Hypertrophy & Strength / Glute Training

You Can't Feel Hypertrophy

You've been picking exercises by what you feel. Nobody told you the feeling doesn't predict the result.

Field Notes  ·  2026  ·  5 min read

You know the exercises that light you up and the ones that don't, and you've built your training around the difference. Everyone does. The burn is the closest thing to a readout most lifters ever get.

Thirty-four people, nine weeks, one exercise each. The group whose glutes lit up more on the day grew no more glute than the group whose didn't.

Which means the readout you've been using isn't reading what you think it is — and every session spent chasing it is a session where something else could have moved.

What They Actually Did

In 2023, Plotkin and colleagues randomised 34 untrained college-aged people to a single exercise for nine weeks: hip thrust or back squat, nothing else. Two supervised sessions a week, 8–12 reps to volitional failure, sets ramping from three per session in week one to six by week nine.

They measured gluteus maximus by MRI in three separate regions, plus glute medius and minimus, quadriceps, hamstrings and adductors. They tested 3RM squat, hip thrust and deadlift. And during the first session they recorded surface EMG at the glute — the activation reading that fuels most of what you've read about this exercise.

What They Found

Glute growth: no detectable difference. Every gluteal site crossed zero — lower glute max −1.6 cm² (−6.1 to 2.0), mid −0.5, upper −0.5, medius and minimus −1.8. The estimates tip slightly toward the hip thrust and the intervals are far too wide to call it either way.

The squat added thigh on top. Quadriceps +3.6 cm² (0.7 to 6.4), adductors +2.5 (1.2 to 3.9). Those two don't cross zero — that difference is real.

Strength went wherever it was trained: squat 3RM to the squat group, hip thrust 3RM to the thrust group, by a distance in both cases. Deadlift transfer was identical. Zero kilos between them.

The Part Worth The Post

The hip thrust produced greater mean EMG at every gluteal site measured. More signal. More of the thing you feel.

It grew no more glute.

Then the authors checked whether the reading predicted who actually grew. Lower glute max r = 0.18. Mid glute max r = −0.03. Upper glute max r = 0.50. Medius and minimus r = 0.28. One moderate relationship, one at zero, one fractionally negative. Their own words: the ability to forecast muscle growth from EMG was "generally poor and variable", and acute readings taken during a workout "are not predictive of hypertrophic outcomes".

The sensation is real. It just isn't measuring the thing you're trying to change.

Worth knowing who ran it. Bret Contreras is the eighteenth author on this paper — the coach most associated with the hip thrust, and with the activation work that made it popular, on the study finding that its activation advantage didn't show up in the tissue. Nineteen authors, an MRI scanner and a randomised design, and nobody buried the result.

Where It's Weak

Untrained twenty-somethings, nine weeks, one exercise each, 34 people split two ways. Almost anything grows muscle in untrained people, and differences between two good exercises are hardest to detect in exactly that population. "Similar" here means undetectable in this sample, not proven identical — the wide intervals are the tell. Nobody has run this in trained lifters across a year.

The EMG finding is the sturdier half. It compares the same people against themselves rather than one group against another, it held across several analytical approaches, and it's the conclusion the authors state most flatly. That's the part to carry out of this study; the glute-growth comparison is the part to hold loosely.

Test It Now

Take the glute exercise you feel the most. Open your log and find what you were loading it with eight weeks ago, and what you're loading it with today. Reps at that weight, not just the weight.

If the weight and the reps are where they were eight weeks ago, you've been repeating a session, not progressing one. The burn showed up every time. The load didn't.

What It Means For Your Training

Pick exercises for what they load, not what they light up. If you can only run one, the squat delivered the same glute and more thigh. Keep the hip thrust for what it demonstrably does — it built hip thrust strength far better than the squat did, and it transferred to the deadlift just as well. Just stop using the burn as the scoreboard.

The Bottom Line

You can't feel hypertrophy. You can feel effort, tension, blood and discomfort — all real, none of them the measurement.

Nine weeks of supervised, volume-matched training says the exercise that felt like more wasn't more. And when the researchers went looking for a link between the sensation and the result, they couldn't find one they'd trust.

Open your log and run the check. If the load hasn't moved in two months, no amount of feeling is going to have moved the tissue either.

What builds a glute is load and progression over time — and which exercises deliver that for you depends on how you're built and where you can actually load without borrowing range from somewhere else. Get that right and the number on the bar moves.

It's what I write week after week, for lifters and for people training around something. Feeling a muscle work doesn't make the exercise better, or the set more effective. Train the tissue, not the sensation.

References

Plotkin et al. Hip thrust and back squat training elicit similar gluteus muscle hypertrophy and transfer similarly to the deadlift. Frontiers in Physiology. 2023;14:1279170.

Luke Sanders is an osteopath and strength coach based in London. He works with lifters, athletes and people who want to train without paying for it later.

Questions & Answers

Common Questions

It's the most common reason people pick an exercise, and it's the one thing this trial tested directly. The hip thrust produced more gluteal EMG than the squat at every site measured, and both groups finished with the same amount of glute. The sensation tells you a muscle is working. It doesn't tell you how much it will grow.
It works. Nine weeks of it grew glutes as well as nine weeks of squatting, and transferred to the deadlift just as well. Nothing here says drop it. What the trial rules out is the specific claim that it grows glutes better because it feels like more — that part didn't survive the measurement.
Because EMG measures electrical activity during one set, and hypertrophy happens over weeks in response to mechanical tension and progressive load. When the authors correlated the two, the relationship ranged from moderate to slightly negative depending on the site. Their conclusion was that acute readings taken during a workout aren't predictive of what grows.
If you can only run one lift, the squat gave the same glute plus real quadriceps and adductor growth — more total muscle for the same sets. But strength went to whichever exercise was trained, so the answer depends on what you want to be strong at, and on which movements you can load well.
Load and progression over time, in positions you can actually load. Which exercises those are depends on how you're built and what you're currently compensating with — get that wrong and the sets accumulate without the load moving. And form matters in the other direction too: load a position you can't hold, and you build the compensation instead of the muscle.

Want to know what's actually limiting your movement? Book a clinical assessment — we find the root and fix it.

Book Now