Full body, upper/lower and push/pull/legs differ in one thing: how many muscle groups share a training day. Full body puts six groups in every session, upper/lower splits the body in half, and push/pull/legs splits it in thirds. That single choice fixes how often each muscle is trained for a given number of days, which is why the question "which split?" is really the question "how many days can I train?".
This article uses the three split templates exactly as the Zyra Workouts generator defines them. The rosters, the day layout and the set counts below are the app's rules, reproduced — not a recommendation for any individual, and not a trial result.
The three splits and what each day covers
Each split is a list of day templates, and each template is a roster of muscle groups. The generator fills a day by taking one compound exercise per roster muscle first, then adds isolation work until the session's exercise count is reached.
| Split | Day templates | Roster per day | Muscles a week touches (all days) |
|---|---|---|---|
| Full body | Full Body | quads, chest, back, shoulders, hamstrings, core | 6 groups (no direct biceps, glutes, calves, forearms) |
| Upper/lower | Upper · Lower | Upper: chest, back, shoulders, biceps, triceps · Lower: quads, hamstrings, glutes, calves, core | 10 groups (no direct forearms) |
| Push/pull/legs | Push · Pull · Legs | Push: chest, shoulders, triceps · Pull: back, biceps, core · Legs: quads, hamstrings, glutes, calves | 10 groups (no direct forearms) |
Two things follow from the table. First, "no direct sets" is not "no sets": a back squat credits glutes and hamstrings as secondary muscles, and the app counts each secondary set at 0.5 toward that muscle's weekly total. Second, the full-body roster is shorter than the other two combined, so at the same number of exercises per session it spends its slots on the biggest movements and nothing else.
In plain terms: the split decides which muscles get a dedicated slot; secondary credit fills in some of the rest, but not all of it.
Frequency per muscle at 2–6 days
Frequency is how many times a week a muscle appears on a training day. Because the generator cycles through the day templates in order, frequency is a count, not a judgement:
| Days a week | Full body | Upper/lower | Push/pull/legs |
|---|---|---|---|
| 2 | every roster muscle ×2 | everything ×1 | push and pull muscles ×1; legs ×0 |
| 3 | ×3 | upper ×2, lower ×1 | everything ×1 |
| 4 | ×4 | everything ×2 | push ×2, pull and legs ×1 |
| 5 | ×5 | upper ×3, lower ×2 | push and pull ×2, legs ×1 |
| 6 | ×6 | everything ×3 | everything ×2 |
The uneven rows are worth reading twice. Push/pull/legs at 2 days never reaches a legs day, and at 4 or 5 days one or two of the thirds get less than the others. Upper/lower at 3 or 5 days gives the upper body the extra session. The generator does not rebalance this — it runs the templates in order and stops when the week is full.
Frequency is also not the same as weekly sets. Full body at 3 days trains hamstrings three times, but with only a handful of sets each time. The weekly sets calculator adds up what a whole week actually credits to each muscle, against the app's landmarks (chest 12, back 14, shoulders 14, biceps 10, triceps 10, forearms 6, quads 14, hamstrings 10, glutes 10, calves 10, core 10 working sets a week). Those landmarks are the app's dashboard defaults, not a law.
In plain terms: frequency tells you how often a muscle gets a turn; the set count tells you how much work it gets when it does.
How the week is laid out
The generator does not let you pick which weekdays to train. It spreads sessions evenly across a 7-day cycle with one line of arithmetic: training day i (counting from 0) lands on slot floor(i × 7 / days).
| Days | Slots (0 = first day of the cycle) | Pattern |
|---|---|---|
| 2 | 0, 3 | T·R·R·T·R·R·R |
| 3 | 0, 2, 4 | T·R·T·R·T·R·R |
| 4 | 0, 1, 3, 5 | T·T·R·T·R·T·R |
| 5 | 0, 1, 2, 4, 5 | T·T·T·R·T·T·R |
| 6 | 0–5 | T·T·T·T·T·T·R |
The layout is the same for every split. A 4-day full-body week and a 4-day upper/lower week both sit on T·T·R·T·R·T·R; only the roster on each T differs. When a template repeats within the week the days get letters — Upper A, Upper B — and the generator gives each lettered day the same exercise list, because selection depends only on the roster, the session length and the focus settings.
The session length sets how many exercises a day holds: clamp(round(minutes / 12), 3, 8). Thirty minutes buys 3 exercises, 45 buys 4, 60 buys 5, 75 buys 6 and 90 buys 8.
Choosing by days available, not by preference
Put the two tables together and the choice mostly makes itself:
- 2–3 days. Only the full-body roster trains every one of its muscles more than once a week. Upper/lower at 3 days leaves the lower body at once a week; push/pull/legs at 3 days leaves everything at once a week. The 3-day full body plan and 2-day plan are the generated examples.
- 4 days. Upper/lower reaches twice a week for all ten of its muscles (4-day upper/lower). Full body at 4 days reaches four times a week for six muscles with 4 exercises per 45-minute session (4-day full body). Push/pull/legs at 4 days is lopsided.
- 5 days. Both upper/lower and push/pull/legs are lopsided in different directions; full body stays even. The generator supports 5 days but none of the six program pages uses it, for that reason.
- 6 days. Push/pull/legs reaches twice a week for everything with a single rest day (6-day push/pull/legs); upper/lower reaches three times.
What the arithmetic cannot decide is whether you will keep the schedule. A 4-day plan you complete for six cycles does more than a 6-day plan you abandon in week two — not because of any study, but because a missed day is a day of zero sets.
What trials comparing splits found
Everything above is the app's arithmetic. Researchers have also tested the question directly: give two groups the same exercises and the same weekly sets, arrange them as a split or as full-body sessions, and compare what changes.
Trained men, one muscle a day vs every muscle every day. In a 2015 trial, 20 well-trained young men were matched on strength and randomised to a split (2–3 muscle groups per session, so each muscle once a week) or full body (one exercise per muscle, every muscle every session)1. Both groups trained 3 days a week for 8 weeks, with the same 21 exercises, 8–12 reps per set to failure and 90 seconds of rest; 19 finished. Muscle thickness grew more with full body in all three sites measured, but only the elbow-flexor gap was significant (3.2 vs 2.1 mm). Bench press and squat 1RM gains did not differ significantly1. The authors note a novelty effect: 16 of the 19 men normally trained with a split, so full body was the new program1.
Muscle thickness gain after 8 weeks: split vs full body, trained men · mm
Split, each muscle 1×/week Full body, each muscle 3×/week
Untrained people, equal weekly sets. A 2021 trial randomised 67 untrained people to train each muscle twice a week (an A/B split) or four times a week (full body), with equal weekly sets, for 8 weeks2. Both groups got stronger and grew; no measure differed significantly between groups. A 2022 trial in 50 untrained women compared full body twice a week with an upper/lower split four times a week, with the same exercises and weekly sets, for 12 weeks3. Again, no between-group differences: bench press 1RM rose 25.5% with full body and 30.0% with the split, leg press 29.2% and 28.3%, muscle mass 1.9% and 1.7%3.
Gains after 12 weeks: full body twice a week vs upper/lower four times · %
Full body, 2×/week Upper/lower, 4×/week
The pooled view. A 2024 meta-analysis combined 14 studies with 392 healthy adults4. Split and full-body routines produced similar gains in bench press and lower-body strength, and similar growth in the arm muscles, the thigh (vastus lateralis) and lean body mass; no difference came close to significance (p = 0.29 to 0.93). The authors concluded that, with volume equated, people can choose a routine "based on their personal preferences"4.
In plain terms: when the weekly sets are the same, the split did not decide the result in these studies. That is why this page frames the choice by days available: the split is how you fit the sets into the days you have.
What the studies found
| Study | Who was studied | Key result | Year |
|---|---|---|---|
| Schoenfeld et al., *J Strength Cond Res*1 | 19 well-trained young men, 8 weeks, split (1×/week per muscle) vs full body (3×/week) | Elbow flexors grew more with full body; other sites and strength not significantly different | 2015 |
| Evangelista et al., *Einstein*2 | 67 untrained people, 8 weeks, each muscle 2× vs 4× a week, equal sets | Similar strength and muscle thickness gains | 2021 |
| Pedersen et al., *BMC Sports Sci Med Rehabil*3 | 50 untrained women, 12 weeks, full body 2× vs upper/lower 4× a week | No between-group differences in strength, power or muscle mass | 2022 |
| Ramos-Campo et al., *J Strength Cond Res*4 | Meta-analysis, 14 studies, 392 healthy adults | Similar strength and growth with split or full body when volume is equal | 2024 |
Limits worth knowing
- Short and small. Trials ran 8 to 12 weeks with 19 to 67 people; small differences may not show up in that time.
- Volume held equal on purpose. The trials answer "same sets, different arrangement". They do not test a split that lets you do more sets, which the 2015 authors note may be a split's real advantage1.
- Push/pull/legs specifically was not a named arm in the trials we read; the comparisons were split routines in general, including upper/lower, against full body.
- Abstracts only for the 2021, 2022 and 2024 papers; the 2015 trial was read in full except its image tables. The meta-analysis abstract does not state the units of its mean differences.
Latest research
The newest pooled evidence we opened is the 2024 meta-analysis above4: 14 studies, similar strength and muscle growth for split and full-body routines when weekly volume was equated. We found no newer head-to-head trial of split versus full-body training that changes that picture. When one appears, this page will be updated and its date will change.
The short version
A split is a rule for which muscles share a day. Full body trains six groups every session; upper/lower needs 4 days to train everything twice a week; push/pull/legs needs 6. The week is laid out by floor(i × 7 / days) for every split, and the session length fixes the exercise count. Pick the split by the days you can actually train, then check the week's set totals with the weekly sets calculator. The tables are training arithmetic from the app's own rules; in trials with equal weekly sets, split and full-body routines gave similar strength and muscle growth4. Educational overview only — not medical advice.
