Strength training: the best-documented treatment we have – Knowledge

1. Why this article?

There is one treatment in physiotherapy that turns up on the list for a remarkable number of complaints: for osteoarthritis, after a knee replacement, for back pain, for muscle weakness in older age, after an injury, for fall prevention, for osteoporosis. It is not massage, not the ultrasound device and not the tape bandage. It is strength training.

And it is the treatment where we most often have to explain, in the practice, why it is a treatment at all. «I don't want to build muscle, I just want my shoulder to stop hurting.» «I'm 74, it's too late for that.» «I have a resistance band at home, isn't that enough?» «I used to go to a gym, it did nothing for me.»

These sentences are understandable. For decades, strength training was marketed as something young men did with dumbbells, not as medicine. Yet the evidence is now as clear as it rarely is in exercise medicine.

In March 2026, the American College of Sports Medicine – the largest sports medicine association in the world – renewed its recommendations on strength training for the first time since 2009. Rather than collecting individual studies, the working group evaluated 137 review articles, each of which had already summarised many individual studies – together covering more than 30,000 participants [1]. It is, in other words, a summary of summaries. No single new result will easily overturn these conclusions.

The findings come in two halves, and the second is the more interesting one.

The first half: compared with people who did not train, strength training improved strength, muscle mass, muscular endurance, walking speed, balance and several abilities that everyday independence rests on – getting up from a chair, for example. Measured across tens of thousands of people.

The second half: of all the details people argue about in gyms, remarkably few make a demonstrable difference. Whether you train on machines or with dumbbells, whether you lift fast or slowly, whether you go in the morning or the evening, how long you rest between exercises, how sophisticated your training plan is, whether you push to complete exhaustion – for the outcomes that matter, none of it made a consistent difference.

That is good news, and practical news at that. Because if most of it does not matter, only a few things are left that do. Those are what this article is about.

This text is the foundation article for every other guide on this site in which strength plays a role – and that is nearly all of them. It explains what happens inside the muscle, what makes a programme effective, how much is really necessary, why some people progress faster than others, and what is added in physiotherapy when pain, surgery or an illness is involved. The numbers in square brackets refer to the reference list at the end.

One note first: this article does not replace an individual assessment. What your doctor decides for your situation takes precedence.

2. What happens inside the muscle – and why it takes weeks

When you lift weights for the first time in years, something happens in the first weeks that surprises many people: you become considerably stronger without a single gram of muscle being added.

The reason: strength is not only a question of muscle size. It is first of all a question of activation – of how well the nervous system can call up the muscle you already have.

You can picture it as an orchestra. A muscle consists of thousands of fibres, conducted in groups by one nerve cell each. In someone who has not trained for a long time, the groups never all play at once, the tempo is slow, and the muscles involved get in each other's way. Training brings the orchestra back together within weeks – without a single new musician joining. That is why strength rises steeply in the first six to eight weeks while the thigh stays exactly as thick as it was.

This explains two things. First, why children before puberty can get stronger without looking more muscular [18]. Second, why strength training in physiotherapy often works faster than a pure muscle-building calculation would suggest: after knee surgery the thigh muscle has not only become thinner – the nervous system also calls it up less well. Getting that activation back is a matter of weeks, not months.

Only afterwards – and then permanently – does the rebuilding of the muscle fibres themselves follow. At the 2025 conference on the response to training in Jyväskylä, whose results were published in 2026, one of the key points was that this rebuilding happens in more than one dimension [17]: the fibres become thicker as additional contractile filaments are built inside them, but they also become longer as new building blocks are inserted along the whole length of the fibre. On top of that come adaptations to the cell machinery that makes building protein possible in the first place.

This is not an academic detail. A muscle that becomes longer can produce force across a greater range of movement – that is the difference between «strong while seated» and «strong when getting out of a deep armchair».

What matters is the time scale. The nervous adaptation is fast, the structural one slow. Anyone who checks the mirror after four weeks and sees nothing is using the wrong measure. Anyone who notices after four weeks that the stairs are easier is using the right one.

3. What strength training demonstrably achieves

Let us begin with the biggest and most boring question: compared with doing nothing at all – what does it achieve?

For healthy adults, the ACSM evaluation lists the following effects, each measured against groups that did not train [1]:

• More strength – being able to move more weight. Documented in 26 review articles with over 23,000 participants.
• More muscle mass – 12 review articles, around 15,000 participants.
• More muscle power – force and speed at the same time, as when getting up quickly. Why that is something other than plain strength is explained in section 12.
• More muscular endurance: an effort can be sustained longer before strength fades.
• Faster walking, better balance, easier rising from a chair – exactly the abilities that independence in everyday life depends on.

What is remarkable is the range of ways this can be achieved. Not only dumbbells and machines showed these effects, but also training with elastic bands, circuit training – several exercises performed briskly one after the other – and training at home [1]. So anyone who has good reasons not to go to a gym does not lose the effect. They only lose part of the choice of weights.

Going beyond the muscle itself is an older but well-summarised review with the programmatic title «Resistance Training is Medicine» [22]. Some of its figures:

• Adults who do not exercise lose 3 to 8 per cent of their muscle mass per decade.
• Ten weeks of strength training added, in an evaluation with over 1,600 participants, an average of 1.4 kilograms of lean mass and removed 1.8 kilograms of fat.
• Energy expenditure at rest – what the body needs simply to run itself – rose by around 7 per cent.
• Add to that favourable effects on blood pressure, blood lipids and blood sugar, plus an increase in bone density of 1 to 3 per cent.

These numbers are averages from studies with motivated participants. They describe what happens in a group on average – not what will happen to you personally. But the direction is clear: strength training does not only act on the muscle. Through the muscle, it acts on the metabolism of the whole body.

Two limitations belong with this. First, the ACSM paper explicitly covers healthy adults aged 18 and over. Anyone with an illness will find no answer there – for that there are separate guidelines and, on this site, separate articles. Second, the working group itself concedes that many of the underlying individual studies are small and methodologically mediocre. An overview of reviews is only as good as what lies beneath it.

4. The settings that actually count

Now to the real question: when you train – what do you have to get right?

From here on a few technical terms appear that are in every training plan and that you need to know in order to follow the rest. They are quickly explained:

• Repetition. A single complete movement. Once down into a squat and back up: one repetition.
• Set. A series of repetitions performed one after another without a break. Ten squats down and up, then stopping to catch your breath: one set of ten repetitions. Usually a second and a third set of the same exercise follow.
• Maximum weight. The heaviest weight you can move cleanly exactly once in a given exercise. All percentages in this text refer to it. «70 per cent of your maximum weight» therefore means a good two thirds of what you could manage once – a weight with which you will manage roughly ten to twelve repetitions.
• Strength and muscle growth are two different goals. Strength means how much weight you can move. Muscle growth means how thick the muscle becomes. The two are connected, but less closely than people assume – and, as you will see in a moment, one of them partly needs different conditions from the other.

For most people in physiotherapy, incidentally, strength is the goal and not muscle thickness. The muscle is welcome to grow; what matters is whether you can carry the shopping bag and get up from a chair.

4.1 Effort – the one condition that is not negotiable

The main recommendation of the ACSM paper in a single sentence: healthy adults should train with high effort at least twice a week, working all the large muscle groups [1].

The word «effort» deliberately comes before all the numbers. Because the muscle does not respond to a particular weight, nor to a particular number of repetitions. It responds to how close you were to the end of your strength when the set finished.

An example makes this clear. You do twelve repetitions with 20 kilograms and afterwards almost nothing more would have been possible: that works. You do the same twelve repetitions with 20 kilograms but could easily have added another twenty: that barely works at all. Same weight, same number of repetitions, completely different effect – the difference lies solely in how far you were from the end.

This is the most common flaw in home programmes: the exercise is well chosen, the execution is clean, but it is too easy. If you have been pulling the same band for the same number of repetitions for eight weeks and no longer find it difficult, it is time for a stronger band or for more repetitions. Otherwise the programme keeps you moving, but it no longer makes you stronger.

4.2 How heavy? That depends on what you want to achieve

This is the point at which the two goals from the list above part company for the first time. So let us take it slowly and in order.

If you want to become stronger – that is, able to move more weight – heavy helps. Here the overview found a clear relationship: the heavier the training, the greater the gain in strength. As a guide value the working group gives 80 per cent of the maximum weight or more [1]. In practice you do not recognise this by the percentage but by the number of repetitions: a weight with which you manage at most eight clean repetitions is heavy enough. If you manage fifteen, it is too light for this goal.

If, on the other hand, you want to build muscle – a thicker muscle – the weight is almost a free choice. That sounds implausible at first but is well documented: whether people trained with 30 per cent of their maximum weight or with 100 per cent made no difference to muscle growth [1]. The muscle grew equally in both cases.

One condition does apply, however, and it is decisive: the set has to be carried far enough to become strenuous (section 4.1). With a light weight that simply means more repetitions. Someone using 30 per cent of their maximum weight may need twenty-five repetitions before it really becomes hard – and then the muscle grows just as it would with eight heavy repetitions.

This is one of the most liberating findings of recent years, and in physiotherapy one of the most useful. Because many people cannot or do not want to lift heavy: because of a painful shoulder, because of high blood pressure, out of fear of injury, or simply because there are no heavy weights at home. For all of them: lighter works too – the individual set just takes longer.

Summarised in one sentence: heavy is necessary if you want to get stronger. For a thicker muscle any weight will do, as long as you carry the set far enough.

That leaves the third goal, muscle power – force and speed at once, as when rising quickly or catching yourself after a stumble. For this, neither very heavy nor very light weights were best, but moderate ones: 30 to 70 per cent of the maximum weight, moved briskly [1]. Why moderate weights achieve most here, and why it matters particularly with age, is explained in section 12.

4.3 How many sets?

The number of sets is the amount of training. More amount works – but only up to a certain point, and that point is lower than many people assume.

The beginning of the series is well established: one set achieves more than none, two sets achieve more than one [1]. After that the return keeps shrinking. A calculation of the dose-response relationship – the question of how much more each additional portion adds – found the following points at which the curve flattens out [1]:

• For strength: about two to three sets per exercise. A fourth and fifth set add hardly anything.
• For muscle growth: about 18 to 20 sets per muscle group per week. Here the count runs across the whole week and across all exercises that load the same muscle group.

The recommendation is therefore: at least two sets per exercise. More is allowed and helps a little, but every additional set adds less than the one before it.

For anyone who has no sporting ambition but simply wants to stay strong, the lower limit is the more interesting figure. A review on time-saving strength training gives a minimum of four sets per muscle group per week, with a weight that allows six to fifteen clean repetitions [2].

In practice that means, for example: two training sessions per week with two sets per exercise in each – and you already have four sets a week. With three to four exercises, such a session takes about twenty minutes.

4.4 Frequency: twice a week

For strength there was an advantage for at least two sessions per week over one; the upper limit could not be determined from the data [1].

For muscle growth, frequency hardly matters – provided the total amount over the week stays the same. Whether someone completed their twelve weekly sets on a single day or spread them over five days made no demonstrable difference [1]. The review on time-saving training reaches the same conclusion: how much you do in a week counts for more than the question of how many days you spread it over [2].

In practice that means: twice a week is the documented lower limit. How you place the sessions across the week is for your diary to decide, not a rule.

4.5 Completing the movement

«Range of movement» means how far you actually move the joint during an exercise – whether you go all the way down in a squat or only part of the way, whether you pull the arm all the way back when rowing or turn around halfway.

For strength, the full movement was superior to the partial one [1]. For muscle growth the data were not sufficient for a conclusion.

In physiotherapy this is one of the things we correct most often. Half squats, half rowing pulls and half bench presses are the norm in unsupervised programmes – and usually for the same reason: the weight has been chosen too high, so that the full movement is no longer possible. Anyone who wants to work through the whole range therefore often has to take weight off.

That feels like a step backwards, but it is not one. A full squat with 20 kilograms achieves more than a half squat with 30.

The qualification: what «full» means depends on your joint. After a knee replacement, with a stiff shoulder or during acute pain, the full movement is sometimes not possible – then the largest range that stays comfortable is the right one. What is meant is never «until it hurts», but «not artificially cut short».

4.6 Order within the session – and why increasing is necessary

Two smaller but practical points to finish.

First, the order. The exercise you want to get stronger in belongs at the beginning of the session, not at the end [1]. The reason is simple: at the start you are fresh and can move more weight. So if leg strength matters most to you, begin with the legs – not with twenty minutes of warming up and three secondary exercises. For muscle growth, incidentally, the order made no difference.

Second, increasing. That you have to do more weight or more repetitions over time has always counted as a basic principle. The overview adds an interesting nuance here and separates two questions [1]:

• For it to achieve anything at all, increasing is not essential. Even a programme that always stays the same is better than no training.
• For it to keep achieving something, increasing is necessary. The body gets used to a load that never changes and stops adapting.

So anyone who has been doing the same ten repetitions with the same ten kilograms every week for a year loses nothing by it – they hold their position. But they also do not get any stronger. How to increase in practice is set out in section 19.3.

4.7 Lowering counts too

Every movement with a weight has two halves: the lifting, in which the muscle shortens, and the lowering, in which it lengthens under tension. Most people give the second half away – the weight drops rather than being guided.

That is a mistake for which there is evidence. In the overview the muscle grew more when only the lowering half was trained than when only the lifting half was [1]. Deliberately increasing the weight during that phase was equally effective. For strength there were additional advantages for so-called flywheel devices, in which the braking phase is loaded particularly heavily – although there it cannot be cleanly separated whether the advantage comes from the device or from the greater load during lowering [1].

Now comes the qualification, and it matters so that the wrong conclusion is not drawn. The difference shows up in the extreme comparison: lowering only against lifting only. As soon as both halves are performed normally, it no longer matters how slowly you lower (section 9).

The practical recommendation is therefore not «lower twice as slowly». It is: guide the weight on the way back as well, instead of letting it drop. Two to three seconds are enough – not ten.

In physiotherapy this is one of the most frequent corrections of all, and it costs nothing: same weight, same number of repetitions, only the second half of the movement no longer given away.

5. What makes surprisingly little difference – and what is unresolved

5.1 Three statements that do not mean the same thing

A careful distinction is needed here, one that is almost always lost in everyday conversation. The overview keeps three statements strictly apart [1]:

• «Has an effect» – there are enough data, and they show a difference.
• «Has no effect» – there are enough data, and they show no difference.
• «Cannot be determined» – there are too few or too contradictory data.

The second point is a genuine finding, the third merely an admission. On the internet the two are regularly lumped together.

Falling into the category «shows no consistent difference» were:

• Training to muscular failure – no advantage for strength, muscle mass or muscle power (section 6).
• Machines versus free weights for strength (section 8).
• Unstable surfaces – balance cushions, wobble boards and the like – for strength.
• Movement tempo and the total time under tension (section 9).
• Time of day – morning versus evening.
• Rest length between sets – under one minute versus over one minute, for strength.
• Set structures such as cluster sets or contrast sets (section 10).
• Sophisticated training plans for muscle growth. What is meant are plans that vary weight and repetitions according to a fixed scheme over weeks and months – «periodisation» in technical language. Whether someone followed such a plan or always did the same thing led to the same result; the working group states explicitly that these plans are less important than previously assumed [1].

You can read this list in two ways. Disappointed – so many refinements, and none of them counts. Or relieved: there is no trick you are missing.

Anyone who does the following four things has the essentials in place:

• train twice a week,
• make it strenuous enough that only a few repetitions would still have been possible at the end of each set,
• perform the movements fully rather than half,
• and after a few months move more than at the start – more weight, more repetitions, or both.

Everything else is taste, availability and enjoyment. The working group writes exactly that: a programme someone actually keeps up is worth more than one that matches the specifications perfectly [1].

The figures behind this make clear why that is not merely a kind remark. In the United States only about 30 per cent of adults do any form of strength training twice a week; nearly 60 per cent do none at all. Among older people the working group estimates participation in programmes of the kind described here at 10 to 15 per cent [1]. The gap between the best training and no training is many times larger than the gap between the best and merely good.

5.2 Unresolved does not mean disproved

The third category – «cannot be said from the available data» – is surprisingly large. Popular summaries usually skip it, because no advice can be made out of it. For the following questions the data simply were not sufficient for an answer [1]:

• Blood flow restriction – training with light weights and a cuff constricting the limb. In the studies it is nearly always compared with heavier training, so the effect of the cuff cannot be separated from that of the weight. This matters for physiotherapy, because the method is readily recommended for situations in which heavy weights are forbidden – after surgery, for example. It may well be sensible; documented in the sense of this overview it is not.
• The full versus the partial movement – but only for muscle growth. For strength the full movement is documented as superior; whether it also makes the muscle grow more is open.
• The rest length between sets, again for muscle growth.
• Sophisticated training plans for strength – a single review saw a slight advantage, overall the data were not sufficient.
• Combining strength and endurance training in the same session. The question of whether endurance training holds back gains in strength has been asked so often and answered so poorly that no recommendation could be derived from it.
• Machines versus free weights for muscle growth (for strength, by contrast: no difference, see section 8).

Anyone who gets one of these questions answered with certainty is hearing an opinion, not evidence. That is no disgrace – in practice one has to decide even where data are missing. But the difference deserves to be named.

6. Does it have to go to muscular failure?

«The last set has to burn, otherwise it is worthless.» This rule persists stubbornly, and it has been refuted.

First the term. Muscular failure means: you continue a set until literally not one further repetition succeeds. The weight comes to a halt halfway even though you are pushing with all your strength. That is something different from «it becomes strenuous» or «it burns» – it is the actual end.

Whether this end is necessary is shown particularly vividly by an Australian study, because it uses a clever device: instead of comparing two groups of people, it compared the two legs of the same person [3]. That removes all the differences that otherwise get in the way – age, diet, sleep and disposition are identical in both legs.

Eighteen training-experienced people – twelve men, six women – trained twice a week for eight weeks on the leg press and leg extension machine. One leg went to muscular failure every time. The other leg stopped one to two repetitions earlier.

The result: the thickness of the thigh muscle increased practically identically in both legs – 0.181 against 0.182 centimetres. That is not «almost the same», that is a dead heat.

Only the price differed. The leg trained to failure was considerably more exhausted: in the following sets the participants managed fewer repetitions with it, and they moved the weight more slowly. In short: the same effect, more exhaustion.

The large overview reaches the same conclusion and goes one step further. It explicitly advises older people against training to muscular failure [1]. Two reasons are given: at maximal effort blood pressure rises sharply, and anyone who is completely exhausted no longer performs the movement cleanly – which increases the risk of injury.

What is recommended instead is to end the set shortly before muscular failure: at the point where two to three repetitions would still have been possible.

Which raises the question of how you are supposed to know that two to three would still have been possible without trying it. That is exactly what the next section is about – and the answer is surprisingly well researched.

7. «Two repetitions in reserve» – the most useful number in strength training

The classic way of setting a training weight goes via the maximum weight: you test once what is possible at most and then calculate percentages of it. This has two drawbacks. The test itself is unsuitable for many people, and the result goes out of date – daily form fluctuates, and strength rises during the programme while the percentage from the old test stays where it was.

The alternative turns the question around. Instead of calculating in advance what percentage the weight should be, you ask yourself a single question at the end of each set:

«How many repetitions could I just have managed cleanly?»

What is meant here is explicitly repetitions – not sets, not minutes. Say you have done ten squats, and your sense of it is: an eleventh and a twelfth would still have worked, at the thirteenth it would have been over. Then you have two repetitions in reserve. Specialists abbreviate this to RIR, from repetitions in reserve [4]. Zero repetitions in reserve means nothing more was possible – in other words, muscular failure.

The target range for most people is two to three repetitions in reserve. That is strenuous enough to work, and it leaves enough of a safety margin for the movement to stay clean.

Can this be judged reliably at all? Surprisingly well. Fifteen young men with little training experience were tested at three identically structured appointments: they were to increase the weight in the deadlift and bench press until exactly one repetition would still have been possible at the end of the set. The weights they chose agreed well from appointment to appointment – they differed by only about 3 to 6 per cent [5]. People can therefore sense fairly reliably how close they are to the end.

The same study incidentally shows why percentages alone are misleading. «One repetition in reserve» corresponded to quite different percentages in the deadlift depending on how long the set was: 88 per cent of the maximum weight in a set of three repetitions, 84 per cent at five and 79 per cent at eight. The same felt effort – three different numbers on the weight plate. Here your own perception is not the less accurate measure but the more useful one.

Where the judgement is weak. An Australian study had twenty trained men estimate their reserve and then actually carry the set to the end – so the estimate could be compared with reality [6]. Two results matter in practice.

• In the first set the participants regularly judged wrongly – and always in the same direction. They believed they were close to the end although considerably more would have been possible. From the second and third set onwards the judgement became more accurate.
• Whether they knew how heavy the weight was made no difference to the accuracy.

From this follows a very practical rule: the first set is almost always chosen too light. If you notice that it was easier than expected and therefore add weight for the second set, you are doing it right – not wrong. That is exactly what the first set is for.

Does this method achieve more than a fixed plan? A British study had 31 training-experienced men do squats twice a week for twelve weeks. One group worked to a fixed plan with percentages from a test at the beginning. The other chose the weight from set to set according to how many repetitions they still had in reserve. Both groups became stronger – the second considerably more so [7]:

• Front squat: plus 11.7 per cent against plus 8.3 per cent.
• Back squat: plus 10.8 per cent against plus 7.1 per cent.

The reason is plain: anyone who becomes stronger during the programme can also lift more. The second group noticed this every week and added weight accordingly. The first group carried on calculating from a test value that had meanwhile gone out of date – and therefore trained increasingly too light.

One caveat belongs with this: the ACSM paper states that there are too few data to recommend an exact number [1]. Two to three repetitions in reserve is a sensible working range – not a measured optimum.

How to apply it in everyday life:

• Choose a weight where, at the end of the planned set, you have the feeling that two to three repetitions would still have been possible.
• If the set was too light, add weight for the very next set – not next week.
• After each session note three things: the weight, the number of repetitions, and how many more would have been possible. Without notes people remember too optimistically and never increase.
• If you cannot judge at all where your limit lies: once – supervised, and on a safe exercise such as the leg press – carry a set genuinely to muscular failure. After that you will know what «two left» feels like. That single calibration is worth the effort.

8. Machines or free weights?

Few questions in the fitness world are answered with such conviction and supported so poorly.

A 2023 meta-analysis evaluated thirteen studies with a combined 1,016 participants that compared the two directly [8]. The result can be summarised in one sentence: you get strong at what you practise.

• Those who trained with free weights performed better in tests with free weights.
• Those who trained on machines tended towards better results in tests on machines.
• Looked at across both kinds of test, however, there were no differences – neither in strength during movement, nor in strength while holding, nor in jump height, nor in muscle growth.

A word of explanation: free weights are dumbbells, barbells, kettlebells – anything moved freely in space that you have to keep in balance yourself. Machines guide the movement along a fixed path; the machine takes care of the balance.

The individual studies confirm the picture. 36 men without training experience trained for ten weeks either on machines, with free weights, or switching from machines to free weights after five weeks. No differences appeared between the groups in muscle circumference, strength and movement quality – and those who switched mid-programme lost nothing by it [9]. In a further study with 46 people, 26 of them women, muscle thickness in the upper arm and thigh increased equally in both groups over eight weeks. In only one respect was the machine group ahead: in bench pressing on the machine – precisely the movement they had practised [10].

One study stands out from the rest, and is interesting for exactly that reason. 32 physically active people between 60 and 86 trained twice a week for six months – one half on machines, the other with barbell and squats. In leg strength the difference was large: 113 per cent gain with the free weights against 44 per cent on the machines [11].

Even so, the authors explicitly do not conclude that free weights are generally better – the group was small and consisted of already sporty people. They found another result more important: the free-weight group rated enjoyment, motivation and everyday usefulness higher. And what people enjoy, they keep doing longer.

This means the practical question has to be put differently from usual. Not: which is better? But: which suits you?

• Machines are easier to learn, need less supervision, can be dosed finely, and are the better entry point when balance is a problem, when weakness is pronounced, or when there is fear.
• Free weights and body-weight exercises resemble everyday movements more closely – standing up, lifting, carrying – and train stabilisation along the way. They need more instruction.
• Elastic bands are documented as effective for strength, muscle mass and parts of everyday function [1], and at home they are often the only realistic option. Their weakness is dosage: beyond a certain level of strength the resistance is no longer great enough.

In physiotherapy we mix the two for a simple reason: the start should be safe, the goal should be close to everyday life. What begins as a leg press later becomes a squat – not because the leg press is bad, but because you do not have one at home.

9. How fast should you lift?

Movement tempo is often sold as a refinement for the advanced. In the data it is above all one thing: surprisingly unimportant.

A detailed review from 2021 worked through the state of research on this [12]. The key finding: for muscle growth it made no difference whether a single repetition took half a second or eight seconds. That is an enormous range – from brisk to very markedly slow – and the result was similar throughout.

Only with extremely slow lifting of ten seconds or more per repetition does it become questionable. The reason, however, is an indirect one: only something that is light anyway can be moved that slowly – and then the effort the muscle responds to is missing.

A Brazilian study examined the question particularly cleanly [13]. 38 untrained men trained the bench press for ten weeks in two variants set up so that the muscle spent the same length of time under tension in each – namely 36 seconds per set:

• One group did twelve repetitions of three seconds each.
• The other did six repetitions of six seconds each.

The outcome was the same in both groups: the same gain in muscle, the same gain in strength. Twice as many repetitions at half the duration – no difference.

For strength the tempo question resolves itself anyway: anyone lifting a very heavy weight moves slowly of necessity, no matter how fast they would like to [12]. What counts is the intention to push briskly, not the speed that actually results.

In practice a single rule remains: lower under control, push or pull briskly, do not use momentum. If you see sequences of digits such as «3-1-2-0» in training plans, denoting the seconds for each phase of the movement, you may safely ignore them.

10. Short cuts: cluster sets, supersets and training under time pressure

If most of it does not matter, one question suggests itself: can the time be shortened without losing the effect? To a good extent the answer is yes.

Cluster sets are so called because a set is not pushed through in one go but divided into small clusters. Instead of twelve repetitions without interruption you do three lots of four repetitions with roughly twenty seconds of rest in between. In the end it is the same twelve repetitions, but you are less exhausted along the way and can therefore move a heavier weight cleanly.

A systematic review of nine studies with 172 participants confirms exactly that: heavier weights, more total volume, less fatigue [14]. Whether more comes out of it in the end is another question – and the answer is no. A Spanish study again compared both forms in the two legs of the same people over eight weeks, with the same number of sets and repetitions and the same effort. Muscle thickness and muscle mass increased equally in both legs [15]. The large overview likewise finds no difference between such set forms [1].

Cluster sets are therefore no miracle weapon, but a useful tool when someone is supposed to train heavy and loses clean execution in the process. In physiotherapy we use the principle regularly – for instance with people whose technique runs away from them after the eighth repetition.

When time is short, the review on time-saving strength training is the most practical source on the subject [2]. Its recommendations:

• Prefer exercises that use several joints at once. A squat loads hip, knee and ankle in one go; a leg extension machine only the knee. If you have little time, take the squat. Three exercises are enough as a basic framework: one for the legs (squat or leg press), one pulling exercise for the upper body (rowing, pull-up) and one pushing exercise (bench press, press-up).
• At least four sets per muscle group per week, with six to fifteen repetitions per set.
• Fill the breaks instead of sitting them out. Anyone who does the next exercise during the rest period of the first – leg press and rowing alternately, for instance – halves the training time without doing less. Such forms go by names like supersets or drop sets. The authors' caveat: they suit muscle growth better than pure strength, and long-term data are lacking.
• Warm up only with the exercise itself. One or two lighter sets of the exercise that is coming next are enough – a general warm-up programme is not needed.
• Stretch only if flexibility is your goal. It is not a necessary part of a strength session.

Added together, an effective programme can be fitted into two sessions of twenty to thirty minutes a week for most people. Lack of time is the most frequently cited reason for not doing strength training – and the easiest one to refute.

11. Keeping is far cheaper than building

One of the most reassuring findings concerns not the building but the keeping.

In an older, frequently cited study, 52 people trained two or three times a week for ten weeks. Their workload was then cut. Those who dropped back to twice or even only once a week kept the strength they had gained over the following twelve weeks – some even added a little. Only those who stopped altogether lost it [2].

A second study ran for 32 weeks and looked at younger and older participants separately. After sixteen weeks of building, one session per week was enough to maintain the strength gained in both age groups. For muscle thickness, however, a difference appeared: the younger participants kept their gain, while in the older ones it returned to the starting value despite the weekly session [2]. The Jyväskylä conference reaches the same conclusion: older people need more training than younger ones to keep muscle mass they have built [17].

For everyday planning this is good news. Holidays, a project at work, a bout of flu: reduce instead of stopping. One short session per week keeps most of what you have worked for. And the reverse holds too: anyone who wants to stay strong over years does not need a high workload – they need a low one that never breaks off entirely.

How load and recovery build up over weeks can be tried out in our learning game Randvoll.

12. Strength, power and rate of force development – three related but different things

When you stumble, what decides the outcome is not how much you can lift. It is how quickly you can produce force – within the two tenths of a second before you fall.

In everyday conversation this gets called «explosive strength». That umbrella term is imprecise, however, because two distinct quantities hide behind it, measured differently and trained partly differently. Since the distinction has practical consequences, here are the three terms kept apart:

• Strength – how much. The weight you can move at all. Measured in kilograms.
• Muscle power – how much times how fast. Power is force multiplied by movement velocity, measured in watts. It is highest at moderate weights, and the reason is simple: a very heavy weight you necessarily move slowly; a very light one you move fast but with little force. The product of the two peaks in between. That is why 30 to 70 per cent of the maximum weight keeps coming up in connection with power.
• Rate of force development – how steeply. How fast force rises from zero, measured as force per unit of time. This is not about the whole movement but about its very first fractions of a second.

The difference between the last two can be shown with the stumble. Reaching your full strength easily takes three tenths of a second – by then you are already down. What actually counts in that moment happens even earlier: in the first 50 to 75 milliseconds, the force produced depends above all on how quickly the nervous system can raise the firing rate of the muscle fibres [23]. That is the rate of force development. Muscle power, by contrast, describes the whole movement – getting up out of the chair, for instance.

12.1 Power training: moderate load, maximal intent

For power there is a clear recipe [1]: moderately heavy weights – 30 to 70 per cent of the maximum weight – and the intention to perform the lifting part of the movement as briskly as possible. Together with a low volume, so that the movement does not become slow through fatigue; as a rule of thumb, at most around 24 repetitions in total per exercise, that is roughly three sets of eight.

What is decisive here is the intent, not the visible result: what counts is that you try to push briskly. How fast it actually looks is secondary – at 70 per cent of your maximum weight even the best attempt still looks fairly leisurely.

This form of training is called power training, and it is the reason this section belongs in a patient guide: power training improved everyday functions that classic heavy strength training did not improve to the same extent – walking performance and stair climbing [1].

How large that advantage is has been investigated by our own practice. A meta-analysis by Marielle Tschopp, Martin Sattelmayer and Roger Hilfiker pooled eleven randomised trials with 377 people over the age of 60 and compared power training with conventional, slowly performed strength training [24]. The result: an advantage for power training in functional outcomes, with an effect size of 0.32 (confidence interval 0.06 to 0.57).

And now the honest reading of that number. An effect size of 0.32 counts as small. The confidence interval reaches down close to zero – so the true effect could also be very slight. And when participants were asked about their own everyday function instead of being tested, the difference was no longer statistically secure (0.16; confidence interval −0.17 to 0.49). On safety, no firm conclusion could be drawn from the eleven trials.

Translated: power training is feasible for older people and on average somewhat better than slow strength training – but the difference is small. Anyone deciding between «train slowly» and «do not train at all» is making the far more important choice.

12.2 And the rate of force development?

Here it becomes interesting, because the finding contradicts the obvious expectation.

One might assume that the steepness of the rise can only be trained with fast movements. The authoritative review on the subject states, however, that it can be improved both by explosive-type and by heavy-resistance training – in both cases mainly because the nervous system activates the muscle more rapidly [23].

That is an important reassurance. It means that anyone training heavy does not lose the steepness of their force rise. And conversely it means the two forms of training are not opponents but work through the same mechanism – the faster activation already familiar from section 2.

A second qualification belongs with it: the rate of force development is difficult to measure reliably [23]. You will therefore rarely encounter it as a measured value in practice – unlike your walking speed or the time for five chair rises.

The practical consequence of 12.1 and 12.2 is pleasingly simple: what connects both quantities is the intention to be brisk – regardless of the weight. Whether you do that with 40 or with 80 per cent of your maximum weight matters less than whether you do it at all.

12.3 What this means in everyday life

How far the effect can reach is shown by a study of 26 people with mild to moderate Parkinson's disease. For three months they trained twice a week with light weights and high movement speed, supplemented by balance and agility drills. What improved was not only the maximum weight and peak power, but also the slowing of movement itself – in arms and legs – as well as quality of life, measured with a disease-specific questionnaire [19]. This is a small study and no proof for everyone; but it shows that the line between «training» and «treatment» is not a sharp one here.

For most people over sixty a plain rule can be drawn from this: part of the programme should be brisk. Rising from a chair as fast as can be done cleanly, and sitting back down slowly. A calf raise briskly up, controlled down. Not everything – but not nothing.

13. Why some people progress faster than others

Two people, the same programme, the same period of time. One increases her squat weight by thirty per cent, the other by only eight. Why? And are there people in whom strength training simply does not take?

This question was the subject of a dedicated conference at the University of Jyväskylä in 2025; the results were published in 2026 [17]. The answer is illuminating – and to a good extent it casts doubt on the question itself.

First: a large part of the apparent differences is not a difference at all. If you measure a person's muscle thickness twice without anything happening in between, two different values already come out – every measuring device has an inaccuracy, and the body fluctuates from day to day. So anyone who classifies a person as «responds well» or «responds poorly» on the basis of a single measurement before and a single one after is to a considerable extent sorting noise. To separate genuine differences from chance, you would have to put the same person through the same programme repeatedly – such studies are laborious, and there have been few of them so far.

Second: the search for «the gene» has so far been unsuccessful. Individual gene variants could not be confirmed in further investigations. A combined score from eleven variants did explain around 47 per cent of the differences in 440 people – but a genuinely reliable score of that kind would require investigations in over 100,000 people. In short: there is no genetic test that tells you how to train. Even though such tests are sold.

Third: what actually helps is disappointingly ordinary. As the most reliable levers when progress is weak, the conference names [17]: more sets, a suitable frequency, enough effort within the individual set, and regular increases over time. Which exercises someone chooses is secondary. And the summary of the whole thing, from Stuart Phillips's lecture: the most important controllable factor is regular training with slowly rising load – over years, not over weeks.

On dietary supplements the conference gives a sober figure: creatine and long-term protein supplementation produced an additional 0.5 to 1.0 kilograms of muscle mass over twelve weeks and more [17]. That is measurable, but small – and it replaces not one single training session.

And one sentence from the practical conclusions that deserves to be passed on verbatim: not all older people respond poorly – chronological age is no reason not to try [17].

14. Women and men

Strength training research has historically been conducted on men, and that still has consequences today. A 2026 review summarises what can be said about differences between the sexes [16].

The differences that do exist:

• Men lift more kilograms on average – they have more muscle mass and a higher proportion of fast, powerful muscle fibres.
• Women fatigue more slowly on average, especially when holding a position, and have a higher proportion of slow, enduring fibres.

And now the point that actually matters. The question is not who lifts more, but: who improves more through training? On this the review states: the percentage gain in strength and muscle mass is largely the same in women and men. A woman who trains for twelve weeks gains proportionally about as much as a man – she simply starts from a different baseline.

The review closes with a recommendation worth remembering: because the differences within each group are far larger than those between the groups, an individually tailored approach makes more sense than a specification by sex [16]. Put differently: what counts is what this one person can do – not what women or men can do on average. How the course of training develops over months and years in women, incidentally, remains considerably less well researched than in men. That is a gap, not an answer.

Two things follow for practice. The widespread worry that strength training makes women undesirably muscular finds no basis in the gain figures – muscle mass builds up slowly, in women as in men, and certainly not by accident. And the equally widespread habit of giving women lighter weights and more repetitions as a matter of course finds none either.

15. Children and adolescents

One myth persists particularly stubbornly here: that strength training stunts growth or damages the growth plates.

The American Academy of Pediatrics has published a clinical report on this, reaffirmed in 2024 [18]. The core statements:

• Well-designed programmes have shown no negative effects on the growth plates, on height growth or on cardiovascular health.
• Children before puberty become stronger without becoming more muscular – through activation, as described in section 2. Visible muscle gain only arrives with puberty.
• In injury statistics, injuries to children connected with strength training come predominantly from home equipment in unsupervised situations – not from instructed programmes.
• Strength training does not impair the development of endurance; combined programmes often work better than one form of training alone.

What is striking is the shift of the question that the report explicitly names: the worry used to be what happens when a child lifts weights. Today it is what happens if it does not – against a background of measurably declining muscular fitness in children over the decades [18].

Also useful for parents is the concept of training age. What a child may do depends not on their year of birth but on their previous experience, their movement quality and their maturity. A ten-year-old with three years of instructed experience may be further along than a fourteen-year-old who is just starting.

In practice: strength training for children does not mean a weight bench. It means body weight, bands, medicine balls, climbing, jumping – with the emphasis on technique and supervision, not on weight.

16. Older people: the group with the most to gain

If strength training has the standing of a medication anywhere, it is here. And it is precisely here that participation is lowest – the ACSM working group estimates it at 10 to 15 per cent [1].

How much is possible is shown by a Brazilian study of 132 community-dwelling people aged 60 and over whose muscle strength was already measurably reduced – in technical language, probable sarcopenia. For one year they trained twice a week in a mixed programme of strength, balance and mobility [20]. After twelve months the following had improved:

• Standing up, walking, turning, sitting down (the so-called timed up-and-go test): 7 per cent faster.
• Standing on one leg: held 53 per cent longer.
• Rising from a chair five times in a row: 10 per cent faster.
• Walking speed at the usual pace: 13 per cent higher.
• The proportion of those whose grip strength was still below the threshold fell from 88.6 to 67.4 per cent. In roughly a fifth of the group, in other words, the criterion for reduced muscle strength disappeared entirely.

The caveat is important and is named by the authors themselves: there was no control group. The participants were not randomly allocated but took part voluntarily in an existing offer. Part of the improvement may be down to selection, to getting used to the tests, and to the attention itself. What the study does demonstrate is feasibility over a whole year in a group considered difficult – and the order of magnitude of what is possible.

Sarcopenia, it should be said, is explicitly not an unavoidable sign of ageing but a condition in its own right with diagnostic criteria; we cover this in detail in the article Muscle weakness in older age.

And nutrition? A meta-analysis of nine studies with 496 people with sarcopenia examined whether protein building blocks taken as supplements – amino acids, leucine, HMB and the like – add anything on top of strength training [21]. What was compared, in other words, was «training plus powder» against «training alone».

The result is split in two: for grip strength, walking speed and the tests of physical performance there was an advantage – but not for muscle mass itself. Two limitations matter: the authors themselves rate the reliability of this evidence as low to very low. And the benefit depended on how much protein someone was already eating – those who were already getting enough gained little.

Translated, this means: anyone who eats a balanced diet with enough protein does not need powders. Anyone who eats too little – and that is common in older age – should solve that through food first and discuss the question of supplements with their family doctor. In any case, the main part of the effect comes from the training.

What runs differently when training with older people, compared with the textbook programme:

• Not to muscular failure. Sets end while two to three repetitions would still be possible – for this group the ACSM working group explicitly advises against going to the very limit [1]. See section 6.
• More sets per week to keep what has been achieved than younger people need for the same purpose [17].
• Part of the training should be brisk, not only heavy – power training for muscle power (section 12.1). The benefit is documented, but small [24].
• Think in years, not in weeks. The impressive figures above come from a whole year of training, not from eight weeks.

What strength and balance together achieve for fall prevention is set out in the article Fall prevention; why a hospital stay costs strength particularly quickly, in the article Leaving hospital weaker.

17. Is strength training safe?

Fear of injury is one of the most frequently cited reasons for not starting strength training – particularly among older people. The evidence on this is surprisingly reassuring.

The ACSM overview refers to an evaluation of over 38,000 participants from exercise studies, among them over 6,700 in strength training and over 11,000 older people: training did not increase the risk of serious adverse events. Milder complaints – pain, fatigue, swelling – occurred no more often than with endurance training [1].

Even clearer is an evaluation of 23 studies with 1,174 people with narrowed coronary arteries – that is, a known heart condition. In these studies 63 cardiovascular events occurred in total, none of them fatal. All 63 happened during endurance training, not a single one during strength training [1]. During strength training there were 20 musculoskeletal complaints; these went back to pre-existing problems – mostly knee osteoarthritis – and could be resolved by reducing the weight or changing the body position.

This does not mean that nothing can happen during strength training. It means: the risk is low compared with forms of exercise generally regarded as harmless. And when an exercise causes complaints, it should almost always be adapted rather than dropped.

The situation is similar for children: the injuries appearing in the statistics come predominantly from unsupervised use of home equipment, not from instructed programmes [18].

Where caution is warranted: with untreated severely raised blood pressure, with certain heart conditions, in the first weeks after surgery, with recent fractures, and with acute complaints of unclear cause. For all of these the rule is: get it assessed, then adapt – do not simply wait.

18. What is added in physiotherapy

Everything said so far comes from studies in healthy adults. In physiotherapy, healthy adults are rarely the ones sitting in front of us. What changes?

First: you do not start at zero, you start below it. After knee surgery, after two weeks in hospital or after six months of avoiding activity because of back pain, two things have been lost: muscle mass – and the nervous system's ability to call the muscle up fully (section 2). The first weeks therefore often serve not to build but to reconnect. That explains why progress at the beginning is frequently surprisingly fast: nothing is being built anew, something is being reconnected.

Second: pain has a say. In healthy people, managing the load is only about weight and recovery. Where there are complaints, a third question is added: how much pain is acceptable during and after the exercise?

As a rough working rule in many areas: tolerable pain during the exercise is acceptable as long as it settles back to the usual level within 24 hours. If you have more complaints the next morning than the day before, it was too much. Exactly how this rule looks, however, depends strongly on the condition – an overloaded tendon has different limits from osteoarthritis, a fresh muscle injury different ones from persistent back pain. This is the point at which an individual assessment cannot be replaced.

Third: it is measured, not guessed. The great advantage of strength training as a treatment is that it can be put into numbers: weight, repetitions, comparison between the two sides, results of simple tests. A programme without a measurement at the start is a programme where nobody can say afterwards whether it worked.

Fourth: there is a detour via the healthy side. If one leg may not be loaded after surgery, it is still worth training the other. Because training one side also makes the untrained opposite side stronger – a well-documented effect that works through the nervous system rather than through the muscle itself [1]. This does not replace training the affected side. But it bridges weeks in which nothing at all would otherwise be possible there.

For individual conditions our other guide articles go into depth: Osteoarthritis, Knee replacement, Back pain, Muscle weakness in older age, Frailty, Fall prevention and Leaving hospital weaker. Why passive treatments do not replace exercise is set out in the article Massage.

19. What a workable programme looks like

From everything said so far a basic framework can be built. It is deliberately kept simple – the refinements, as we have seen, are largely a matter of taste.

19.1 The basic framework

The table shows three columns for three different goals. For most people in physiotherapy the first column is the right one. If you like, mix them: two exercises heavy for strength, for instance, and one as power training for muscle power.

The rate of force development from section 12.2 deliberately has no column of its own: it improves through both heavy training and power training [23] and therefore needs no separate plan.

19.2 Which exercises

Four to six exercises cover the whole body [2]. What matters is not the individual exercise but that every direction of movement is represented:

• Straightening the legs against resistance – squat, leg press, or rising from a chair with a weight in your hands.
• Straightening the hip, that is bringing the upper body up out of a forward bend – deadlift, lifting the pelvis while lying down (bridge).
• Pulling with the arms – rowing on a machine or with a band, lat pulldown, pull-up.
• Pushing with the arms – bench press, press-up, pressing a weight overhead.
• Add as needed – calf raises, exercises for the trunk, targeted exercises for the body region that concerns you.

The summary of research on the response to training puts it like this: you have a great deal of freedom in choosing exercises, and little in effort, sets and progression [17]. If you dislike an exercise or cannot perform it, simply swap it for another from the same line.

19.3 How to increase

Increasing follows a simple order: more repetitions first, then more weight.

An example. You start with 8 repetitions at a weight that feels right. Over the following weeks you work your way up to 12 repetitions with the same weight. Once 12 succeed reliably, you increase the weight – and at first manage only 8 again. From there the same path starts over. This up and down is not a setback; it is exactly the pattern you want to see.

• Keep notes. A sheet of paper is enough: date, exercise, weight, repetitions. Almost everyone remembers their progress too optimistically.
• Start lower after a break. After two weeks of holiday or illness, begin with around 20 per cent less weight and work your way back over one to two weeks.
• Expect the plateau. The first two to three months move fast, after that it becomes hard going. That does not mean something is wrong – the curve looks like that for everyone.

19.4 A realistic timetable

• Weeks 1 to 2: learning the movements cleanly and measuring the baseline values. Deliberately start too light. Muscle soreness during this time is normal and says nothing about whether the training was any good.
• Weeks 3 to 8: strength rises noticeably. At this stage that comes almost entirely from the nervous system calling the muscle up better (section 2). Stairs, shopping bags and rising from a chair become easier long before anything can be seen on the muscle itself.
• Months 3 to 6: now the muscle grows measurably too. At the same time the gains become smaller – from here on, regularity counts for more than progression.
• From month 6: the programme becomes a habit. This is where it is decided whether a treatment turns into a lasting habit.
• Long term: twice a week. In stressful periods once – reduce instead of stopping (section 11).

19.5 How to measure for yourself whether it is working

Muscle circumference is the worst of all measures of success: it changes late, unevenly, and can hardly be captured reliably with a tape measure. Take instead measures that mean something and that you can repeat at home. In the study of people with reduced muscle strength, these very tests were the outcomes [20]:

• Rising from a chair five times without using your hands, timed in seconds. The most practical strength test there is for the legs.
• Walking speed over a measured distance – four or ten metres at your usual pace, for instance.
• Standing on one leg, eyes open, timed until you put the foot down.
• Timed up-and-go: stand up from a chair, walk three metres, turn, come back, sit down.

Add to that the training data themselves: weight, repetitions and your estimate of the reserve. These notes are the most sensitive early warning system you have – progress and standstill show up there weeks earlier than in any test.

A word of caution that comes straight from the research. The summary of work on the response to training stresses that a single pair of measurements – one before, one after – reflects chance to a considerable extent: measurement inaccuracy and natural day-to-day variation [17]. That holds in the laboratory and all the more so in the living room.

The consequence is simple: measure twice before you start – on two different days, under the same conditions. The difference between those two values is your personal noise. A later change is only worth something once it is clearly larger than that gap. And do not measure weekly but every six to eight weeks; in between, what you are mostly measuring is your form on the day.

20. Nine common misunderstandings

«I'm too old for that.»
People aged 60 and over whose muscle strength was already reduced clearly improved their balance, their walking speed and their ability to rise from a chair in one year of training [20]. The summary of current research states explicitly: chronological age is no reason not to try [17].

«If I don't go to complete exhaustion, it's worthless.»
Not so. When one leg was trained to muscular failure and the other leg of the same person stopped one to two repetitions earlier, the muscle grew equally in both legs – the one leg was simply far more exhausted [3]. Older people are even explicitly advised against going to muscular failure: blood pressure rises sharply, and when you are completely exhausted, clean execution suffers [1].

«Only heavy weights build muscle.»
For muscle growth this is not true: whether people trained with 30 or with 100 per cent of their maximum weight made no difference to muscle growth – provided the set was carried far enough [1]. Anyone who wants to become stronger, that is to move more weight, does need heavy weights. So it depends on what you want to achieve (section 4.2).

«Machines are for beginners, free weights for the advanced.»
In direct comparison, strength, jumping power and muscle growth did not differ [8]. Those who train on machines get stronger above all on machines; those who train with dumbbells, above all with dumbbells. Choose according to your goal, safety, and what you enjoy.

«Lifting slowly achieves more.»
Whether a repetition took half a second or eight seconds led to similar results [12]. And when a study pitted twelve fast repetitions against six slow ones – at the same total duration – both produced the same outcome [13].

«I don't have time.»
The documented minimum is four sets per muscle group per week with three to four basic exercises [2] – that corresponds to two sessions of twenty to thirty minutes. To keep what you have achieved, even less is enough [2].

«Strength training is dangerous for the heart and joints.»
In 23 studies of people with narrowed coronary arteries, all 63 cardiovascular events occurred during endurance training – none during strength training [1]. And in an evaluation of over 38,000 participants, training did not increase the number of serious events [1].

«Strength training damages children's growth.»
Well-designed programmes showed no negative effects on the growth plates, on height growth or on the cardiovascular system [18]. The injuries that appear in accident statistics come predominantly from home equipment that children were handling unsupervised.

«Women should train lighter and with more repetitions.»
Women gain proportionally about as much strength and muscle mass from strength training as men [16]. The recommendation is therefore to adapt the training to the individual person – not to their sex.

21. When to get in touch

Have this assessed by a doctor before starting:

• Chest pain, breathlessness, dizziness or palpitations on exertion.
• A known heart condition, untreated severely raised blood pressure, an aneurysm.
• Recent surgery, a recent fracture, known advanced osteoporosis with vertebral fractures.
• New weakness, numbness or altered sensation in an arm or leg.
• Unintended weight loss, pain at night, fever – regardless of training.

Raise this in physiotherapy:

• You have been training with the same weight and the same number of repetitions for weeks – then it is time to increase.
• You do not know how heavy your programme should be, or it feels too light.
• The pain after exercising does not settle again within a day.
• One side of the body stays clearly weaker than the other and the difference is not getting smaller.
• You are supposed to return to everyday life or to sport after surgery or an injury, but nothing was ever measured.
• You do not trust yourself to make a movement although it would in fact be allowed.

What to expect in our practice: a baseline measurement – strength compared side to side, mobility, simple functional tests such as rising from a chair, walking speed, balance – and from it a programme with defined criteria for progressing, a home programme for the time between appointments, and regular follow-up measurements. And then the handover: at some point the programme should carry on without us. That is not a discharge, it is the goal.

22. In summary

Strength training is the best-documented single measure we have to offer in physiotherapy. The current summary of 137 systematic reviews shows effects on strength, muscle mass, muscle power, endurance, walking speed, balance and everyday functions [1] – and beyond that on metabolism, blood pressure and bone [22].

What makes a programme effective can be listed in five points [1], [2]:

• train at least twice a week;
• work all the large muscle groups – legs, back, chest;
• carry each set far enough that only two to three repetitions would still have been possible at the end;
• perform the movements fully rather than half;
• do at least two sets per exercise – and after some weeks manage more weight or more repetitions than at the start.

What you do not need, by contrast: to go to complete exhaustion [3], particular equipment [8], a particular lifting tempo [12], a sophisticated training plan [1], a dietary supplement [17] – and certainly not a particular age [17], [18], [20].

And what you absolutely do need is not in the tables of the studies but in their footnotes: regularity over years. The biggest gap in this field is not between a good programme and a very good one. It is between a programme and none – and it is on that side of the gap that the majority of the population stands [1].

Two sessions a week, seriously for half a year and permanently thereafter. That is the whole price.

References

Every Digital Object Identifier (DOI) was individually verified against the Crossref register. The links in the reference list lead via the DOI service to the publishers' pages. Some of these are located outside Switzerland and the EU. When you click them, your IP address is transmitted to the provider in question – on our own site this does not happen.

[1] Currier BS, D'Souza AC, Fiatarone Singh MA, et al. American College of Sports Medicine Position Stand. Resistance Training Prescription for Muscle Function, Hypertrophy, and Physical Performance in Healthy Adults: An Overview of Reviews. Medicine & Science in Sports & Exercise. 2026;58(4):851–872. https://doi.org/10.1249/MSS.0000000000003897

[2] Iversen VM, Norum M, Schoenfeld BJ, Fimland MS. No Time to Lift? Designing Time-Efficient Training Programs for Strength and Hypertrophy: A Narrative Review. Sports Medicine. 2021;51(10):2079–2095. https://doi.org/10.1007/s40279-021-01490-1

[3] Refalo MC, Helms ER, Robinson ZP, Hamilton DL, Fyfe JJ. Similar muscle hypertrophy following eight weeks of resistance training to momentary muscular failure or with repetitions-in-reserve in resistance-trained individuals. Journal of Sports Sciences. 2024;42(1):85–101. https://doi.org/10.1080/02640414.2024.2321021

[4] Helms ER, Cronin J, Storey A, Zourdos MC. Application of the Repetitions in Reserve-Based Rating of Perceived Exertion Scale for Resistance Training. Strength and Conditioning Journal. 2016;38(4):42–49. https://doi.org/10.1519/SSC.0000000000000218

[5] Lovegrove S, Hughes LJ, Mansfield SK, Read PJ, Price P, Patterson SD. Repetitions in Reserve Is a Reliable Tool for Prescribing Resistance Training Load. Journal of Strength and Conditioning Research. 2022;36(10):2696–2700. https://doi.org/10.1519/JSC.0000000000003952

[6] Mansfield SK, Peiffer JJ, Hughes LJ, Scott BR. Estimating Repetitions in Reserve for Resistance Exercise: An Analysis of Factors Which Impact on Prediction Accuracy. Journal of Strength and Conditioning Research. Online vorab veröffentlicht am 31. August 2020. https://doi.org/10.1519/JSC.0000000000003779

[7] Graham T, Cleather DJ. Autoregulation by "Repetitions in Reserve" Leads to Greater Improvements in Strength Over a 12-Week Training Program Than Fixed Loading. Journal of Strength and Conditioning Research. 2021;35(9):2451–2456. https://doi.org/10.1519/JSC.0000000000003164

[8] Haugen ME, Vårvik FT, Larsen S, Haugen AS, van den Tillaar R, Bjørnsen T. Effect of free-weight vs. machine-based strength training on maximal strength, hypertrophy and jump performance – a systematic review and meta-analysis. BMC Sports Science, Medicine and Rehabilitation. 2023;15(1):103. https://doi.org/10.1186/s13102-023-00713-4

[9] Aerenhouts D, D'Hondt E. Using Machines or Free Weights for Resistance Training in Novice Males? A Randomized Parallel Trial. International Journal of Environmental Research and Public Health. 2020;17(21):7848. https://doi.org/10.3390/ijerph17217848

[10] Schwanbeck SR, Cornish SM, Barss T, Chilibeck PD. Effects of Training With Free Weights Versus Machines on Muscle Mass, Strength, Free Testosterone, and Free Cortisol Levels. Journal of Strength and Conditioning Research. 2020;34(7):1851–1859. https://doi.org/10.1519/JSC.0000000000003349

[11] Schott N, Johnen B, Holfelder B. Effects of free weights and machine training on muscular strength in high-functioning older adults. Experimental Gerontology. 2019;122:15–24. https://doi.org/10.1016/j.exger.2019.03.012

[12] Wilk M, Zajac A, Tufano JJ. The Influence of Movement Tempo During Resistance Training on Muscular Strength and Hypertrophy Responses: A Review. Sports Medicine. 2021;51(8):1629–1650. https://doi.org/10.1007/s40279-021-01465-2

[13] Martins-Costa HC, Lacerda LT, Diniz RCR, et al. Equalization of Training Protocols by Time Under Tension Determines the Magnitude of Changes in Strength and Muscular Hypertrophy. Journal of Strength and Conditioning Research. 2022;36(7):1770–1780. https://doi.org/10.1519/JSC.0000000000004004

[14] Alarcón-Rivera M, Benavides-Roca L, Salazar Orellana C, Guzmán-Muñoz E. Efectos del entrenamiento cluster sobre la hipertrofia muscular: Una revisión sistemática. MHSalud. 2024;21(1):e16859. https://doi.org/10.15359/mhs.21-1.16859

[15] Vargas-Molina S, García-Sillero M, Maroto-Izquierdo S, et al. Cluster sets and traditional sets elicit similar muscular hypertrophy: a volume and effort-matched study in resistance-trained individuals. European Journal of Applied Physiology. 2025;125(6):1725–1734. https://doi.org/10.1007/s00421-025-05712-6

[16] Cidrais M, Mendonça GV. Sex differences in resistance training: a brief narrative review. Sport Sciences for Health. 2026;22(1):66. https://doi.org/10.1007/s11332-026-01650-8

[17] Roberts MD, Bamman MM, Hornberger TA, et al. Research into resistance training response heterogeneity: a summary of the 2025 conference at the University of Jyväskylä. Journal of Applied Physiology. 2026;140(6):1761–1776. https://doi.org/10.1152/japplphysiol.00289.2026

[18] Stricker PR, Faigenbaum AD, McCambridge TM; Council on Sports Medicine and Fitness. Resistance Training for Children and Adolescents. Pediatrics. 2020;145(6):e20201011. https://doi.org/10.1542/peds.2020-1011

[19] Ni M, Signorile JF, Balachandran A, Potiaumpai M. Power training induced change in bradykinesia and muscle power in Parkinson's disease. Parkinsonism & Related Disorders. 2016;23:37–44. https://doi.org/10.1016/j.parkreldis.2015.11.028

[20] Marzetti E, Álvarez-Bustos A, da Silva Aguiar S, de Oliveira Gonçalves I, Rodríguez-Sánchez I, Coelho-Junior HJ. Evaluating the Impact of a One-Year Exercise Program on Probable Sarcopenia in Older Adults. Journal of Aging Research. 2026;2026:5534766. https://doi.org/10.1155/jare/5534766

[21] Xie C, Yan R, Tao R. Combined resistance training and amino acid-based supplementation for sarcopenia in older adults: a systematic review and meta-analysis. BMC Musculoskeletal Disorders. 2026;27(1):35. https://doi.org/10.1186/s12891-025-09436-8

[22] Westcott WL. Resistance Training is Medicine: Effects of Strength Training on Health. Current Sports Medicine Reports. 2012;11(4):209–216. https://doi.org/10.1249/JSR.0b013e31825dabb8

[23] Maffiuletti NA, Aagaard P, Blazevich AJ, Folland J, Tillin N, Duchateau J. Rate of force development: physiological and methodological considerations. European Journal of Applied Physiology. 2016;116(6):1091–1116. https://doi.org/10.1007/s00421-016-3346-6

[24] Tschopp M, Sattelmayer MK, Hilfiker R. Is power training or conventional resistance training better for function in elderly persons? A meta-analysis. Age and Ageing. 2011;40(5):549–556. https://doi.org/10.1093/ageing/afr005