Dumbbell exercises for chest

Most lists of these are fifteen exercises with a stock photo each and no reason to pick one over another. This is ten, because ten is what the library holds, and every one has a demonstration and a written page naming the muscle driving it and the muscles only steadying you.

Three things worth settling before the list, because they decide how you use it: what an incline really does, whether there is an inner chest, and what happens if you have no bench.

The short version

The ten, and what each one is for

Four are done on a flat bench, five on an incline, and one on the floor. That split is the most useful thing about the list, so it is how the list is ordered.

Flat

The dumbbell bench press is the one to learn first and the one to do most. Your chest drives it, the triceps and front delts press alongside, and you can keep adding weight to it for years. The hammer press is the same press with the palms facing each other and the elbows tucked, which suits a shoulder that complains about the flared position. The close chest press keeps the two bells squeezed together throughout, so the chest stays under tension across the whole press rather than resting at the top.

The flat fly is the odd one out and is worth understanding rather than just doing. Its muscle panel lists no triceps at all, because a fly holds the elbow angle fixed for the whole repetition and the triceps have no joint left to move. That is the entire difference between a fly and a press, and it is why swapping one for the other is not a swap.

Incline

The incline bench press is the flat press tilted, and the next section is about what that tilt actually buys. The incline close chest press is that tilt with the squeeze held throughout. The alternating incline press is the same tilt one arm at a time, which asks your midsection to stop you rotating. The incline hammer press to hammer curl adds a curl to the top of it, so it is a chest and biceps sequence rather than a chest exercise.

The incline fly comes with a correction most lists do not make. It is sold everywhere as upper chest isolation. On this site the front delt is listed as a co-primary rather than as a muscle that only steadies you, because the pooled shoulder research ranks it highest of all for that head. It is a chest and front delt exercise. Knowing that is the difference between wondering why your shoulders are tired and expecting it.

Floor

The wide dumbbell push-up is the only one of the ten that needs no bench. Gripping the handles lets your chest travel below your hands, which a push-up on the floor will not.

What an incline actually does

Every page that recommends an incline says 30 to 45 degrees and none of them says why. The study behind that number is worth reading properly, because it does not say what it is quoted as saying.

Rodríguez-Ridao and colleagues put 30 trained adults on a bench at 0, 15, 30, 45 and 60 degrees, with the load set at 60% of the one-rep max measured at each angle. The abstract reports that the upper portion of the chest was most active at 30 degrees, and that sentence is why the internet believes 30 degrees is the upper chest angle.

Its own comparisons say otherwise. The upper portion at 30 degrees was not significantly different from flat, at p = 0.688. Nor was 15 degrees (p = 0.977) or 45 degrees (p = 1.000). "Most active" there means the highest number among numbers that could not be told apart.

What the same study does show significantly is the cost. The middle and lower portions were highest at flat and dropped significantly at every single incline. The front deltoid was significantly more active than every other muscle measured at 45 and 60 degrees. And at 60 degrees the upper portion of the chest was significantly less active than it was lying flat. So inclining does not reliably buy upper chest, and it does reliably cost you the rest of the muscle while handing work to the shoulder.

The strongest population anybody has tested agrees. Saeterbakken and colleagues measured 12 athletes who compete at bench press, at national and international level, across incline, flat and decline. Pectoralis major activation did not differ significantly between the three. What differed was the arms and the load: the triceps did 58.5 to 62.6% less on the incline, the biceps 48.3 to 68.7% more, and the 6RM weights were 18.5 to 21.5% lower.

Here is the part nobody reconciles, and it points the other way. Every number above is electrical activity during a set, and activity during a set has not been validated as a predictor of muscle built over months. The one study that measured muscle instead found for the incline. Chaves and colleagues trained 47 untrained men once a week for eight weeks. The incline group gained more thickness at the upper site than the flat group, a mean difference of 0.62 cm. Only one of three measured sites differed, and the group doing both exercises did not win. The authors' own conclusion was that flat, incline or both "can render similar change in general strength".

So: untrained men, once a week, eight weeks, one site out of three. That is the whole of the evidence that inclining builds an upper chest, and it is more than the activation studies give it. What follows from both together is modest. Keep a flat press as the main one. If you incline, set the bench at the bottom of its range rather than the top, because the only thing the angle research shows clearly is what you lose as it gets steeper. Every incline exercise here says 30 to 45 degrees in its steps; 30 is the end to use.

There is no inner chest

The phrase is everywhere on this subject. One of the pages ranking above this one calls a squeeze press an "inner chest burner" and carries no anatomy at all.

Pectoralis major has two heads, the clavicular and the sternocostal, named for where they start. They start in different places and they have different nerves. They end in the same place: the insertion is a single site on the upper arm bone, at the lateral lip of the intertubercular sulcus of the humerus.

That single ending is the whole answer. The muscle starts across the inner half of your collarbone, your sternum, seven costal cartilages, a rib and the sheet over your abdominal wall, and it all converges on one point on your arm. The part you mean by "inner" is the origin, the end bolted to your chest. It does not travel, and there is nothing there for an exercise to shorten toward that is not already shortening on every repetition you do.

The training evidence agrees with the anatomy. Pinto and colleagues had 15 untrained men train the pec deck for 12 weeks, and measured both heads by ultrasound. The clavicular head grew 17% and 18%, the sternocostal head 21% and 21%. Their conclusion was that the results "do not support regional hypertrophy after pec deck exercise in untrained men". The same authors note, and it is worth repeating, that "research on the pectoralis major is notably limited despite it being a frequently targeted muscle group".

A squeeze press is still a good exercise. Holding the bells together keeps tension on the chest through the top of the press, where an ordinary press lets it off. That is a real reason to do it. It is not a region.

Press or fly, and why they are not interchangeable

A press bends and straightens your elbow. A fly does not: the elbow angle is set before the first repetition and held there. That means the triceps have no joint to work across, which is why the flat fly on this site lists no triceps where every press lists them.

The nearest comparison measured is a barbell bench press against a dumbbell fly, in 17 resistance-trained men with nearly five years of training behind them. The press produced higher activation across the whole repetition for the chest, front delt and triceps, reported as a range of 8 to 81% across those muscles. The biceps ran the other way, 57 to 86% more active in the fly.

Read that carefully. Two things change at once between those conditions: the bar becomes two bells, and the press becomes a fly. It is not a dumbbell press against a dumbbell fly, and nobody has published that comparison for muscle gained. What it supports is the ordinary conclusion: press for the load, fly for the tension in a position a press cannot reach, and do not treat one as a substitute for the other.

What a dumbbell costs against a barbell

One number, and it is a useful one. Saeterbakken and colleagues tested 12 resistance-trained men for a one-rep max in a Smith machine, with a barbell and with dumbbells. The dumbbell load was 17% less than the barbell and 14% less than the machine.

The activation did not follow the load. Chest and front delt activity did not differ between the three. What changed was the arms. The biceps worked harder as the weight got harder to balance, and the triceps did less with dumbbells than with either the barbell or the machine.

So the trade is real and it is not the one usually described. You will handle meaningfully less weight, and the muscle you came for is being asked to do a comparable job. What that study did not measure is range of motion, which is the thing every article asserts about dumbbells. As far as this research found, nobody has measured it.

If you have no bench

This is where an honest answer is worth more than a longer list. Nine of these ten exercises need a bench, and five of those need it to incline. Only the wide dumbbell push-up does not.

Pages chasing this question pad the gap with floor presses. There is no study comparing a floor press with a bench press, for activation, for range of motion, or for muscle gained. Not one. Every confident sentence you will read about the floor press traces back to somebody's opinion. That includes the page ranking sixth for this term, which asserts that a floor press without a bench beats one with a bench, and explains nothing.

The substitute that does have evidence is not a dumbbell exercise at all. Blackard and colleagues used a force plate to equate the load between a push-up and a bench press, and found no significant difference in chest or triceps activation between them. Their conclusion was that "external load is more important than boundary condition in describing human activity", which is a formal way of saying the muscle does not know what you are holding. Over eight weeks it holds up. Kikuchi and Nakazato trained 18 men twice a week, on either a bench press at 40% of one-rep max or push-ups adjusted to that same load. Both groups gained comparable thickness and strength, and only the biceps favored the bench.

So with no bench, do the wide dumbbell push-up and make it hard, rather than doing a worse version of a press. The handles are the point: they let your chest drop below your hands, which the floor will not allow. When that gets easy, the load has to come from somewhere, and adding repetitions is not adding load.

Common questions

What is the best dumbbell exercise for the chest?

The flat dumbbell bench press, for most people. Your chest is the muscle driving it and you can add weight to it for years. The activation research also finds the middle and lower portions of the muscle most active at a flat angle. Everything else on this page is a variation on it, or a narrower version of it.

Can you build a chest with dumbbells alone?

Yes. The bar is not the thing holding you back. In 12 resistance-trained men the dumbbell one-rep max was 17% lower than the barbell, but activation of the pectoralis major and the front deltoid did not differ between the two. You handle less weight and the muscle is asked to do a comparable job.

How do you hit all three chest muscles?

There are not three. Pectoralis major has two heads, the clavicular and the sternocostal, named for where they start, and pectoralis minor sits underneath and does not press. The research that divides the muscle further divides it into upper, middle and lower portions of one muscle, and nobody measures an inner one.

Is there an exercise for the inner chest?

No, and the anatomy is the reason. Pectoralis major starts across the clavicle, the sternum, seven costal cartilages, a rib and the abdominal wall, and all of it inserts at one place on the upper arm bone. The part you mean by inner is the end fixed to your chest, which does not travel. When 15 men trained one exercise for 12 weeks, the two heads grew 17% and 21%, and the authors said their results do not support regional growth.

Does an incline bench press work the upper chest?

The honest answer is that the evidence disagrees with itself. The most-cited activation study found 30 degrees no different from flat for the upper portion, and 12 competitive bench pressers showed no pectoralis difference across incline, flat and decline at all. But the one study that measured muscle rather than activation found the incline group gained more thickness at the upper site. What inclining reliably does is cost you the middle and lower portions and hand work to the front deltoid.

What angle should the bench be?

If you incline, use the bottom of the range rather than the top. Activation of the front deltoid was significantly higher than every other muscle measured at 45 and 60 degrees, and at 60 degrees the upper portion of the chest was significantly less active than on a flat bench. Past about 45 degrees you are increasingly training a shoulder press.

Can you train your chest with dumbbells and no bench?

You can, and you should know what you are giving up. Nine of the ten dumbbell chest exercises here need a bench. The honest substitute is a push-up rather than a dumbbell exercise. With the load equated by force plate, a push-up and a bench press produced no significant difference in chest activation. Over 8 weeks, push-ups loaded to match a bench press at 40% of one-rep max produced comparable gains.

How many chest exercises should you do?

Nobody has measured it for the chest. The one study on chest volume found one set as good as three over 12 weeks in untrained men. Every meta-analysis showing that more sets do more measured arms and thighs rather than the chest. Two or three exercises covering a flat press, an incline and a fly is a reasonable shape, not a finding.

Where these figures come from

Every figure on this page comes from one of the records below, each read in full. Where a study measured electrical activity rather than muscle built, the page says so in the same sentence as the number.

• Rodríguez-Ridao D, Antequera-Vique JA, Martín-Fuentes I, Muyor JM (2020). Effect of five bench inclinations on the electromyographic activity of the pectoralis major, anterior deltoid, and triceps brachii during the bench press exercise. International Journal of Environmental Research and Public Health 17(19):7339. PMID 33049982. The angle study, and the p values its abstract leaves out.
• Saeterbakken AH, Mo DA, Scott S, Andersen V (2017). The effects of bench press variations in competitive athletes on muscle activity and performance. Journal of Human Kinetics 57:61-71. PMID 28713459. Twelve competitive bench pressers, and no chest difference across three bench angles.
• Chaves SFN, Rocha-Júnior VA, Encarnação IGA, Martins-Costa HC, Freitas EDS, Coelho DB, Franco FSC, Loenneke JP, Bottaro M, Ferreira-Júnior JB (2020). Effects of horizontal and incline bench press on neuromuscular adaptations in untrained young men. International Journal of Exercise Science 13(6):859-872. PMID 32922646. The only study that measured muscle rather than activation, and the one that favors the incline.
• Vigotsky AD, Halperin I, Trajano GS, Vieira TM (2022). Longing for a longitudinal proxy: acutely measured surface EMG amplitude is not a validated predictor of muscle hypertrophy. Sports Medicine 52(2):193-199. PMID 35006527. Why every activation number above carries a caveat.
• Solari F, Burns B. Anatomy, thorax, pectoralis major. StatPearls, NCBI Bookshelf NBK525991. PMID 30252247. The origin, the single insertion and the two heads.
• Fung L, Wong B, Ravichandiran K, Agur A, Rindlisbacher T, Elmaraghy A (2009). Three-dimensional study of pectoralis major muscle and tendon architecture. Clinical Anatomy 22(4):500-8. PMID 19291757. Eleven cadavers, 6 to 7 sternal segments, and no twist in the muscle or the tendon.
• Pinto MD, Ughini C, Nunes JP, Cadore EL, Pinto RS (2025). Pectoralis clavicular and sternocostal thicknesses increase similarly in response to one and three sets of pec deck resistance training in untrained young men. Journal of Strength and Conditioning Research 39(5):523-530. PMID 40266636. The regional growth question, measured.
• Saeterbakken AH, van den Tillaar R, Fimland MS (2011). A comparison of muscle activity and 1-RM strength of three chest-press exercises with different stability requirements. Journal of Sports Sciences 29(5):533-8. PMID 21225489. The 17%, and the activation that did not follow it.
• Solstad TE, Andersen V, Shaw M, Hoel EM, Vonheim A, Saeterbakken AH (2020). A comparison of muscle activation between barbell bench press and dumbbell flyes in resistance-trained males. Journal of Sports Science and Medicine 19(4):645-651. PMID 33239937. A barbell press against a dumbbell fly, which is two changes at once.
• Blackard DO, Jensen RL, Ebben WP (1999). Use of EMG analysis in challenging kinetic chain terminology. Medicine and Science in Sports and Exercise 31(3):443-8. PMID 10188750. Load equated by force plate, and no difference left.
• Kikuchi N, Nakazato K (2017). Low-load bench press and push-up induce similar muscle hypertrophy and strength gain. Journal of Exercise Science and Fitness 15(1):37-42. PMID 29541130. Eight weeks of the same comparison.

The exercise counts are from this site's own catalogue, counted on 12 September 2026. The demonstration clips are licensed footage, not filmed here. Nothing is sold on this site and no brand of dumbbell is recommended.

Five things this page does not claim. That an incline builds your upper chest. That 30 degrees is the best angle. That a decline press beats a flat one for the lower chest. That any exercise trains an inner chest. That a floor press is better or worse than a bench press. The last two comparisons have never been done, and the rest is set out above with the evidence pointing both ways, which is the honest state of it.

If a lift hurts rather than being hard, stop and see a physiotherapist.