How To Do the Overhead Press

Start here

The overhead press is the lift where your whole body has to agree with you.

Everything under the bar is either holding position or leaking force. The bar path is short, the margin is small, and there is nowhere to hide a weak brace.

Before the technique, one thing about the evidence you are about to read, because it changes how confidently anybody should be talking to you about this lift.

The literature is thinner than the confidence around it

A review in Sports Medicine examined weightlifting overhead pressing derivatives, covering their historical, technical, kinetic and kinematic mechanisms along with guidelines for using them as training tools.

Only 13 articles were found in a search of electronic databases.

Thirteen. For the push press, push jerk and split jerk taken together, from the front rack and from the back.

That is not a reason to avoid the lift. It is a reason to be suspicious of anybody giving you a very precise rule about it, because the empirical base underneath those rules is small.

What the review does support is why these movements are worth training. They require the ability to develop high forces rapidly through an impulsive triple extension of the hips, knees and ankles, which is mechanically similar to many sporting tasks, and they may be beneficial for enhancing power development and maximal strength.

Triple extension is the point. The pressing derivatives are not shoulder isolation with extra steps. They are a whole-body force expression that happens to finish overhead.

That review covers the weightlifting derivatives specifically, the push press and jerks, rather than the strict standing press. Take the principle about the body underneath, and treat the 13-article count as the honest state of the evidence.

Set the feet and lock the floor

The press starts at the ground, which sounds like a slogan until you press without it.

Feet about hip width, toes forward or very slightly out. Grip the floor. Weight through the middle of the foot, not the heels and not the toes.

Squeeze your glutes hard and brace your midsection before the bar moves. If your ribs flare and your lower back extends to get the bar up, you have turned a press into a standing incline press and moved the load somewhere it does not belong.

The cue that fixes most of it: think about pressing yourself away from the floor rather than pressing the bar away from you.

This positioning guidance is standard coaching practice rather than a finding from the sources cited here.

Grip and the rack position

Hands just outside shoulder width. Forearms as close to vertical under the bar as your build allows, because a vertical forearm is a direct line of force and an angled one is a lever you have to fight.

The bar sits on the front delts at the start, not floating in front of them. Elbows slightly ahead of the bar, not flared wide and not tucked so far under that your wrists bend back.

Wrists stacked over the elbows, knuckles up. A collapsed wrist is the most common reason people feel the press in their wrists and blame their shoulders.

And get your chin out of the way rather than pushing the bar around your face. Small head movement backward, then back to neutral once the bar clears.

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

What the arm position research suggests about going overhead

There is biomechanical work on how loaded the shoulder gets in different arm positions, and it is worth understanding even though it was not done on lifters.

A cadaveric biomechanical study evaluated six shoulders using a custom shoulder testing system with a pressure-mapping sensor, measuring contact area and contact pressures between the subcoracoid arch and the subscapularis complex. Eight arm positions were assessed, including 20 and 60 degrees of forward flexion and of abduction, each with maximal internal and maximal external rotation.

The overall incidence of contact was 52.1 percent across all tests. Critically, no contact was observed at 20 degrees of abduction in the maximal internal rotation position, and the mean contact area differed significantly across the other seven positions.

Mean and peak contact pressures did not differ significantly across positions, at p values of 0.188 and 0.065.

Read that carefully, because the honest summary is mixed. Which position the arm was in changed how much contact area there was. It did not significantly change the pressure.

The practical takeaway is narrow and it is about rotation: arm position and rotation together determine what happens inside the joint, which is the mechanical argument for keeping your forearms vertical and your elbows in a consistent path rather than letting them drift wherever the weight pulls them.

Six cadaveric shoulders with a mean age of 68. This is laboratory biomechanics on tissue, not lifters pressing weight, and it does not describe what happens in your shoulder under a barbell. It is not a diagnosis of anything and it is not rehabilitation guidance. Shoulder pain is a matter for a qualified professional, not an article.

Press to full lockout, and why range of motion matters

Finish the rep. Bar over the middle of the foot, arms straight, head through, shoulders shrugged up slightly at the top.

A half-locked press is not a shorter press, it is a different exercise, and there is evidence bearing on why the range matters.

A review examined the interplay between muscle length, range of motion and exercise selection, synthesising long-term studies comparing hypertrophy after isometric training at different muscle lengths, training with different ranges of motion, and exercises loading muscles at different lengths and resistance profiles.

Different lines of investigation suggest muscle length plays a role, with training at longer muscle lengths eliciting more favourable muscle growth.

The review is careful about what it can conclude, and so am I: it states directly that it is difficult to identify whether muscle length per se is the main factor, which limits the ability to draw practical conclusions.

So the direction of the evidence favours full range over partial, without the mechanism being settled.

On the way down, control it. Lower to the front delts under tension rather than dropping the bar onto your chest and bouncing. The eccentric is training too.

This review addressed muscle length and range of motion generally, not the overhead press, and it explicitly flags its own difficulty in isolating the mechanism.

Choosing the variation, and where the injuries cluster

Strict press, push press, seated press, dumbbell press. They are not interchangeable and the choice matters more than the fine detail of your grip.

Strict standing press: least weight, most whole-body control demand. Push press: uses the leg drive that review described, letting you move more load and train the triple extension. Seated press: removes the leg contribution and the balance demand, which makes it easier to load the shoulders and harder to cheat. Dumbbells: each side works independently, and the path is free rather than fixed.

On exercise selection generally, an integrative review analysed the relationship between exercise selection and musculoskeletal injuries in fitness centres, following PERSiST reporting guidelines across four databases, with 39 peer-reviewed articles meeting inclusion criteria.

Most injuries occur to the shoulders, elbows, vertebrae of the spine and knees.

Shoulders lead that list, and the overhead press is a shoulder-dominant lift, so the case for building it patiently rather than testing it weekly is straightforward.

The review is clear that injury causes are multifactorial, so exercise selection is one contributing factor among several rather than the explanation.

Start lighter than your ego wants. The press progresses slowly for almost everybody, and there is no version of this lift where forcing the load works better than earning it.

Why any of this is worth your time

The press is honest. It tells you immediately whether your brace, your position and your patience are real.

The evidence base for its variations is thin, which means the confident rules you hear are mostly convention rather than data. The parts that do have support are simple: the lift is a whole-body force expression, full range of motion is favoured over partial, shoulders are where gym injuries cluster, and arm position changes what happens inside the joint.

That is enough to train well on. Feet set, ribs down, forearms vertical, full lockout, controlled descent, load added slowly.

None of this is medical advice. Pain in the shoulder during pressing is a reason to see a qualified professional, not a reason to change your grip and push on.