The Best Training Split for Your Goals | Complete Guide
For most people the best training split is an upper and lower body split run four days a week, or full body work if you only have two or three days. That is the practical answer, and the more interesting one sits underneath it: when weekly volume is held constant, the split you choose has almost no measurable effect on how much muscle or strength you build.
A meta-analysis of 14 studies and 392 participants compared split routines against full body routines and found no significant difference in bench press strength, lower limb strength, elbow extensor size, elbow flexor size, vastus lateralis size or lean body mass (Ramos-Campo and colleagues, Journal of Strength and Conditioning Research, 2024, volume 38, pages 1330 to 1340). So the question worth answering is not which split is best, but which one lets you fit the right amount of work into the days you actually have.
The main training splits compared
Split | Sessions per week | Times each muscle group is trained per week | Direct sets per major muscle group per week | Typical session length | Who it suits |
|---|---|---|---|---|---|
Full body | 2 to 4 | 2 to 4 | 6 to 12 | 45 to 70 minutes | Two to four days available, newer lifters, shift workers, anyone who misses sessions often and wants every muscle covered each time |
Upper and lower | 4 (also runs as 2, 3 or 6) | 2 | 10 to 16 | 60 to 75 minutes | Most people training three to four days. Enough volume per muscle without long sessions |
Push, pull, legs | 3 or 6 | 1 on a three-day week, 2 on a six-day week | 6 to 9 on three days, 12 to 20 on six days | 60 to 90 minutes | People who can reliably train five or six days and want high volume per muscle group |
Body part split | 4 to 6 | 1 | 12 to 20, all in one session | 60 to 90 minutes | Experienced lifters who prefer long sessions on one or two muscle groups and train five or more days |
Source and derivation: the sessions and frequency columns describe how each split is conventionally run. The set and session length columns are arithmetic, not measurement, assuming 15 to 20 hard working sets per session and 2 to 3 minutes per set including rest. Direct sets count only work targeting the named muscle group, following the direct and indirect classification used by Pelland and colleagues (Sports Medicine, 2026, volume 56, pages 481 to 505). No study has measured session length across these formats, so treat that column as a planning estimate.
The honest answer: split choice barely matters
The Ramos-Campo meta-analysis is the most direct evidence on this question. Mean differences were near zero across every outcome and none reached significance: bench press 1.19 kg (95% CI minus 1.28 to 3.65, p equals 0.34), lower limb strength 2.47 kg (95% CI minus 2.11 to 7.05, p equals 0.29), lean body mass minus 0.07 kg (95% CI minus 1.59 to 1.44, p equals 0.92). Cross-sectional area at the elbow extensors, elbow flexors and vastus lateralis behaved the same way.
Two limits. Most included studies used untrained or recreationally trained participants, and most ran 8 to 12 weeks. Long enough to detect a moderate difference, not a small one accumulating over years. Split structure is not a large lever, which is not the same as zero.
How often should you train each muscle group?
Schoenfeld, Ogborn and Krieger meta-analysed ten studies in Sports Medicine in 2016 (volume 46, pages 1689 to 1697). Using frequency as a binary predictor, higher frequency produced a larger hypertrophy effect size than lower frequency, 0.49 plus or minus 0.08 against 0.30 plus or minus 0.07, p equals 0.002. Their conclusion was widely repeated: train each major muscle group at least twice a week. They noted they could not tell whether three times beat twice.
The same group revisited it three years later with 25 studies and reached a different conclusion (Schoenfeld, Grgic and Krieger, Journal of Sports Sciences, 2019, volume 37, pages 1286 to 1295). On a volume-equated basis there was no significant difference between higher and lower frequency. That held for direct measures of growth, for resistance-trained participants, and for upper and lower body separately. Higher frequency did win in studies that had not equated volume, though the gap between one day and three or more was described as modest. Their conclusion: individuals can choose a weekly frequency per muscle group based on personal preference.
The most recent analysis agrees. Pelland and colleagues meta-regressed 67 studies and 2,058 participants and found the posterior probability of a frequency effect on hypertrophy was less than 100%, which they read as compatible with negligible effects (Sports Medicine, 2026, volume 56, pages 481 to 505).
Strength is a slightly different story. Grgic and colleagues analysed 22 studies and found effect sizes rising with frequency: 0.74, 0.82, 0.93 and 1.08 for one, two, three and four or more sessions a week, p equals 0.003 (Sports Medicine, 2018, volume 48, pages 1207 to 1220). Their volume-equated subgroup showed no significant effect, p equals 0.421, so the advantage was largely volume in disguise. Ralston and colleagues found the same across 12 studies, effect size 0.03 (95% CI minus 0.20 to 0.27, p equals 0.78), with upper body strength the one exception (Sports Medicine Open, 2018, volume 4, article 36).
Pelland's group is the dissent. For strength they found a 100% posterior probability that gains increase with frequency, with diminishing returns. So for maximal strength, spreading the same work across more sessions may be worth a little. For muscle size, it is not.
Frequency is a scheduling tool. It sets how much volume fits into a session before the last sets fall apart, and not much else.
How much volume per muscle group per week
Schoenfeld, Ogborn and Krieger meta-regressed 34 treatment groups from 15 studies and found a significant effect of weekly sets on muscle size, p equals 0.002 (Journal of Sports Sciences, 2017, volume 35, pages 1073 to 1082). Each additional weekly set was associated with an increase in effect size of 0.023, corresponding to a 0.37 percentage point increase in muscle growth. Comparing higher and lower volume conditions within each study, the effect size difference was 0.241, equating to 3.9 percentage points more growth.
One caveat gets dropped from most summaries. When the same authors split weekly sets into three bands, fewer than five, five to nine, and ten or more, the effect only reached a trend, p equals 0.074. The graded relationship is well supported. The band thresholds are less firmly established than they are usually made to sound.
Pelland and colleagues confirmed the direction with more data: across 67 studies the posterior probability that hypertrophy increases with volume was 100%, though their models showed diminishing returns. They counted indirect sets at half value, so your pressing partly counts towards triceps and your rowing towards biceps.
A defensible target is ten to sixteen hard direct sets per major muscle group per week, taken close to failure. Look back at the first table. A three-day full body week reaches the bottom of that band and no further. Four days upper and lower sits inside it. Six-day push, pull, legs can exceed it. That is the real difference between splits: capacity, not quality.
How long between sessions for the same muscle
MacDougall and colleagues measured muscle protein synthesis after heavy resistance exercise in six young men using a labelled leucine infusion (Canadian Journal of Applied Physiology, 1995, volume 20, pages 480 to 486). Synthesis rose by 50% at four hours and 109% at 24 hours, and by 36 hours had returned to within 14% of the untrained control arm, a non-significant difference. Six participants and a dated method, so hold it loosely, but it is the origin of the 48-hour convention.
How hard you train changes that floor. Moran-Navarro and colleagues put ten resistance-trained men through three protocols, tracking recovery out to 72 hours (European Journal of Applied Physiology, 2017, volume 117, pages 2387 to 2399). Sets taken to failure produced a larger acute drop in jump height and bar velocity than the same total volume in shorter sets, and recovery between 24 and 48 hours was significantly slower. Creatine kinase followed the same pattern.
So the gap depends on what you did. Two or three sets stopped a repetition short of failure can be repeated after about 48 hours. Sixteen sets taken to failure cannot, which is part of why a body part split leaves a full week between sessions. That is the design responding to the load, not a flaw in it.
How many days a week should you lift?
Pick the number you can hold for a year, then choose the split that fits it. Two days is enough to make real progress with full body sessions. Three to four covers most people. Five or six is only better if you genuinely attend, because the extra volume is the whole benefit.
There is reasonable evidence that asking for more frequency is not what breaks adherence. Perri and colleagues randomised 379 sedentary adults to walk 30 minutes a day at either 3 to 4 or 5 to 7 days per week, at moderate or higher intensity, over six months (Health Psychology, 2002, volume 21, pages 452 to 458). Higher frequency increased total exercise completed, p equals 0.0001, with no decline in the percentage of the prescription completed. Higher intensity did the opposite: adherence dropped, p equals 0.02, and less exercise was completed.
That trial used walking, so applying it here is an inference rather than a finding. Still, more frequent, shorter, more moderate sessions held up better than fewer and harder ones. A qualitative systematic review of 35 studies on lifestyle intervention adherence agrees, naming self-regulation skills, social support and personalised goals as the recurring facilitators (Deslippe and colleagues, International Journal of Behavioral Nutrition and Physical Activity, 2023, volume 20, article 14).
Four sessions of 60 minutes are more likely to survive a busy month than three of 90.
Three weekly schedules
Worked examples, not prescriptions. Counts are working sets only, excluding warm-ups, every set taken to within one or two repetitions of failure.
Three days a week, full body
Day | Session | Main work | Working sets |
|---|---|---|---|
Monday | Full body A | Back squat 3, bench press 3, chest-supported row 3, Romanian deadlift 3, lateral raise 2, triceps extension 2 | 16 |
Tuesday | Rest or easy walking | None | 0 |
Wednesday | Full body B | Trap bar deadlift 3, overhead press 3, lat pulldown 3, leg press 3, biceps curl 2, face pull 2 | 16 |
Thursday | Rest or easy walking | None | 0 |
Friday | Full body C | Split squat 3, incline press 3, seated cable row 3, leg curl 3, lateral raise 2, triceps extension 2 | 16 |
Saturday and Sunday | Rest | None | 0 |
Source: constructed example. Set counts are arithmetic, and the volume target is drawn from Schoenfeld, Ogborn and Krieger, Journal of Sports Sciences, 2017.
Weekly direct sets: chest 6, back 9, quadriceps 9, hamstrings 6, shoulders 4 direct plus pressing, arms 6. That sits at the lower end of the useful band. If three days is what you have, add a set to the compound lifts rather than more exercises, and expect about 60 minutes.
Four days a week, upper and lower
Day | Session | Main work | Working sets |
|---|---|---|---|
Monday | Upper A | Bench press 4, barbell row 4, overhead press 3, lat pulldown 3, lateral raise 2, biceps curl 2, triceps extension 2 | 20 |
Tuesday | Lower A | Back squat 4, Romanian deadlift 3, leg press 3, leg curl 3, calf raise 3 | 16 |
Wednesday | Rest | None | 0 |
Thursday | Upper B | Incline press 4, chest-supported row 4, machine chest press 3, cable pulldown 3, rear delt fly 2, biceps curl 2, triceps extension 2 | 20 |
Friday | Lower B | Hip hinge 3, Bulgarian split squat 3, leg extension 3, leg curl 3, calf raise 3 | 15 |
Saturday and Sunday | Rest | None | 0 |
Source: constructed example, built to the same volume target.
Weekly direct sets: chest 11, back 14, quadriceps 13, hamstrings 9, shoulders 7 direct plus pressing, arms 8. Every major muscle group lands inside the ten to sixteen band, sessions run 60 to 75 minutes, and each muscle gets at least 48 hours before it is trained again. That is why this is the default.
Five days a week, upper, lower, push, pull, legs
Day | Session | Main work | Working sets |
|---|---|---|---|
Monday | Upper | Bench press 4, barbell row 4, overhead press 3, lat pulldown 3, biceps curl 2, triceps extension 2 | 18 |
Tuesday | Lower | Back squat 4, Romanian deadlift 3, leg press 3, leg curl 3, calf raise 3 | 16 |
Wednesday | Rest | None | 0 |
Thursday | Push | Incline press 4, seated shoulder press 3, cable fly 3, lateral raise 3, triceps extension 3 | 16 |
Friday | Pull | Weighted pull-up 4, chest-supported row 4, cable row 3, rear delt fly 3, biceps curl 3 | 17 |
Saturday | Legs | Front squat 4, Bulgarian split squat 3, leg curl 4, leg extension 3, calf raise 3 | 17 |
Sunday | Rest | None | 0 |
Source: constructed example, built to the same volume target.
Weekly direct sets: chest 11, back 18, quadriceps 17, hamstrings 10, shoulders 12 direct, arms 10. Back and quadriceps now sit at the top of the band, and sets above about sixteen are worth less than the ones below. Five days buys volume, not a better split.
A body part split runs on the same arithmetic: chest Monday, back Tuesday, legs Wednesday, shoulders Thursday, arms Friday puts 12 to 20 sets into one session per muscle group and leaves seven days before you return. It builds muscle comparably when total weekly volume matches. The cost is that a missed Tuesday means your back gets nothing that week, which is the practical argument against it rather than a physiological one.
Where your own data fits
Every number above is a population average, and the recovery research shows how much the right gap between sessions depends on how hard you trained and how well you slept. Your heart rate variability, sleep and session-to-session strength can be measured rather than guessed at. hlth.coach reads your wearable and body composition data and adjusts training, recovery and nutrition against what your body is doing that week, which is a more useful input than a fixed template.
The short version
Split structure is not a meaningful variable. Fourteen studies and 392 people showed no significant difference between split and full body routines when volume was equated.
Nor is frequency, for muscle size. The 2019 analysis of 25 studies found no difference once weekly volume was matched, revising the twice-a-week advice the same authors gave in 2016.
For maximal strength, higher frequency may add a small amount. The evidence is mixed and the volume-equated subgroups mostly show no effect.
Weekly volume does matter. Each extra set was associated with 0.37 percentage points more growth, with the curve flattening. Aim for ten to sixteen hard direct sets per major muscle group.
Leave about 48 hours between sessions for the same muscle, longer if every set went to failure.
Choose by days you can reliably attend: two to three suits full body, three to four upper and lower, five to six push, pull, legs or a body part split.
More frequent sessions did not reduce adherence in a 379-person randomised trial. Harder ones did.
Common questions
What is the best training split?
The one you complete consistently. Once weekly volume and frequency per muscle group are matched, split choice makes little difference to muscle growth. The comparison table above sets out the realistic options by how many days a week you can actually train.
Is push pull legs better than upper lower?
Not inherently. Upper lower reaches most muscle groups twice a week on four days. Push pull legs does the same on six days, or once a week on three. Pick the one that fits your week rather than the one with the better reputation.
How many days a week should I lift?
Three to five suits most people, and three well-executed sessions beat five half-finished ones. The deciding factor is how many days you can hold across a busy month, not the best week you have ever had.
How often should I train each muscle group?
Training a muscle roughly twice a week is a sensible default when volume is equated, which is what drives the popularity of full body and upper lower splits at lower training frequencies.
Is a body part split effective?
It can be, though it usually means training each muscle once a week, which makes weekly volume harder to accumulate and each session longer. It suits high training frequency and plenty of time, not a three day week.
Can I change my split?
Yes, and switching when your schedule changes is better than abandoning training because the split no longer fits. Keep weekly volume roughly stable through the change rather than resetting.
Sources
Ramos-Campo DJ, Benito-Peinado PJ, Andreu-Caravaca L, Rojo-Tirado MA, Rubio-Arias JA. Efficacy of split versus full-body resistance training on strength and muscle growth: a systematic review with meta-analysis. Journal of Strength and Conditioning Research, 2024, 38(7), 1330 to 1340.
Schoenfeld BJ, Ogborn D, Krieger JW. Effects of resistance training frequency on measures of muscle hypertrophy: a systematic review and meta-analysis. Sports Medicine, 2016, 46(11), 1689 to 1697.
Schoenfeld BJ, Grgic J, Krieger J. How many times per week should a muscle be trained to maximize muscle hypertrophy? A systematic review and meta-analysis of studies examining the effects of resistance training frequency. Journal of Sports Sciences, 2019, 37(11), 1286 to 1295.
Schoenfeld BJ, Ogborn D, Krieger JW. Dose-response relationship between weekly resistance training volume and increases in muscle mass: a systematic review and meta-analysis. Journal of Sports Sciences, 2017, 35(11), 1073 to 1082.
Pelland JC, Remmert JF, Robinson ZP, Hinson SR, Zourdos MC. The resistance training dose response: meta-regressions exploring the effects of weekly volume and frequency on muscle hypertrophy and strength gains. Sports Medicine, 2026, 56(2), 481 to 505.
Grgic J, Schoenfeld BJ, Davies TB, Lazinica B, Krieger JW, Pedisic Z. Effect of resistance training frequency on gains in muscular strength: a systematic review and meta-analysis. Sports Medicine, 2018, 48(5), 1207 to 1220.
Ralston GW, Kilgore L, Wyatt FB, Buchan D, Baker JS. Weekly training frequency effects on strength gain: a meta-analysis. Sports Medicine Open, 2018, 4(1), 36.
MacDougall JD, Gibala MJ, Tarnopolsky MA, MacDonald JR, Interisano SA, Yarasheski KE. The time course for elevated muscle protein synthesis following heavy resistance exercise. Canadian Journal of Applied Physiology, 1995, 20(4), 480 to 486.
Moran-Navarro R, Perez CE, Mora-Rodriguez R, de la Cruz-Sanchez E, Gonzalez-Badillo JJ, Sanchez-Medina L, Pallares JG. Time course of recovery following resistance training leading or not to failure. European Journal of Applied Physiology, 2017, 117(12), 2387 to 2399.
Perri MG, Anton SD, Durning PE, Ketterson TU, Sydeman SJ, Berlant NE, Kanasky WF, Newton RL, Limacher MC, Martin AD. Adherence to exercise prescriptions: effects of prescribing moderate versus higher levels of intensity and frequency. Health Psychology, 2002, 21(5), 452 to 458.
Deslippe AL, Soanes A, Bouchaud CC, Beckenstein H, Slim M, Plourde H, Cohen TR. Barriers and facilitators to diet, physical activity and lifestyle behavior intervention adherence: a qualitative systematic review of the literature. International Journal of Behavioral Nutrition and Physical Activity, 2023, 20(1), 14.
This article is general information and is not medical advice. Speak with your doctor or an accredited exercise physiologist before starting a new training program, particularly if you have an existing health condition or injury.