Training Frequency for Strength: How Many Days You Need

The question behind every training split

You have a certain number of days you can realistically get to the gym. Three, maybe four, five in a good week. Every training split answers the same question: is training frequency for strength a variable in its own right, or does four days beat three only because four days is where more sets fit?

If the number of sessions is what drives strength, then a week where life cuts you down to two days is a week you lost ground. If it is the total work that drives strength, then a two-day week where you get the sets in is a normal week, and you can build your schedule around your life instead of the other way around.

A 2018 meta-analysis in Sports Medicine is the best pooled evidence we have found, and it points at the total work — with a caveat about who was studied that is worth weighing before you rearrange anything.

What the researchers did
Frequency helps because it adds volume
Training frequency in the subgroups: read them as leads
The limitation that matters most for you
What training frequency for strength means if you already train
Common Questions About Training Frequency
How Pick It Up Helps
Further Reading & References

What the researchers did

Grgić and colleagues — a group including Brad Schoenfeld and James Krieger — pooled 22 studies comparing different resistance training frequencies and ran a random-effects meta-analysis on strength outcomes. The pooled sample was 912 participants. Individual studies ran from 11 to 152 people, with a median of 29. Programs averaged about 12 weeks, ranging from 6 to 24.

Frequency was sorted into 1, 2, 3, or 4+ sessions per week. Twenty-one of the 22 studies measured strength with a one-rep max test. The most common comparison, in 14 studies, was two sessions a week against three.

On a modified Downs and Black quality checklist the studies scored 13 to 22, averaging 18 — four rated good, the rest moderate.

Frequency helps because it adds volume

Look at all 22 studies together and frequency looks like it matters. Standardised effect sizes for strength gain climbed with the number of sessions: 0.74 at once a week, 0.82 at twice, 0.93 at three times, and 1.08 at four or more. That trend was statistically significant (p = 0.003).

Then the researchers looked only at studies that held weekly training volume constant — same total work, spread across different numbers of days. The trend vanished.

Two panels of effect sizes for strength gain by weekly training frequency. Across all 22 studies the effect sizes rise from 0.74 to 1.08 as frequency increases; among only the studies that matched weekly volume they do not rise, sitting at 0.53, 0.80, 0.64 and 0.58

In the volume-equated studies the effect sizes were 0.53, 0.80, 0.64 and 0.58 for one, two, three and four-plus sessions — no ordering, no trend, and no statistical significance (p = 0.421). The authors’ conclusion, with “RT” their shorthand for resistance training:

The higher muscular strength gains associated with higher RT frequencies are observed largely because of the additional training volume.

The confidence intervals overlap heavily on both sides of that chart, even in the panel where the trend was significant. The interval for four or more sessions a week ran from 0.74 to 1.42, and the interval for a single session ran from 0.48 to 1.01. These are not cleanly separated conditions.

That squares with what we have written about breaking through a strength plateau: the levers that move a stalled lift are the ones that change how much progressive work you accumulate, not the ones that rearrange the calendar.

Training frequency in the subgroups: read them as leads

The meta-analysis also splits its results six ways — by exercise type, upper versus lower body, training to failure, age, and sex. None of those splits is volume-equated. Volume is one of the six moderators, analysed on its own; every other subgroup is drawn from the full pool of 22 studies, mixed designs and all. Each therefore re-introduces the confound the section above removed, so a significant p value inside one tells you less than it looks.

The sex result shows why. Frequency came out significant for women (p = 0.030) and not for men (p = 0.190) — and the authors explain that as an artifact of which studies happened to exist rather than anything about women. Four of the five studies in men used a volume-equated design; only two of the eight studies in women did. The women’s pool was mostly the kind of study where more sessions also meant more total work, which is the design that manufactures a frequency effect. It is not a finding about sex.

One split comes with a mechanism rather than just a p value. Frequency mattered for one-rep max tests on multi-joint exercises (p < 0.001) but not on single-joint ones (p = 0.324). The authors read that as motor learning: complex movements “require a precise timing of muscle recruitment and coordination and a higher degree motor proficiency,” so training them more often gives “more opportunities for ‘practicing’ the test/exercise.” Squatting twice a week rather than once may be making you better at squatting, which is not quite the same thing as making you stronger. It comes from the same un-equated pool, so treat it as a lead, not a result.

Our guide to exercise anatomy for strength training covers how compound patterns recruit across muscle groups, and why a bench press asks more of you than an isolation movement.

The limitation that matters most for you

Of the 912 participants, only 56 had lifted before. Three of the 22 studies used people with previous resistance training experience; the other 19 studied untrained participants. The authors could not run a trained-only analysis: not enough studies to pool. In their words:

The most apparent limitation of the current review is the small number of studies that included trained participants. This lack of empirical evidence limits the generalizability of the findings to trained individuals.

If you already train, you are not the population this evidence describes. The three trained-lifter studies are at least consistent: all three compared once a week against three times a week with volume equated, and two of them found no meaningful difference in upper or lower body strength. The third found an advantage for three sessions on the leg press, and no difference on the other eight one-rep max tests it ran. The longest of the three lasted 12 weeks.

Two further cautions:

The authors mention an unpublished 15-week study of 16 Norwegian powerlifters in which six volume-equated sessions a week beat three on squat and bench. They flag it as unpublished and not yet scrutinised. So do we — it points the other way, and is not evidence to act on yet.
In studies this short, on people this untrained, part of what a one-rep max test measures is skill at the test. Practising the test itself produces strength gains comparable to conventional training in that population, which is a reason to expect the effect sizes above to shrink in people who already know how to grind out a heavy single.

None of this makes the finding useless. It makes it a strong prior rather than a settled answer, and the prior points toward a flexible calendar.

What training frequency for strength means if you already train

Your weekly volume is the training variable. The number of days is logistics.

If you currently do 16 hard sets for your upper body across four sessions and your schedule collapses to three, this meta-analysis suggests that moving those 16 sets into three sessions is closer to a lateral move than a loss. What would cost you is doing 12 sets instead of 16 because one day disappeared.

The researchers name one limit. Beyond some point a session gets long enough that fatigue degrades the last sets in it — the fifth set of squats is not the same stimulus as the first. Their words, with RT again standing for resistance training:

Higher RT frequencies allow a distribution of training volume throughout the week while keeping the performance on each RT session high.

That is a different argument from “more sessions build more strength.” More days protects the quality of your sets once weekly volume gets high enough that one or two sessions cannot hold it. Nobody has established where that threshold sits for a given lifter, and this meta-analysis did not try to. What that leaves you:

Set your weekly volume first, then divide it. Pick It Up programmes 10–15 hard sets per muscle group per week, which is where the volume research points. More sets do keep adding gains, but the returns flatten out past roughly 15 sets a week for most people: Stronger By Science’s reviews of the dose-response literature and of more recent volume work put a doubling of volume at roughly 50% more growth, which makes the jump from 15 sets to 30 a poor trade for the hours it costs. The split serves the number, not the reverse.
Protect the number when your week breaks. Move the sets when you can, and skip them when you cannot find another day that week. It is consistency week to week that matters, not any one day or even any one week.
Add a day when session quality starts dropping, not on principle. Last two sets of a session consistently well below the first two is the point where spreading the work out earns its keep, and it is the mechanism the researchers propose.
Give compound lifts more exposure than isolation work. The clearest frequency effect was on multi-joint one-rep max tests, and the authors put that down to practice at the movement. Squatting twice a week rather than once costs you nothing if the weekly sets are the same.
Do not overhaul a program that is working. If you are progressing on a four-day split, this research is not a reason to change it. It is a reason to stop treating the number four as the thing that is working.

Common Questions About Training Frequency

Does training frequency matter more for upper body? It did in this analysis — upper body strength showed a significant frequency effect (p = 0.004) while lower body did not (p = 0.070). Neither subgroup was volume-equated, though, so treat it as a lead rather than a rule.

Should I switch from a split routine to full body? Not on the basis of this paper. It compares numbers of sessions, not how those sessions are organised. Either structure can deliver the same weekly volume.

Is more frequency better for advanced lifters? The consensus has not settled it, and this study cannot settle it either. The wider frequency literature has the same shortage of trained subjects, so it is a gap in the field rather than a quirk of one paper. The volume side is not in dispute: an advanced lifter carrying more weekly sets may need more days to fit them in without the last sets falling apart.

How Pick It Up Helps

The mechanism the researchers point to — sessions long enough that the last sets degrade — bites hardest when you are fitting a week’s volume into fewer days, and it is hard to see while it is happening. Set five coming in well under set one shows up in the log rather than in the moment.

Pick It Up adjusts targets during the session. As you report the weight, reps and difficulty you hit, the app updates the target for your next set — so a session that runs long gets targets that reflect the fatigue you are carrying, rather than a number written before you walked in. That holds the quality of late sets whether you are training twice a week or five times.

It also handles the distribution. Pick It Up asks how many days a week you can train and how long you have each day, then spreads the sets across those days, so the weekly volume lands where it should whether your week has three slots in it or six.

Start a free trial and have all the science taken care of for you, so you can get back to building a better you.