Push Pull Legs vs Upper Lower (The Answer Is Your Calendar, Not the Split)

Neither split builds more muscle than the other. Once the weekly hard sets are equal, push/pull/legs and upper/lower produce the same result, and the argument you've been reading online is about something else entirely. What actually decides it is a number you already know: how many days a week you can train. Three days, run upper/lower. Four days, it's close enough to be a coin flip and upper/lower still edges it. Five or six days, push/pull/legs distributes the work better. And there is one setup to avoid, which is unfortunately the one most guys with a job end up running: a 3-day push/pull/legs, because it trains every muscle exactly once a week. My position on this is simple: adherence beats program. Hit each muscle often enough and the exact split barely matters.

Is push pull legs or upper lower better for building muscle?

Neither. Once your weekly hard sets per muscle are equal, the name on the split stops doing work.

The cleanest evidence is Schoenfeld, Grgic and Krieger's 2019 meta-analysis in the Journal of Sports Sciences — 25 studies comparing higher against lower training frequency. On a volume-equated basis, the result was no significant difference in muscle growth. Their separate meta-regression of the studies that didn't equate volume did favor training more often, by a margin the authors themselves called modest — which is extra sets showing up as extra days. Their own recommendation is the part worth reading twice. They tell you to choose your frequency "based on personal preference." That is a meta-analysis of 25 studies concluding, in print, that you should pick the one you like.

What that leaves standing is volume. The 2017 dose-response meta-analysis from the same group put roughly 0.37% more growth on each additional weekly set across 34 treatment groups, which is the whole reason a split exists — it's a container for sets, not a stimulus of its own. So the real question isn't PPL versus upper/lower, it's how often you should train each muscle group and whether your split delivers that. Two containers, same contents, same result.

Why does every comparison pit 6-day PPL against 4-day upper/lower?

Because that's a rigged comparison, and once you see it you can't unsee it. Every "PPL vs upper/lower" article on page one lines up a six-day push/pull/legs against a four-day upper/lower. That's not one split against another. That's six training days against four, with the split names painted on top. Of course the six-day version looks bigger. It got two more sessions.

Run the day count honestly and it collapses. Saric and colleagues, 2019, in the Journal of Strength and Conditioning Research, took 27 resistance-trained men and gave them the same total weekly volume across either 3 sessions or 6 sessions per week for six weeks. Upper-body strength went up 4% on three days and 6% on six days. Lower-body strength went up 22% on three days and 18% on six days — the three-day group won that one. Elbow-extensor thickness went up 14% versus 11%, again favoring three days. None of the between-group differences were statistically significant.

Be honest about what that study is: six weeks, 27 men. It's small and it's short, and I wouldn't hang a training philosophy on it alone. It also isn't alone. Colquhoun and colleagues, 2018 ran the same 3-versus-6 design in 28 trained men with volume and intensity equated and found the same nothing: squat 1RM up 16.8 kg versus 16.7 kg, fat-free mass up 1.7 kg versus 2.6 kg, no group differences on anything measured. Two small studies pointing the same way as the 25-study pooled data is the honest state of the evidence — and it's the thing the popular version of this argument never does, which is hold volume constant while changing the day count. Every version you've read changes both at once and then credits the split.

So how do you actually pick? Count your training days.

Start with the number of days you'll genuinely show up, not the number you'd like to. Then the split follows from it, because a split's only job is fitting your weekly sets into the days you have.

• 3 days: upper/lower. Alternate them — upper, lower, upper one week, lower, upper, lower the next. Every muscle gets trained roughly three times every two weeks. A 3-day PPL, by contrast, gets each muscle once a week. Same gym time, very different frequency.
• 4 days: coin flip, and upper/lower edges it. Two uppers and two lowers land every muscle twice, every week, on the same days. Push/pull/legs across four days either rolls off the calendar or turns into a hybrid (more on that below).
• 5-6 days: push/pull/legs. Now PPL is the coherent choice. Six days is two full rotations, so every muscle gets hit twice a week; five days rolls off the calendar and lands each muscle about every fourth day. Either way the sessions stay short because each one owns three or four muscles — push takes chest, shoulders and triceps, pull takes back and biceps, and the third day is everything below the waist — and the movement pattern per day stops overlapping the day before it.

The thing that changes with day count isn't the muscle, it's the session length. Six days of PPL means six short sessions. Three days of upper/lower means three long ones — and if that pushes you past the point where you start cutting sets to get out of there, how long a workout should be is the constraint that decides your split, not your preference.

One more filter, and it's the one people skip: the split you can complete after a bad week at work beats the one that looks better on paper. If day five and six are the ones that die first, you don't run a six-day program. You run a four-day program that gets finished.

Why is a 3-day push pull legs the one version to avoid?

Because it trains every single muscle exactly once a week — the one configuration the frequency evidence argues against, and the version most guys with a full-time job drift into without noticing.

Schoenfeld, Ogborn and Krieger's 2016 meta-analysis in Sports Medicine pooled 10 studies on training frequency. Higher frequency produced an effect size of 0.49 ± 0.08 against 0.30 ± 0.07 for lower frequency, P = 0.002. Their volume-equated read of the one-to-three-day range landed in the same place: train a major muscle group at least twice a week to maximize growth.

Now hold that next to the 2019 paper above, which found no frequency effect at all. Those two results are not identical, and I'm not going to pretend they are. Two things separate them. The 2016 pooled effect size includes studies that didn't hold weekly volume constant, so some of that 0.49 is extra sets rather than extra days. And the 2019 pool covers the whole frequency range, including two-versus-three and three-versus-four comparisons where the signal genuinely disappears. The 2016 result is specifically about the once-a-week end, which is where the evidence is most consistent — and neither paper has once a week coming out ahead of twice. That's the asymmetry that matters. Going from once to twice might do nothing, or it might be worth what the 2016 numbers say. It has never been worth less.

Then there's the practical failure. On a 3-day PPL, missing one session means that muscle group gets nothing for two weeks. On a 3-day upper/lower, a missed session still leaves the other two covering everything you own.

Does the split change how strong you get, or just how big?

It can change strength more than size, and that's the one place where day count earns real credit.

Pelland and colleagues' 2026 meta-regressions in Sports Medicine are the biggest read on this so far — 67 studies, 2,058 participants. On hypertrophy, the effect of frequency was compatible with negligible: more sessions per week, holding sets constant, doesn't reliably grow you more. On strength, the same analysis put the posterior probability of a frequency effect at 100%, with clear diminishing returns as sessions pile up.

That fits how strength works. Strength is partly a skill, and you get better at a lift by doing the lift more often. So a six-day PPL that has you pressing twice a week may well move your bench faster than a three-day setup — with the returns shrinking each time you add a session. If your goal is a bigger squat number specifically, frequency is a lever. If your goal is looking like you lift, it mostly isn't.

What I run at 37 with a full-time job, and how you'd know it's working

I lift four days a week: push, pull, legs, upper. That's upper/lower's frequency wearing push/pull/legs' exercise order — my upper body gets trained twice every week, my legs get one hard session, and the day names still tell me what I'm doing when I walk in. I've written out the exact 4-day push/pull/legs/upper split elsewhere, so I won't re-spec it here.

I'll be straight about the limits of that. Four days is what I run and the only version I can speak to first-hand, so I'm not going to hand you a verdict on a six-day program from the outside — and there's no dramatic switching story here, no burnout arc, no "and then I found the perfect split." I'm 37, I work full time as a software engineer, and four days is simply what my calendar supports, so four days is what I built around. That's not a research finding. It's a constraint, and it's the same constraint the day count is supposed to respect.

The part that isn't preference: what happens inside those sessions. 2-3 hard sets per exercise, taken close enough to failure that the last rep is genuinely in doubt. Two near-failure sets beat four junk sets, and no split rescues four junk sets.

Which brings you to the only signal that tells you whether any of this is working. Not soreness, not the pump, not whether the split has a name people respect. Your logbook trend over 6-8 weeks. Pick the split your days support, log every set, and look at whether the weight or the reps at a given weight are climbing. If they are, you picked right and the debate is over. If they're flat for a month and you didn't miss sessions, the fix is more sets or harder sets — changing the letters on your split will do nothing. That logbook is the whole job, and no split hands you one. The Momentum Method is that four-day layout written as a day-one starting standard rather than my personal program — the set-and-rep spec and the daily scorecard attached, so the measuring isn't a separate thing you have to remember to do.