Cable vs Dumbbell Lateral Raise: Which Builds More Side Delts?

For side delt size, the cable lateral raise did not beat the dumbbell version in the only head-to-head trial we found. Twenty-four trained lifters did one version with each arm for eight weeks. The side delt thickened by roughly 3 to 5% on both sides, and the statistics favoured "no difference" over "cable is better".

That result surprised the researchers, who had predicted a cable win. Below we cover what the trial did, why the prediction was reasonable, where the evidence is thin, and how to program and log both versions so your own training history settles the question for you.

Is the cable lateral raise better than the dumbbell lateral raise?

No, not on the evidence we found. Larsen and colleagues (2025, Frontiers in Physiology) trained one arm with cable lateral raises and the other with dumbbells, and both arms grew about the same. The authors summed it up this way:

"Our data suggest that both dumbbell and cable lateral raises are similarly effective for increasing lateral deltoid muscle hypertrophy in resistance-trained lifters."

Larsen et al., Frontiers in Physiology, 2025

A single trial cannot rule out a small difference, and we cover its limits further down. For a lifter choosing between two stations in a gym, though, the practical reading is that neither version has a proven edge.

What did the lateral raise trial actually measure?

The trial was a within-person comparison. Twenty-four resistance-trained adults (16 men and 8 women, with 3 to 16 years of lifting experience) had each arm randomly assigned to cable or dumbbell lateral raises. Lateral deltoid thickness was measured with B-mode ultrasound at two sites, near the shoulder (proximal) and further down the arm (distal).

The Bayesian analysis gave "moderate" support for the null hypothesis (no difference) at both the distal site (Bayes factor 0.27) and the proximal site (0.22). A Bayes factor below 1 favours the null, and the smaller it gets, the stronger that support. The combined analysis across both sites was stronger again (Bayes factor below 0.01).

Participants also followed a standardised lower-body programme and an optional upper-body programme, and they were told to avoid shoulder presses, front raises, other lateral raise variations and steep incline pressing. That kept other deltoid work from muddying the result.

Why did researchers expect cable raises to win?

Because the two lifts load the muscle differently across the movement. The authors reasoned that the cable version places peak torque near the muscle's longest length, whereas the dumbbell version loads the side delt most when it is shorter. A broader body of work suggests that training at longer muscle lengths may help growth in some muscles.

Kassiano and colleagues (2023, Journal of Strength and Conditioning Research) reviewed range of motion and hypertrophy and concluded that the quadriceps, biceps and triceps tend to grow best with longer-length training. Nobody had tested the idea on the side delt until this trial.

The result looks less surprising next to other work with trained lifters. Wolf and colleagues (2025, PeerJ) trained 30 resistance-trained people for eight weeks and found lengthened partial repetitions produced similar growth to full range of motion in the elbow flexors and extensors. Longer-length training does not appear to be a switch that makes every variation better, especially in experienced lifters. Our full range of motion vs partial reps guide covers that wider question.

What are the limits of the evidence?

• One trial, 8 weeks. Trained lifters grow slowly, so a short block gives small changes to compare.
• Small changes against measurement error. The ultrasound's typical error was about 0.4 mm, while thickness changes were around 0.5 to 1 mm. That makes small real differences hard to detect, and the authors say so.
• Matched range of motion. A cable can keep tension on at the bottom of the lift where a dumbbell cannot. Because both arms used the same range, the trial did not test a cable version that exploits that. The authors suggest the stretched position may have been under-emphasised.
• No control group and two measurement sites. The deltoid may also have changed at sites nobody measured.
• Nutrition was not tracked. Participants kept their usual diet and were encouraged to eat a little more and reach at least 1.6 g of protein per kg of body mass, but intake was not recorded.
• Muscle thickness, not appearance or performance. The trial did not test shoulder width visually or overhead pressing strength.

The authors also noted that set-by-set effort and total volume are probably bigger drivers of growth than the exercise variation, and both were matched here.

How close to failure should you take lateral raises?

Close. Robinson and colleagues (2024, Sports Medicine) pooled studies in a meta-regression and found that muscle growth tends to improve as sets end closer to failure, while strength gains were similar across a wide range of reps in reserve. They called the analysis exploratory, and reps in reserve were estimated rather than measured.

The lateral raise trial took every set to failure, but the failure point was clear and safe: the arm could no longer reach 90 degrees. A light isolation lift is a reasonable place to push hard because there is little technical risk. Our guide to training to failure vs RIR explains the trade-offs, and the RPE vs RIR guide shows how to rate sets.

How should you program cable and dumbbell lateral raises?

Everything in this section is our reasoning from the evidence above, not a protocol the trial tested. The only trial numbers used are the 12 to 16 rep range, double progression and about 10 weekly sets, all in trained lifters.

1. Choose the version you can load progressively. Dumbbells jump in whole steps, often 1 to 2 kg, which is a large jump on a lift this light. Many cable stacks and some dumbbell racks offer smaller steps. The trial used 0.25 kg steps, so a set-up with small increments helps.
2. Use the 12 to 16 rep range. Light isolation work suits higher reps. Keep each rep controlled at roughly a second up and a second down.
3. Match your range of motion. Raise to about shoulder height and lower under control. Swinging the torso or shrugging shifts work away from the side delt.
4. Take sets close to failure. Stop when you cannot reach shoulder height with a straight path, or a rep or two before that if your form breaks first.
5. Progress with double progression. Add reps until every set reaches 16, then add the smallest load you can. See our double progression guide.
6. Count your weekly sets. Pressing also trains the side delt. A heavy overhead press day contributes, so check how many sets per muscle group before adding more raises. The trial's participants averaged roughly 6 to 7 weekly lateral deltoid sets before the study, counting presses as half a set.
7. Rotate if you like. Alternating blocks of cable and dumbbell raises is reasonable. The trial gives no reason to expect a cost, and variety can keep joints comfortable.

Which lateral raise should you choose?

Choose the one you will do consistently, progress steadily and perform without shoulder pain. The trial found no growth advantage for either, so practical factors decide. These recommendations are our judgement, not trial findings.

Some participants in the trial reported shoulder pain in the strict frontal plane and were allowed to raise the arm midway between the frontal and scapular plane. Comfort matters more than a perfect line.

How do you track cable and dumbbell lateral raises?

Log them as separate exercises. A 10 kg dumbbell and a 10 kg cable stack setting do not load the muscle in the same way, so comparing the numbers directly makes your history misleading.

IronLedger records weight, reps, RPE or RIR and a note on every set. The weight is pre-filled from the target and editable per set, so you can enter what you actually lifted. A practical set-up for lateral raises:

• Weight and reps: the load per arm and the reps completed, aiming for the top of your rep range.
• RPE or RIR: how close each set was to failure, so 14 reps at RIR 3 is not mistaken for 14 reps at RIR 0.
• Notes: the pulley height, which hand, or "slight forward lean". Free text stays with the set in your history.

IronLedger also tracks personal records and estimated 1RM from what you log, but a lateral raise is not a lift you would test for a one-rep max. For this lift, the more useful view is the per-lift trend: reps and load across weeks. Run one block of each version, compare the trends, and keep the one that moves.

What mistakes should you avoid with lateral raises?

• Treating one trial as settled. Twenty-four lifters over eight weeks support a preference, not a rule.
• Comparing weights across versions. Cable and dumbbell numbers are not interchangeable. Compare each lift only to itself.
• Chasing load with momentum. Heavy swinging and shrugging reduce side delt work. Keep the tempo near a second up and a second down.
• Stopping far short of failure on every set. The meta-regression points toward effort mattering for growth.
• Changing several things at once. Swap one variable per block so you can tell what helped.

Keep learning

Related articles

Comparison

Overhead Triceps Extension vs Pushdown: Which Builds More Triceps?

11 min read

Comparison

Overhead Press vs Bench Press: What the EMG Research Shows

14 min read

Comparison

Full Range of Motion vs Partial Reps: What the Research Shows

14 min read