positions in the last 500 m than in the first.

The context

Posture advice usually has two shapes. One is "sit tall, shoulders back and down", aimed at a visibly collapsing rower. The other is "keep the back neutral", which sounds more technical and is usually worse, because it invites you to brace the trunk and hold it.

Both describe a photograph. Neither survives contact with a stroke, where the spine flexes and extends several times a minute for as long as the piece lasts — and neither survives contact with the research, which is specific about what changes.

Caldwell and colleagues instrumented sixteen school rowers and measured lumbar flexion from markers on L1 and S1, together with electromyography from three erector spinae muscles, at three stages across a rowing trial. Lumbar flexion increased significantly over the course of the trial. So did the magnitude of erector spinae activity — while the median frequency of that signal decreased in every muscle examined, which is the signature of fatigue. The authors concluded that rowers reach relatively high levels of lumbar flexion during the stroke, that these levels increase as the piece goes on, and that erector spinae fatigue may itself be part of why the flexion increases [2]. Reid and McNair had identified risk factors for low back pain in rowers well before any of this was instrumented [3], and Cady-McCrea and colleagues' recent clinical review covers the biomechanics, injury and treatment picture that all of this sits inside [4].

Nugent and colleagues pooled 22 studies and 429 participants into a systematic review, and their picture is consistent with that. Rowers without a history of low back pain showed neutral or anterior pelvic rotation at the catch, greater hip range of motion, and a flatter low back at the finish. Rowers with low back pain showed greater posterior pelvic rotation at the catch and greater hip extension at the finish. In both groups, fatigue produced increased lumbar flexion at the catch — and the effect was greater on the ergometer than on the water [1].

That last clause is the one an indoor-rowing article should lead with. The machine the reader is sitting on shows the effect more strongly than the boat does, and the review found insufficient evidence to recommend one ergometer type over the other for avoiding low back pain [1].

The core idea

Posture on the rower is not a position. It is the behaviour of a column that changes shape — and, on this evidence, a column that changes shape predictably as you tire.

Three things have to stay alive at once.

The spine has to be long rather than locked. The useful sensation is length — sitting tall through the crown of the head, ribs stacked over the pelvis — rather than rigidity. Soper and Hume describe an ideal technique in which the trunk's job is to arrive at the right angle on time [5], which is a timing job, not a posture job. Secher's physiology review frames the same thing as a single coordinated sequence through legs, trunk and arms, in which no single segment can be optimised on its own [7]. A spine held rigid through the drive cannot absorb the difference between what the legs are producing and what the arms have not yet taken, and that load gets redistributed somewhere less able to carry it.

The pelvis has to be supported, not held. On the rower the pelvis is carried by the seat for the whole stroke, so it needs no muscular support, and asking for any tends to tuck it under. This is not a style preference — pelvic rotation at the catch is one of the few things that separates rowers with a history of low back pain from those without [1].

The ribcage has to keep moving. This is the one people most often lose, and it fails for a different reason than the other two. Posture cues of the "open the chest" variety tend to be delivered at a moment when the ribcage needs to expand, and holding the chest open is a strong way to stop it doing so. Yamashita and colleagues recorded trunk muscle activity during ergometer rowing and found it differed between rowers with and without low back pain [6]. Breathing is not decoration; if the ribcage cannot move, the breath is shallow and the rower compensates by gripping.

This is also why "brace harder" is the wrong response to most posture discomfort. Bracing reduces movement, and reduced movement is the mechanism — but note that the erector spinae need to be doing something. The Caldwell data is not an argument for switching the back off; it is an argument that those muscles are load-bearing and they tire [2]. What you are watching for is their fatigue, not their absence.

The how-to

Run this on an easy piece, at a rate you would describe as too slow — 14 to 16 — so the transitions are separable.

1. Find length before you find uprightness. At the catch, before the drive, feel the crown of your head lifting away from your ribs. That is length. If you are also pinching your shoulder blades, you have found rigidity instead and lost the mobility you were aiming for. Both the manufacturer's technique material and the federation coaching material describe the trunk in motion through a sequence rather than at rest [9] [10].

2. Let the pelvis be heavy. Sit into the seat rather than lifting out of it. The tell is that your lower back settles rather than holding itself in place. A forward lean at the catch is correct and it is supposed to be there; what you are managing is not the angle but the rotation.

3. Watch the finish, once, at the start. The systematic review found a flatter low back at the finish in rowers without low back pain, and more hip extension there in rowers with it [1]. This is a single look at the start of a piece, not something to monitor throughout — it tells you which of the two you default to.

4. Breathe into the recovery. It is the long, quiet part of the stroke and it is where your ribcage should visibly expand. If you cannot feel your ribs move there, you are bracing, and you will feel it in the first few minutes of any piece.

5. Treat late-sequence change as information. Because flexion rises across a trial in step with erector spinae fatigue [2], a back that has quietly rounded by the last 500 m is reporting that the piece has outlasted your trunk. The response is to read it — shorten the piece, or lower the rate — rather than to brace harder and push on. A hard rower racing a 2k would be well advised to know that the ergometer is where this shows up most [1].

6. Do not expect the monitor to describe any of it. Hagerman's applied physiology review is largely about what the ergometer can and cannot tell you about a rower: power and rate are instantaneous, heart rate lags behind them, and none of it measures the shape you are holding [8]. If trunk position is what you are working on, it needs eyes or a coach, not a screen.

The trade-offs

Two honest limits.

The first is that this is an article about the ergometer, and the ergometer is not quite a boat. On the rower the pelvis is supported continuously by a seat, which makes a supported trunk easier than it is on the water. The systematic review is direct about this too: the fatigue-related increase in catch flexion is greater on the ergometer [1], so a rower reading their own back on a machine is seeing the exaggerated version of the thing.

The second is that the mechanism is well described and the prescription is not. Caldwell's sample was sixteen school rowers, and the review it sits in pools studies of varying quality — a mean modified Quality Index score of 16.7 out of 28 across the included set [1]. What is well supported is that lumbar flexion is fatigue-dependent and that its pattern differs between rowers with and without low back pain. What is not established is a safe angle, a safe number of strokes, or a safe number of minutes. Anyone who tells you there is a threshold has oversold it.

What to do tomorrow

Take one piece and drop the rate to something you would not normally choose. Row twenty strokes and watch three things: whether the crown of your head stays long at the catch, whether your lower back settles into the seat rather than holding itself up, and whether you can feel your ribs expand on the way back.

If the length is there but the ribs are not moving, breathe more deliberately on the recovery and nothing else. If the back feels strained or painful rather than simply tired, stop and get it looked at — fatigue-related flexion is expected, pain is a different signal. And if everything above holds early and rounds off late, that is the pattern the studies describe, and the useful response is to shorten the piece rather than to fight it with your back.