The Science of Five Reps
Walk into almost any gym and most people train the same way: five sets of an exercise, 10 to 12 reps each, until they're drenched and sore — then sore for two more days at home. It's practically everyone's default.
But there's an easier road with a far higher success rate: three sets of five reps — or even a single set of five.
Even a busy person can train this way: warm up with a few light sets, rest three minutes, then focus on one hard set of five. One exercise a day, done, wrap up — the whole thing takes 15 minutes. Even with very little training each day, you keep making progress.
And even with a solid base and plenty of time to train, the method doesn't need to get complicated: a handful of exercises, three days a week, three whole-body movements per day, each for just 3 to 5 reps, three sets only.
It sounds like too little. But the key isn't "how much" — it's "strong enough." Training requires a stimulus that's strong enough, not a body full of fatigue.
So why does almost everyone believe "more pain, more grind, more effective"? Because fatigue and suffering have been romanticized. Think of the training montage in every movie: the hero, monk-like, drenched in sweat and gritting his teeth, pushes through one gauntlet of pain after another, and finally emerges transformed into someone superhuman. The image is so stirring that we mistake suffering for getting stronger itself. But real training looks nothing like the movies — it's quiet, plain, regular, even a little boring. And the people who thrash themselves the way the movies depict tend to quit the fastest, right after their initial burst of progress.
Training Is Sending Your Body a Message
At its core, training is a message you send your body, asking it to adapt to a higher demand. Take an example: a Zercher split squat with 90 kg for 5 reps. That movement tells the body, "We need to hold up 90 kg on one side." To meet this strength demand, the body has to grow the corresponding equipment: more force-producing myofibrils, recruitment of stronger fast-twitch fibers, and a whole set of matching neural, tissue, and energy-system adaptations.
The same exercise with a different weight and rep count sends a completely different message. Switch to 50 kg for 10 reps and you're no longer transmitting a "strength demand" but an "energy and metabolic demand" — the body reads it as "we need to do many reps at a moderate weight," so it grows more of the energy stores and metabolic machinery inside the muscle cells to handle "lots of repetitions."
And sending the stimulus isn't enough on its own — the body actually rebuilds and gets stronger during the rest and recovery afterward. Training is really three steps:
After adapting, it's not just muscle that gets upgraded. Squatting down and standing back up under 90 kg on one leg sends the message to every tissue involved in the movement, telling them all to get stronger: muscles grow more myofibrils and recruit stronger fast-twitch fibers; the nervous system learns to call up more and higher-order motor units, fire them faster, and coordinate them better — which is also why beginners see strength shoot up in the first few weeks while their muscles have barely grown, because much of getting stronger is a neural "skill" the body learns first [14]. Tendons stiffen so they transmit force without leaking it [15]. Even bones become denser from bearing the load [16]. This whole package of "adaptation up" is something free-weight, multi-joint exercises are especially good at triggering — because they simultaneously demand that you stabilize, balance, and coordinate force across your whole body, engaging a wider range of tissue than isolation work. (Machines aren't useless — they train muscle, nerve, tendon, and bone too; the difference is in movement specificity, stability demand, and degrees of freedom.) It's also why strong lifters build their strength training around a few heavy multi-joint movements, not a pile of fragmented little exercises.
The single-leg squat adds a layer that an ordinary squat can't. With the same external load, standing on two feet with a barbell on your shoulders is stable and symmetrical; but making one side the primary working leg — squatting all the way down and driving back up — puts a far higher relative demand on that leg, and forces the body to keep producing and correcting force in a state that could tip over at any moment. The demands on the nervous system and on balance are much higher.
Back to the message itself. This whole process — demand in, adaptation up — inevitably comes with some soreness and fatigue, but those are byproducts, not the goal. Soreness and tiredness are the cost you pay; getting stronger is the revenue you actually book. Judging a company isn't about how much it spends (spending less is better); it's about how much revenue it brings in. The good companies keep costs down and revenue up — and training works the same way.
So here's the question: training for "strength" (fives) versus training for "size" (10 to 12 reps) — which is better for your health and for resisting aging, and which gives the better cost-to-benefit ratio?
You Both Get Bigger, but Grow Different Things
"Doing fives" isn't really about the number of reps — it's about the weight. A weight you can only lift 5 times is roughly 85% of your 1RM; one you can do 10 to 12 times is only about 65 to 75% [1]. So "fives" = "a near-limit weight," and "twelves" = "a relatively light weight." And heavy and light grow two genuinely different things.
Muscle receives two main types of training stimulus [2]. Heavy weight delivers mechanical tension — the force that directly pulls the fibers taut — which drives the muscle to build more myofibrils, the tissue that actually contracts and produces force. Light weight and high reps deliver metabolic stress — the "burn" from accumulated metabolites during repeated contractions — which drives expansion of the sarcoplasm (the muscle cell's fluid): more glycogen, enzymes, and water, the metabolic machinery.
So "getting bigger" happens two ways, and what grows is somewhat different — and the evidence lines up. In one study, trainees did high-volume, high-rep work; after six weeks the muscle had grown, but the concentration of myosin and actin (the contractile proteins) per unit of muscle had actually dropped — a good chunk of the growth was sarcoplasm, not all contractile tissue [12]. Conversely, training with heavy weight and low reps grew more type 2 fiber cross-sectional area, with contractile protein increasing in step [13]. (Whether "sarcoplasmic hypertrophy" counts as a distinct phenomenon is still debated in the field; but the broad direction — high load biases toward contractile tissue and strength, high volume biases toward sarcoplasm — is well supported.)
Put the most certain point first: to build maximal strength, heavy loads and low reps are clearly more effective — a conclusion that large meta-analyses converge on [5]. The finer distinction — heavy load grows contractile tissue, light load grows sarcoplasm — is a supported but still-debated hypothesis, so don't treat it as gospel. What's certain is the direction: fives use a near-limit weight and are the most favorable for maximal strength; 10 to 12 reps (bodybuilding's favorite "hypertrophy range") is very good at making muscle bigger, and girth genuinely grows, but its efficiency at converting into maximal strength lags behind heavy weight [4]. Which produces a result that seems contradictory at first: if both high and low loads are taken to failure, the "size" they build is roughly similar, yet "maximal strength" still clearly favors heavy load.
What about all those post-workout "sensations"? Each has its own source, and none of them equals results: the in-the-moment burn comes from metabolite buildup, the pump is swelling from blood rushing in, next-day soreness (DOMS) is mostly tied to eccentric load and mild inflammation, and drenching sweat is just the body regulating temperature. Any of these can accompany training, but not one of them alone proves you trained effectively — they're costs or side effects, not the outcome itself. You can train with a huge amount of sensation and grow bigger while your maximal strength barely follows; you can also feel almost nothing while your maximal strength climbs steadily. That's what "lifting without soreness" means: if strength is the goal, you don't need those sensations to prove anything.
The road to bigger muscles is wide — heavy or light both work, as long as the volume is sufficient and each set is close enough to failure; you can even train yourself big and exhausted with light weight and high reps. But the road to greater maximal strength has only one lane and no shortcut: heavy enough. For size, hit the volume and it grows; for strength, heavy is non-negotiable.
The First Thing Aging Takes Is Strength and Type 2
Here we have to ask a bigger question: of these two things, which do you actually need? Especially as you start to age.
First, something many people never realize: what we call "lost youth" is, physiologically, in large part lost type 2 muscle fiber. That youthful feeling of springing up, breaking into a sprint, reacting fast, producing explosive power — it all runs on fast, powerful type 2. And aging takes that first.
Aging doesn't shed muscle evenly. The neurons that control type 2 are large, fast motor neurons, and they degenerate and die first; once the nerve is cut, the type 2 fibers it managed atrophy with it. The result is that type 2 gets killed off preferentially, and maximal strength and power collapse disproportionately [17]. Areas rich in fast-twitch fiber, like the thighs, decline faster than anything else.
And here's the most crucial point: what really determines your later life is the quality of your muscle, not the quantity. One large study tracking nearly two thousand older adults for three years found that leg strength declined 3 to 4% per year — three times faster than the loss of muscle mass (about 1% per year); more striking still, even in years when muscle mass increased, strength kept falling [18]. Another study is even more direct: what predicts mortality in older adults is strength, not muscle mass [19].
The look of aging is tied to strength too. Picture two 70-year-olds: one shuffles along, hunched and unsteady; the other stands tall and moves briskly — from behind, they could pass for 40. What separates them isn't the size of their muscle; it's how much force-producing muscle they've kept, especially type 2.
Then there's the killer of the elderly: falls. Falls are rarely caused by "not enough muscle" alone — more often it's "too slow to react": when your foot slips, can you summon enough force within a fraction of a second while correcting your balance to stay upright? That contest is won by neural reaction speed, balance control, and instantly explosive type 2 — all three together. And that is exactly what a loaded single-leg squat trains: producing force in instability, correcting while producing force. It trains not just muscle but unilateral strength, postural control, and balance — the very foundations that "staying steady and reacting in time" depend on.
Connect it all and it becomes clear: aging isn't just muscle dropping away — it's type 2 atrophying, neurons dying, and bone thinning, the whole system declining together. And to hit the brakes on this entire slide at once, heavy enough strength training is about the most comprehensive single move there is — it fights sarcopenia, neural degeneration, and osteoporosis simultaneously (research confirms that progressive resistance training can restore both strength and hip bone density in older adults, and the heavier the load, the better for bone density) [16].
So back to the question: what aging takes is exactly what fives training gives — maximal strength, type 2, force-producing myofibrils, and the integrity of the whole system. The 10-to-12-rep, girth-focused style isn't useless against aging (muscle mass is lost too), but what you lose fastest and most fatally is strength and type 2 — and that's what to protect first. Against the strength, type 2, and bone that aging strips away, heavy enough strength training is the road with the most complete evidence and the most consistent results. What it trains, rep by rep, is winning that lost youth back.
Weight Decides Which Fibers You Use — and What It Costs
Earlier we said "weight grows type 2." Here's the mechanism behind it — and an answer to a question you might have: can't I also train type 2 with light weight, lots of reps, taken to failure? You can, but the price is very different.
Human muscle fibers come in roughly three kinds. Slow-twitch (Type I) contract slowly and produce little force, but barely fatigue — running on aerobic metabolism, they can go for hours, and they handle low-intensity daily work like walking and standing. Fast-twitch come in two kinds: Type IIa is the in-between type; Type IIx is the other extreme, and the strongest card the body holds. The gap is concrete in numbers: IIx contracts several times faster than slow-twitch, and its peak power is roughly ten times that of slow-twitch [8][9]. But such powerful fiber is expensive to use once — it runs on phosphocreatine and anaerobic glycolysis, and an all-out effort is exhausted in a few seconds to a minute; a single bout to failure burns phosphocreatine down to about 15% of its stores, and fully replenishing it takes oxygen and roughly three to five minutes [10]. So IIx is a scarce resource the body "rations" — the trump card it plays when chased by a lion, never needed for strolling around (one study actually measured lion muscle fiber and found their IIx produces about three times the power per unit area of a well-trained human — no wonder you can't outrun one [11]).
We're not usually chased by lions, so when does the body play this card? You have to actively produce force, and that depends on "how much force is needed right now." This is the size principle [3]: fibers are recruited in order of size, small to large, according to the force demand. With a weight you can only lift 5 times, the demand for force output is high from the very start — slow-twitch can't handle it, so the body quickly mobilizes the highest-order motor units and brings the fast-twitch fibers into play. You're training the most valuable fibers almost from the first rep, and you don't have to run yourself into the ground.
A light weight you can do 12 times is a different story: for the first few reps the force demand is low, the higher-order motor units aren't needed yet, and the body sends out the cheap slow-twitch fibers to carry the load. You have to grind those slow-twitch fibers toward exhaustion, close to failure, before the high-threshold units governing fast-twitch are forced to join in. So high reps aren't incapable of training IIx — you just have to eat all the fatigue of the first dozen reps to earn its participation in the last few.
To train the same type 2, low reps get it at the lowest cost from the very start, while high reps take the long way around and leave you half-dead to get there. Framed by the question "can you keep training for many years," this cost gap is what decides the outcome.
So Why Exactly Five — Not 12, and Not 1?
If heavy beats light and the direction is clear, then how many reps exactly? Why not a higher 12, and why not a lower 1 or 2?
First, why not go higher. Beyond the "long way around, high cost" already covered, high reps have a very practical problem: you can't gauge intensity accurately. For training to work, every set has to be pushed close enough to failure (coaches often describe this with RPE — RPE 8 is roughly "1 to 2 reps left"). But that's hard to do at high reps — when a set takes 10 or 12 reps to reach failure, the soreness, breathlessness, and burn often make you stop before you're truly near the limit, without quite realizing it. In one study, subjects took a set to failure at 70% 1RM (averaging 16 reps) and were asked to call out when they believed they had "5 reps left, 3 left, 1 left": when they were still far from failure, their judgment was off by as much as 5 reps [6]. In other words, at high reps you think you nailed it, but you may be stopping every set four or five reps short of failure, quietly discounting your intensity. Low reps are far more accurate — the rep target is clear and failure is right in front of you, so it's harder to fool yourself (though not impossible); and because of that, you'll dare to use, and can actually handle, heavier weight.
Now, why not go lower. Wouldn't heavier weight for just 1 or 2 reps be even more hardcore? Push into that range and the stimulus is indeed stronger, but the cost is higher too: these weights are extremely close to the limit, the margin for error is tiny, any slight breakdown in form has to be absorbed the hard way, and it demands more technique, safety, and mental preparation; accumulating stimulus across many sets is harder as well. And a large part of strength is movement skill, which needs many repetitions under heavy enough load — at 1 or 2 reps you barely get any. For a generally healthy person who wants to train for the long haul, these extra costs usually aren't worth it.
Five reps lands right on the sweet spot for all these conditions: heavy enough to deliver high mechanical tension and quickly recruit high-order motor units (that's training strength); yet few enough to dial in intensity precisely, maintain technique safely, and still accumulate stimulus across many sets. Too heavy and it's dangerous and hard to master; too light and the intensity gets discounted and you take the long way around. To be clear: 5 isn't a magic number — anything from 1 to 8 reps has value for strength, and it varies slightly by exercise (heavy multi-joint movements often land at 3 to 6 reps). It's simply the practical compromise where "the weight is heavy enough, each set is short enough, and technique and reserve are preserved" — the real point has always been "heavy enough, few reps," not landing exactly on 5.
Important: "dialing in intensity" is not the same as "taking every set to failure." What you want is to approach failure (about 1 to 2 reps left), not hit the wall every set. Deliberately training to failure sharply raises fatigue and slows recovery, with no added benefit for a strength goal.
The Real Key: Can You Keep Training Forever?
All the costs and benefits above ultimately funnel into one ledger: the long term.
Training is a cycle you repeat for many years — stimulus in, go home and recover, stimulus in again. What truly decides whether you get stronger isn't how hard you went on any one day, but how many years you can repeat this cycle. Train ferociously and burn out in three months, and it's as if you never trained; train just right for ten uninterrupted years, and you've genuinely remade your body. So any friction that makes this cycle harder, more painful, or less appealing to come back to is a source of failure. Which is why you choose the "dose" you can complete for the long haul: one set of five isn't the ideal volume — a few more sets would give more stimulus — but it's the minimum effective dose, and one set you can finish every day beats five sets that are theoretically perfect but never actually done.
Seen this way, the benefits of fives are no longer just physiological. It's easy — a set lasts a few seconds, leaves you barely sore or tired, and you finish with enough energy left to live your life and the appetite to come back tomorrow. It's low-pressure — none of the on-the-edge tension of 1 or 2 reps, none of the dread of a set of 12 that tires you out before you even begin. And it delivers something rarely mentioned but crucial: a sense of progress. Low reps plus gradual added weight means you're always going head-to-head with a concrete number — 2.5 kg more than last week, clear as day. That feeling of "I'm progressing" is the strongest fuel for wanting to keep training. Chasing soreness and the pump, by contrast, gives progress that's hard to quantify, and so it lacks that feedback.
The newest and largest body of evidence puts this plainly. In 2026, the American College of Sports Medicine (ACSM) published its first revised position stand on resistance training in seventeen years, integrating 137 systematic reviews and data on more than thirty thousand people [7]. It doesn't say "it's all the same however you train" — quite the opposite: it gives goal-specific prescriptions. To build maximal strength, you need heavy enough loads (80% 1RM and above), 2 to 3 sets per exercise. It even calls out directly that many people who think they're "training strength" are actually using lighter weights and higher reps, building only endurance and muscle size, with strength never really increasing.
What it says "doesn't matter as much" isn't the load itself — load is always the key to strength — but the details that have been over-mythologized: whether to train to failure (leaving 2 to 3 reps in reserve actually recovers better, with the same effect), barbell versus machine, whether to run complex periodization. Take care of the foundation — heavy enough, enough volume, regular — and you needn't fret over the rest. And fives happen to be the spot in that foundation that takes the least effort and is easiest to stick with long-term; being able to keep training is the only way to turn a foundation into results.
Soliday implements the same core loop: five-rep strength work, gradual loading, and clear progress. It logs each set as you train and, following "complete all sets → add weight; miss → repeat or deload," calculates the weight to use next. Free, no account required. You just focus on the next clean set.
This Isn't One Amateur's Opinion
You might be thinking: this sounds reasonable, but is it just one person's take? Quite the opposite. Look at the best-selling strength books and the most widely run strength programs in the field, and you'll find nearly all of them are built on "fives."
Mark Rippetoe's Starting Strength, the acknowledged beginner's bible, centers on a few big barbell lifts done for three sets of five (3×5) — add weight, log it, repeat. Jim Wendler's 5/3/1, one of the most widely run intermediate programs in the world, has three sets of five in the first week of its four-week cycle and is built around the low rep counts of 5, 3, and 1. Michael Matthews's Bigger Leaner Stronger, one of the best-selling fitness books for years, also has a main program of heavy weight for 4 to 6 reps per set (about 85% 1RM).
More interesting still, beyond similar rep counts, the philosophy behind these programs is identical to what we've laid out: start light, add weight slowly, avoid taking every set to failure, keep the exercise selection minimal, and then run it for the long haul. Whether you reason down from the physiology or converge up from decades of hands-on experience, the answer lands in the same place — heavy enough weight, few reps, and keep training.
So: Three Sets of Five, Just Do a Little
For both "getting stronger" and "fighting aging," five is the point that's just right on every count — heavy enough that what grows is force-producing myofibrils, maximal strength, and type 2 (intensity dialed in); each set easily approaches failure (consistent); with reserve to spare for technique and low risk (safe); low metabolic cost, a set over in seconds (comfortable). And precisely because of this, it's the one most likely to keep you going year after year — and consistency is what all of this truly hinges on.
This isn't the only way to train, of course. If you love high reps, enjoy the pump, or you want to build size, that road works too — the trade-offs are just different. But if what you want is real strength, and a young body that can still move freely in old age, then try this: pick a few big compound movements, do three sets of 5 reps per exercise, add a little weight each session and adapt slowly, and when you're down to only 1 or 2 reps in reserve, deload and build back up. The exact rhythm of adding weight has its own article; how to choose the exercises has its own article too.
One more thing to be clear about: strength training isn't everything. If you could pick only one, it would be this — strength and type 2 are what aging takes first and what's hardest to restore any other way; but the ideal longevity mix is resistance training to hold onto strength and fast-twitch fiber, aerobic training to maintain heart, lungs, and metabolism, plus some fast, explosive movements to preserve power (especially important for mobility in older age). They aren't in competition — they're complementary.
Do little, without soreness or sweat — yet every rep is winning back, bit by bit, the strength, the type 2, and the young body that aging is taking away.
Further Reading (a Three-Part Series)
- This article (Part 1): The Science of Five Reps — why heavy enough and few reps is the best value
- Part 2: How to Add Weight — how to load up precisely and continuously
- Part 3: Choosing the Exercises — which movements to pick, and why
References
- [1] Nuzzo JL, Pinto MD, Nosaka K, Steele J. "Maximal Number of Repetitions at Percentages of the One Repetition Maximum: A Meta-Regression and Moderator Analysis of Sex, Age, Training Status, and Exercise." Sports Med 54(2):303-321, 2024 — roughly 7,000 subjects; about 85% 1RM averages 6 reps and 80% about 9, with achievable reps varying by person and by exercise.
- [2] Schoenfeld BJ. "The mechanisms of muscle hypertrophy and their application to resistance training." J Strength Cond Res 24(10):2857-2872, 2010 — three main drivers of muscle growth: mechanical tension (the primary driver), muscle damage, and metabolic stress.
- [3] Henneman E. "Relation between size of neurons and their susceptibility to discharge." Science 126:1345-1347, 1957 — the size principle of motor units: recruited in order from small to large, with large units mobilized only under high load or near failure.
- [4] Campos GE, et al. "Muscular adaptations in response to three different resistance-training regimens: specificity of repetition maximum training zones." Eur J Appl Physiol 88(1-2):50-60, 2002 — the low-rep group (3-5RM) gained the most maximal strength; the very-high-rep group (20-28RM) gained the most muscular endurance.
- [5] Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. "Strength and Hypertrophy Adaptations Between Low- vs. High-Load Resistance Training: A Systematic Review and Meta-analysis." J Strength Cond Res 31(12):3508-3523, 2017 — when both high and low loads are taken to failure, hypertrophy is similar, but maximal strength clearly favors high load.
- [6] Zourdos MC, et al. "Proximity to failure and total repetitions performed in a set influences accuracy of intraset repetitions in reserve-based rating of perceived exertion." J Strength Cond Res 35(Suppl 1):S158-S165, 2021 — the further from failure and the more reps in a set, the less accurate the judgment of "reps remaining."
- [7] Currier BS, et al. "Resistance Training Prescription for Muscle Function, Hypertrophy, and Physical Performance in Healthy Adults: An Overview of Reviews." Med Sci Sports Exerc 58(4):851-872, 2026 — 137 reviews, more than 30,000 people: maximal strength requires high load (80% 1RM and above); training to failure, equipment, and complex periodization have limited impact.
- [8] Schiaffino S, Reggiani C. "Fiber types in mammalian skeletal muscles." Physiol Rev 91(4):1447-1531, 2011 — the authoritative review on muscle fiber typing.
- [9] Widrick JJ, et al. "Force-velocity and force-power properties of single muscle fibers from elite master runners and sedentary men." Am J Physiol 271(2 Pt 1):C676-C683, 1996 — in human single fibers, peak power of IIx is about 2x that of IIa, and IIa about 5x that of Type I.
- [10] Harris RC, et al. "The time course of phosphorylcreatine resynthesis during recovery of the quadriceps muscle in man." Pflügers Arch 367(2):137-142, 1976 — after failure, phosphocreatine drops to about 15%, and recovery depends on oxygen and takes several minutes.
- [11] Kohn TA, Noakes TD. "Lion and caracal type IIx single muscle fibre force and power exceed that of trained humans." J Exp Biol 216(6):960-969, 2013 — lion IIx single-fiber power is roughly 3x that of humans.
- [12] Haun CT, et al. "Muscle fiber hypertrophy in response to 6 weeks of high-volume resistance training in trained young men is largely attributed to sarcoplasmic hypertrophy." PLoS ONE 14(6):e0215267, 2019 — after six weeks of high-volume training, the growth was largely sarcoplasmic.
- [13] Roberts MD, et al. "Sarcoplasmic Hypertrophy in Skeletal Muscle: A Scientific 'Unicorn' or Resistance Training Adaptation?" Front Physiol 11:816, 2020 — high load tends to grow contractile protein and strength in step (the topic remains debated).
- [14] Aagaard P, et al. "Increased rate of force development and neural drive of human skeletal muscle following resistance training." J Appl Physiol 93(4):1318-1326, 2002 — a large share of strength gains comes from neural adaptation.
- [15] Bohm S, Mersmann F, Arampatzis A. "Human tendon adaptation in response to mechanical loading." Sports Med Open 1:7, 2015 — the tendon's main adaptation is stiffening, and high load is most effective.
- [16] O'Bryan SJ, et al. "Progressive Resistance Training for Concomitant Increases in Muscle Strength and Bone Mineral Density in Older Adults." Sports Med 52(8):1939-1960, 2022 — progressive resistance training can raise both lower-limb strength and hip bone density in older adults, and heavier loads are more favorable for bone density.
- [17] Larsson L, et al. "Sarcopenia: Aging-Related Loss of Muscle Mass and Function." Physiol Rev 99(1):427-511, 2019 — aging is driven largely by motor neuron loss, with fast-twitch (type II) fiber damaged preferentially.
- [18] Goodpaster BH, et al. "The loss of skeletal muscle strength, mass, and quality in older adults." J Gerontol A Biol Sci Med Sci 61(10):1059-1064, 2006 — leg strength declines about 3-4% per year, roughly three times the rate of muscle mass loss, and gaining muscle can't stop the strength decline.
- [19] Newman AB, et al. "Strength, but not muscle mass, is associated with mortality in the health, aging and body composition study cohort." J Gerontol A Biol Sci Med Sci 61(1):72-77, 2006 — what predicts all-cause mortality is strength, not muscle mass.