How To Do the Seated Cable Row

Start here

The seated cable row is the most forgiving big back movement in the gym, which is exactly why it gets done badly.

There is no bar to balance, no spine loaded from above, and no floor position to earn. You sit down and pull. So the technique errors are quiet ones: momentum from the hips, a rounding lower back, and elbows that finish in the wrong place.

And it is the exercise where the loudest coaching disagreement is about something you cannot see from the front: whether your shoulder blades should move.

The fixed-versus-free shoulder blade question

Somebody built a study around exactly this.

Fourteen resistance-trained men, averaging 25 years old, performed the seated row in two ways in a randomised crossover design: with a fixed scapular position and with a free scapular position, using their 8-repetition maximum as the load for both variations.

Measurement was high-density surface electromyography, which uses grids of electrodes rather than single sensors. That let them record not just how much a muscle was excited but where within the muscle. They recorded upper, middle and lower trapezius, latissimus dorsi, lateral and posterior deltoid, biceps, triceps and erector spinae.

They computed normalised root mean squared amplitude plus excitation centroids in both the side-to-side and up-and-down planes, and they analysed the concentric and eccentric phases separately.

That separation matters. Pulling in and controlling the weight back out are different mechanical events, and treating a set as one undifferentiated thing hides half of what happens.

I am reporting this study's design rather than its full result table, because the per-muscle and per-phase comparisons are the substance of the paper and are not something to compress from an abstract. What it establishes for you is that the fixed-versus-free question is a real, measurable difference in how the work is distributed, not a matter of taste.

The practical reading: pick one deliberately and be consistent within a set. Letting the scapulae drift free on some reps and locking them on others means you are running two exercises inside one set and progressing neither.

Fourteen resistance-trained men in their twenties, one machine, one load. Small, specific, and a starting point rather than a settled answer.

Set the position before you pull

Feet flat and planted on the platform, knees slightly bent rather than locked.

Sit tall. Chest up, ribs down, a neutral spine you can hold for the whole set. Not arched back into a lean, not hunched forward reaching for the handle.

The most common error here is using the lower back as a lever, rocking backward to start the rep and forward to return the weight. That turns a back exercise into a lumbar exercise with a handle attached.

If you cannot hold the torso still, the weight is too heavy. That is not a compromise, it is the actual answer.

A small amount of forward lean in the stretched position is fine when it comes from the hips and the spine stays neutral. Rounding is different from hinging.

These positioning cues are standard coaching practice rather than findings from the sources cited here.

Elbow path and where the work lands

The elbows decide which muscles do this.

Elbows tucked closer to the ribs and pulled straight back, with the handle arriving low on the torso, biases the lats and puts the pull along the line of the ribcage.

Elbows flared out wider, with the handle arriving higher toward the sternum, shifts the emphasis toward the mid-back, rear delts and traps.

Neither is wrong. They are different exercises sharing a machine, and choosing one on purpose is the entire skill.

Pull with the elbows rather than the hands. Think about driving the elbows behind your body, and let your hands be hooks. Gripping hard is the fastest way to make this a biceps exercise.

At the finish, the handle touches the torso, the elbows are behind the plane of the body, and the shoulder blades are where your chosen technique says they should be. Then control it back out. The eccentric is training, not transport.

Also standard coaching practice, included because a guide that skipped it would be useless.

How much does grip actually change things

Grip variation is the other thing people argue about, and there is evidence from the closely related pulldown.

A study examined electromyographic activity of the major back muscles across seven lat pulldown variants differing in grip type, grip width and trunk inclination. Forty men with at least five years of resistance training experience performed five repetitions at 70 percent of their repetition maximum, with maximal voluntary contraction tests run beforehand to normalise every muscle: latissimus dorsi, posterior deltoid, biceps, middle and lower trapezius, and infraspinatus. Full, concentric and eccentric movements were each recorded.

The multivariate analysis of variance did not return significant differences across those variants.

Seven variations of grip and width and trunk angle, forty experienced lifters, and the multivariate test did not separate them.

The study exists because, in its own words, there are significant gaps in the literature regarding the relationship between exercise modality and specific muscle activation. That is the field admitting the confident grip advice you have heard is not built on much.

This is a pulldown study, a vertical pull, not the horizontal row. And I am reading a partial result, so the per-muscle detail sits in the full paper. The direction is still worth carrying: grip fiddling is a smaller lever than most people believe, and elbow path and load are bigger ones.

Machine or free weight, and what it costs you

The cable row is a machine movement, and there is a meta-analysis on what that means.

A PRISMA systematic review and meta-analysis compared free-weight against machine-based strength training, including only studies that directly compared the two for a minimum of six weeks in adults aged 18 to 60. Thirteen studies were included with a total sample of 1,016 participants, 789 men and 219 women, covering maximal strength, jump performance and hypertrophy.

Strength measured in free-weight tests increased significantly more with free-weight training, with a standardised mean difference of negative 0.210, a confidence interval from negative 0.391 to negative 0.029, and a p value of 0.023. Strength in machine-based tests tended to increase more with machine training.

That is specificity, stated cleanly in both directions. You get better at the thing you actually practise, and the test format matters as much as the training format.

So the honest position on the cable row: it is an excellent way to load a horizontal pull with low technical cost and easy progression. It is not a substitute for a barbell row if barbell rowing is what you want to be strong at. Run both if both matter to you.

Thirteen studies and 1,016 participants across a range of exercises and populations, not the seated row specifically.

Why any of this is worth your time

The cable row earns its place because it lets you train a horizontal pull hard without the technical tax of a barbell row, and progression on it is simple.

What the evidence supports is narrower than the confidence around it. Scapular position is a genuine, measurable variable worth choosing deliberately. Grip variations appear to be a weaker lever than commonly claimed. Training is specific, so a machine builds machine strength and a barbell builds barbell strength.

What that leaves you: sit tall, keep the torso still, choose an elbow path and a scapular strategy on purpose, hold both for the whole set, control the way back out, and add load slowly.

This is a training article, not medical advice. Pain in the shoulder or lower back during rowing is a reason to see a qualified professional, not a reason to change your grip and push on.