Push pull legs: how it works and how to set it up

Push pull legs groups your training by movement pattern: chest, shoulders and triceps on push day, back and biceps on pull day, everything below the waist on leg day. Run it six days a week and every muscle group gets trained twice, which is where the split earns its reputation. Run it three days a week, which is common, and each muscle group gets trained once, with every set you do for it crammed into one session.

That last part is the problem.

Push pull legs is the most searched workout split there is. It also gets recommended to plenty of people who shouldn't be running it yet.

I'm not going to talk you out of it. Six-day push pull legs is a genuinely good way to train, and the logic behind the split is sound. But on three days a week each muscle group only gets trained once, and once a week is low frequency for anyone past their first few months of lifting. It also leaves one session to carry a whole week of sets for three muscle groups at once, and that's where it comes apart.

What is a push pull legs split?

Push pull legs splits your training by movement pattern rather than by body part. Push day covers everything you press away from you: chest, front and side delts, triceps. Pull day covers everything you pull toward you: back, rear delts, biceps. Leg day covers the lower body: quads, hamstrings, glutes, calves.

The appeal is that muscles which work together get trained together. Your triceps assist on every chest and shoulder press, so training them on the same day means they get one hard day and then several days off, instead of being hammered on Monday by bench press and again on Tuesday by direct arm work. Your biceps assist on every row and pull-up, so they belong with back day for the same reason. It's why push pull legs holds up better than a classic body part split once you're training four or more days a week.

How many days a week should you run push pull legs?

Running push pull legs three days a week is common: one push session, one pull session, one leg session, and the week is done. Our recommendation is at least four days, and preferably five or six. We'd put six first, five on a rolling schedule close behind, and we'd treat four as the floor. Below four we wouldn't recommend this split at all, because each muscle group comes round too rarely to justify the structure, and you'd get more out of something else.

On six days you run the full rotation twice: push, pull, legs, push, pull, legs, rest. Every muscle group is trained twice a week and your weekly sets for each one are spread across two sessions. On five or four days you run the same three-day rotation continuously without lining it up to a calendar week, so your training days drift and each muscle group comes round roughly every four days on five, every five on four. On three days you run the rotation once, and every muscle group gets trained once a week.

That last version is the one to think hard about.

I ran this split myself through the winter of 2019, on the rolling version: five sessions a week, no fixed days, picking the rotation up wherever I had left it. Aiming for five, I averaged four and a bit, which is the floor rather than the target, and you feel the difference: the rotation stretches, and everything comes round a day or two later than you meant it to. Covid finished it off. I lost gym access in March 2020, a couple of kettlebells became my home gym, and there was no longer any reason to give pulling a day of its own. If you end up in the same position, we have a whole article on kettlebell workouts.

Why three-day push pull legs runs short on volume

Ten hard sets per muscle group per week is where most volume recommendations start, and the meta-analysis evidence has growth still climbing above that. Now work out where those sets have to go on a three-day rotation.

Every set you'll ever do for your chest happens on push day. So does every set for your shoulders, and every set for your triceps. Ten weekly sets for chest, ten for shoulders and six for triceps, which is fewer because they already work on every press, is 26 working sets in one session, before a single warm-up. Nobody holds decent pressing quality through 26 sets.

So people compress. Four sets of bench, three of incline, three of shoulder press, three of lateral raises, three of pushdowns. That's a normal-looking push session, and it gives your chest seven sets and your shoulders six for the entire week: around two thirds of the target, delivered once. That's three-day push pull legs as it's actually run, rather than as it's written down.

What the research says about training frequency

Training a muscle more often does work, and the research is fairly clear about why: more sessions is how you fit in more sets.

Grgic and colleagues' 2018 meta-analysis in Sports Medicine found strength effect sizes rising from 0.74 to 1.08 as training frequency went from once to four or more times a week. But their volume-equated subgroup, where every group did the same number of weekly sets regardless of how many sessions those sets were spread over, found no significant effect. Schoenfeld, Grgic and Krieger's 2019 meta-analysis in the Journal of Sports Sciences pooled 25 studies and reached the same conclusion for muscle growth, and Pelland and colleagues confirmed it in a 2026 meta-regression in Sports Medicine covering 67 studies and 2,058 participants.

Frequency, then, is mostly a delivery mechanism for volume, and that's exactly why once a week is a problem. Volume is what drives the result: Schoenfeld, Ogborn and Krieger's 2017 meta-regression put each additional weekly set per muscle group at roughly 0.37% more growth, a graded dose-response with no obvious cliff in the range studied. So a muscle trained once a week has to take its entire weekly dose in one sitting to match the same muscle trained twice. Ten sets in one session instead of five and five. Twenty instead of ten and ten. Nobody runs that session, and that's what makes three-day push pull legs impractical.

A six-day push pull legs program

Run these six sessions in order, then take a rest day. Each pattern gets a heavy day and a volume day, and the main lift is the same on both: bench on both push days, the barbell row on both pull days, squat on both leg days. That lift is the one you're trying to add weight to, and one attempt a week isn't many. What changes between the two days is the load, the rep range and the accessory work around it. Legs run the other way round to push and pull, volume on day 3 and heavy on day 6, so the heavy squat session never lands the day after the deadlift.

That heavy and volume pairing is the strongest reason to want six days in the first place. Run it on four or five days and it still holds, it just arrives less predictably, because the days drift across the calendar week.

Day 1: Push (heavy)

Day 2: Pull (heavy)

Day 3: Legs (volume)

Day 4: Push (volume)

Day 5: Pull (volume)

Day 6: Legs (heavy)

Count the weekly sets and chest lands at 14, back at 20 across the rows, pulldowns and the deadlift, quads at 17, side delts at 6 direct plus everything they pick up from pressing, and rear delts at 6 from the face pulls and flyes. That's in the range the volume research supports, and you got there without any session running past about an hour.

If you can't train six days, run the same six sessions on a rolling schedule: train four or five days a week, rest whenever you need to, and pick the rotation up where you left it. Your training days drift across the calendar week rather than landing on the same days each week, and each muscle group comes round every four or five days instead of on fixed ones.

If you only have three days, use a different split

I would not run push pull legs on three days a week. I would run full body.

Three full-body sessions train every muscle group three times a week, and because the compound lifts overlap so much, you can cover the whole body in five or six exercises. The volume per muscle group per session is lower, but you get three doses instead of one, and the weekly total comes out ahead of what a single push day can realistically hold. There's a full write-up of that approach, including an A-B-A program, in the 3-day workout split article.

The other advantage is what happens when life interferes, which brings us to the part of this split nobody puts in the program table.

What to do when you miss a session

On six-day push pull legs, push the rotation forward instead of skipping. Miss Wednesday's leg day and Thursday becomes leg day, Friday becomes push, and everything shifts by one. You lose a day, not a muscle group. Because you're training each group twice a week, you'll be back to it within a few days regardless.

On three-day push pull legs, a missed session costs you far more, because all three slots in your week are already spoken for. Miss Wednesday's pull day and you can't move it to Thursday: Friday is leg day, Monday is push day, and the next free pull slot is the following Wednesday. Your back ends up going close to two weeks between sessions. That's the single most practical argument against the three-day version, and it has nothing to do with physiology. It's a scheduling fragility, and the arithmetic of a once-a-week rotation is what creates it.

How to choose exercises for each day

A handful of rules do most of the work here.

Lead with the heaviest compound movement while you're fresh. Bench press first on push day, row or deadlift first on pull day, squat first on leg day. These lifts are where most of your strength progress is measured, and they're the ones that suffer most when you do them tired.

Don't add much direct tricep work to a push day. Your triceps already did about ten sets of hard pressing before you reach the pushdowns. Two or three sets of direct work at the end is plenty. The same logic applies to biceps on pull day.

Keep the heavy deadlift away from the heavy squat. Pull day always lands the day before leg day, so the deadlift and the squat are never far apart. Put the deadlift on the pull day that comes before your volume squat session, and let the heavy squat day follow a pull day with no deadlift in it. Romanian deadlifts, which are more of a hamstring movement, are light enough on the back to sit on leg day.

Cover the front and the back of the lower body. Leg day isn't just quads. Hamstrings and glutes need their own movements, which is why a hinge and a leg curl sit alongside the squats and the quad work.

Train side delts directly. Pressing hits the front delts hard and the side delts much less. If your shoulders are a priority, lateral raises are the one isolation exercise on push day worth protecting.

If you train at home with limited equipment, the structure survives substitutions. Dumbbell press for bench press, goblet or split squats for barbell squats, single-arm rows for barbell rows. What matters is that you can load the movement progressively over time.

How to progress on push pull legs

Push pull legs gives you a structure. It doesn't give you progression, and progression is what actually produces results. For a fuller treatment of this, see the progressive overload guide.

The method I would use is double progression. Pick a rep target range, say 3 sets of 8 to 10. Start at a weight where you can hit 8 on all three sets. Add reps week by week until you hit 10 on all three. Then add weight, usually 2.5 kg on a barbell lift or the smallest jump available on dumbbells, and go back to 8. Repeat.

This works only if you know exactly what you did last time. Not roughly, exactly. Six sessions a week across five exercises each is 30 exercise slots to remember, and nobody remembers 30 slots. This is the problem RepCount was built to solve: it shows what you lifted in the previous session as you start each exercise, so the decision about what to load is already made for you.

A few things that help specifically on this split:

• Build each of the six days as a saved routine. RepCount supports custom routines, so the heavy push day and the volume push day exist as separate templates and you just open the right one.
• Watch the estimated 1RM chart on your main lifts, not the session-to-session numbers. Progress on a six-day split isn't linear week to week. RepCount charts estimated 1RM over time, which smooths out the noise. Our 1RM calculator does the same thing for a single set if you want to check the math.
• Count your weekly sets per muscle group once, when you build the routines. You don't need to track it every week. You need to know that the program you wrote down actually delivers the volume you intended, and that's an easy thing to never check.

If your numbers stop moving across several weeks on this split, the cause is usually volume or recovery rather than the split itself. There's a diagnostic walkthrough in why your training has stalled.

Is push pull legs worth it?

If you're training five or six days a week and you want a structure that keeps related muscle groups together and gives you room for arms and shoulders, yes. It's a good split, and the six-day version does what its reputation says it does.

If you're training three days a week, the split is working against you, and the version being recommended isn't the version that earned the reputation. Train full body instead and come back to push pull legs when your schedule supports it.

RepCount is a free workout tracker for iOS and Android. Build your six push pull legs days as routines, and it shows you what you lifted last time so you always know what to aim for.

Frequently asked questions

How many days a week should you do push pull legs?

Three days a week is common, but our recommendation is at least four, and preferably five or six. We'd put six first, because each muscle group then gets trained twice a week and your weekly sets are spread across two sessions instead of one, with five days on a rolling schedule close behind and four as the floor. Three days a week trains each muscle group only once, and one session rarely holds enough sets to hit the 10 or more per muscle group that most volume recommendations start at.

Can you do push pull legs 3 days a week?

You can, but it's the weakest version of the split. On three days, every set you do for your chest happens on one day, and a single push session also has to cover shoulders and triceps. It's easy to end up with six or seven sets per muscle group for the whole week, which is below what the volume research supports. On three days a week, full body sessions fit more useful work into the same time.

Is push pull legs good for beginners?

It's a reasonable structure for a beginner training six days a week. Below that, a beginner will get more practice on the main lifts, and more weekly sets per muscle group, from full body or upper/lower sessions. Four days is our floor for the split, but a floor is not a recommendation for someone new to lifting: push pull legs makes more sense once you're training five or six days and want somewhere to put the extra volume.

How many sets per muscle group should you do on push pull legs?

Ten or more weekly sets per muscle group is the usual starting recommendation, and the meta-analysis evidence shows gains continuing above that with diminishing returns. On a six-day push pull legs split that means roughly five to eight sets per muscle group per session, split across two sessions a week. On a three-day version you would need all ten or more in one session, which is what makes that version hard to run properly.

What happens if you miss a session on push pull legs?

On a six-day split, push the rotation forward rather than skipping the day. You will hit that muscle group again within a few days, so the cost is small. On a three-day split, all three slots in your week are already taken, so the missed session waits until the same day next week and that muscle group ends up going close to two weeks between sessions. That fragility is one of the strongest arguments against the three-day version.

Is push pull legs better than upper/lower?

Neither is better in itself. Upper/lower fits four days a week well and trains everything twice. Push pull legs fits five or six days and gives you more room per session for shoulders, arms and isolation work. Pick the one that matches the number of days you'll genuinely train, rather than picking the split first and forcing your week around it.