How To Do the Hip Thrust
The lift that built the glute training boom. It does what it says on the muscle, and does much less than promised for everything else.
Start here
The hip thrust is you with your upper back on a bench, a loaded bar across your hips, driving your hips up until your body makes a straight line from knees to shoulders.
It loads hip extension directly, at the top of the range, which no squat or deadlift does in the same way.
Short version of the technique: chin tucked, ribs down, drive through your heels, squeeze at the top, do not arch your lower back to fake height. And then there is the awkward conversation about what it actually does for you.
The meta-analysis nobody wants to read out loud
Researchers followed PRISMA guidelines, searched six databases and pooled 20 randomised controlled trials on hip thrust training, calculating effect sizes for strength, acceleration and jump performance.
Here is what came out.
Acutely, doing hip thrusts before sprinting produced a moderate post-activation performance enhancement on sprinting, effect size 0.55. That depended on protocol, specifically multiple sets and at least four minutes of recovery.
Long term, hip thrust training significantly improved hip thrust strength itself, effect size 0.53. But it failed to transfer to squat strength or to jumping performance.
There was a small transfer to change of direction, effect size 0.25. And when the researchers isolated hip thrust training on its own, excluding programmes that combined it with other work, sprint transfer was borderline at effect size 0.22, p equals 0.05, which the authors described as minimal independent benefit.
That is 20 trials saying the hip thrust largely makes you better at the hip thrust.
I want to be precise about what this does not say. It does not say the exercise is useless. It says its strength gains are specific, and you should not expect them to show up elsewhere on their own.
Where it does earn its place: glute growth
A systematic review and meta-analysis pooled twelve studies on gluteus maximus growth from resistance training and found a moderate overall effect.
On exercise selection, it pointed to prioritising single exercises such as the barbell hip thrust when the goal is glute hypertrophy at the expense of other muscle groups, and noted that back squats through parallel or full range also drive glute growth.
So the two findings live side by side without contradicting each other. The hip thrust is a strong choice for building the muscle. It is a weak choice if your goal is a bigger squat or a higher jump.
Pick it for what it does.
Use it as a primer, not just a builder
The most immediately usable number in the meta-analysis is the acute one.
Hip thrusts before sprinting improved sprinting, effect size 0.55, moderate. But it came with conditions: multiple sets, and at least four minutes of recovery before the sprint.
That recovery window matters. Do a few heavy sets, rest properly, then sprint. Do it as a rushed superset and you are just tired.
This is an acute, same-session effect. It is not a claim that hip thrusts make you faster over months, which is exactly what the long-term arm of the same analysis failed to find.
The bridge family it belongs to
A narrative review catalogued the supine bridge, the floor-based ancestor of the hip thrust, and its variations: single leg, with the hip adductors working, with the abductors working, with the upper trunk elevated, and with the feet elevated on stable or unstable surfaces.
It reviewed how these engage the muscles that stabilise the hip, pelvis and lumbar spine, including gluteus maximus, gluteus medius, hamstrings, erector spinae, multifidus, transversus abdominis and external oblique.
Useful context: if a loaded hip thrust is too much, or your lower back complains, the bridge family gives you graded options that train the same pattern.
This is a narrative review written toward therapeutic and preventive use, not a performance trial.
How to actually do it
Upper back on the bench at about the bottom of your shoulder blades. The bench should not be so high that you cannot get your hips through.
Tuck your chin and look forward, not up. Looking up drags your whole spine into extension.
Keep your ribs down. This is the one that matters most. The extra height people get by arching their lower back is not hip extension, it is lumbar extension, and it is the fastest way to make this lift hurt.
Feet flat, shins roughly vertical at the top.
Drive up until your body is a straight line from knees to shoulders. Stop there. There is no bonus range above straight.
Squeeze, then lower under control.
End the set when you start getting height from your lower back instead of your hips.
Use the hip thrust to build glutes, where a meta-analysis puts it among the priority choices, and to prime sprinting in the same session (multiple sets, at least four minutes rest, moderate effect). Do not expect it to raise your squat or your jump: 20 pooled trials found it improved hip thrust strength but failed to transfer to either, with only minimal independent sprint benefit when trained alone. Chin tucked, ribs down, drive to a straight line from knees to shoulders and no further. When the extra height starts coming from arching your lower back, the set is over.
Limits of this evidence
- The transfer findings are pooled effect sizes across 20 heterogeneous trials, so they describe an average across varied protocols and populations rather than a guaranteed individual outcome.
- The isolated sprint transfer result sat exactly at the conventional significance threshold (p = 0.05), which is weak evidence in either direction.
- The acute sprint benefit is a same-session effect and was protocol dependent, requiring multiple sets and at least four minutes of recovery; it does not indicate long-term speed gains, which the same analysis failed to find.
- The glute hypertrophy meta-analysis pooled only twelve studies and reported a moderate effect for resistance training generally rather than isolating the hip thrust's superiority in a head-to-head design.
- The supine bridge source is a narrative review aimed at therapeutic and preventive use, not a controlled performance study.
- None of these sources measured injury outcomes, so the technique cues here are reasoning about joint position rather than injury evidence.
- Each claim rests on a single source with no independent replication checked. Treat the piece as provisional.
Sources (3)
- Acute and long-term effects of hip thrust training on athletic performance: a systematic review and meta-analysis. PeerJ, 2026
- The impact of resistance training on gluteus maximus hypertrophy: a systematic review and meta-analysis. Frontiers in Physiology, 2025
- Supine Bridge Exercise: A Narrative Review of the Literature (Part I). Cureus, 2025