← All guidesTraining 101: What Actually Decides Whether Training Works?
About six variables decide whether training works, and they are nowhere near equally important. The questions people agonise over — which split, how many days, how slow the reps — sit at the bottom of that list. This page puts them in order, and points at the guide that settles each one.
Updated August 11, 2026
·
Author-reviewed
The answer, with evidence
1. Whether you keep doing it: the heaviest-weighted variable, and the bar is far lower than most people assume — on both sides.
Aerobic side: in a pooled analysis of 661,137 adults and 116,686 deaths, even falling short of the recommended minimum came with 20% lower all-cause mortality; hitting 1–2× the minimum gave 31%, and the benefit topped out around 3–5× (39%). [1] Strength side: across 16 cohorts, muscle-strengthening activity tracked 10–17% lower mortality, the J-shaped curve peaking at roughly 30–60 minutes a week. [2] Same shape twice: the stretch from nothing to something is the steep part.
2. Total volume and progression: what actually drives adaptation, and it pays less with every added unit.
The lifting side is measured most finely: over 67 studies and 2,058 participants, the probability that more weekly sets means more hypertrophy and more strength was 100%, while both best-fit curves showed diminishing returns, considerably more pronounced for strength. [3] On the aerobic side the unit is weekly minutes, and the curve bends the same way [1]. "More works" and "more is worth progressively less" are both true.
3. What you actually load: a muscle you never train and an energy system you never touch both sit out the rest of the plan.
Definitional, not a finding — tissue adapts to the load it receives, and no trial needs to establish that. On the lifting side the gap is a skipped muscle group (the map below is that check); on the aerobic side it is living entirely in the middle, never going properly hard or properly long and easy. It ranks this high not because the evidence is strong but because it is the only variable that fails to zero.
4. How hard the work is: proximity to failure when lifting, how far the cardiorespiratory system gets pushed when doing cardio.
What intensity buys is fitness, and fitness tracks survival steeply: across 122,007 adults, the least fit faced roughly five times the all-cause mortality risk of the most fit — a larger gap than coronary artery disease, smoking, or diabetes. [4] "So train harder" is inferred from that: the study graded fitness, it did not randomise anyone to an intensity. It sits fourth because intensity governs how efficiently a given volume converts, not whether anything happens at all.
5. Recovery and fuel: not the opposite of training but the conditions on which volume turns into adaptation.
Recovery splits by timescale, one guide each: between sets, between days, and sleep. On the fuel side the firmest number is protein — across 49 randomised trials and 1,863 participants, supplementation added 2.49 kg to 1RM and 0.30 kg of fat-free mass: real, but small, and beyond a total intake of about 1.62 g/kg/day more protein added nothing further. [5] Eat enough and the variable is spent; eating more does not buy more.
6. Everything else is tuning: split, frequency, rep speed — hold total volume constant and the differences largely vanish.
25 studies: weekly frequency had no significant effect on hypertrophy on a volume-equated basis [6]; 14 studies and 392 subjects: split and full-body routines differed significantly on neither strength nor muscle growth [7]; 8 studies: repetition durations anywhere from 0.5 to 8 seconds produced similar hypertrophy. [8] Three independent nulls — the hardest evidence on this page, and worth saying plainly that all three come from the lifting literature; the aerobic side has no equally clean set.
The six modules, and where each one is settled
Treat two sessions of twenty minutes a week as the starting point, not as a placeholder until the real programme begins. Whether a plan survives contact with a busy week matters more than how good it looks on paper — pick the version you could still finish during your worst week.
02
Count the volume, then grow it
On the lifting side, count hard sets per muscle per week; on the cardio side, count weekly minutes and how many of them are hard. Fix a number you can currently recover from, then add to it every few weeks — more load, more reps, or more sets all count.
03
Check coverage, do not study anatomy
You do not need to memorise origins and insertions. You need to do two things: look at a body map, find the regions this week never loads, and add one movement for each; then look at how your cardio is distributed, because if every session lands in the same slightly-breathless middle, what is missing is both ends — genuinely easy long work, and genuinely hard intervals. The next section is that map.
04
Let the hard part be hard
When lifting, take sets to within one to three reps of failure. When doing cardio, keep easy work genuinely conversational and hard work genuinely hard — you want both ends, rather than everything parked in the middle.
05
Recover at three timescales, and eat enough
Rest between sets decides the quality of today’s session, rest days decide the load the week can hold, and sleep sets the ceiling on both — the third is the one most often sacrificed and the one that costs the most. On the fuel side there is exactly one thing to manage deliberately: get protein to roughly 1.6 g per kg of bodyweight, past which more buys nothing.
Split, weekly frequency, rep speed, warm-up detail — worth getting right, not worth agonising over while the first five are still unsettled. Come back to these once the rest holds steady.
※
Across both: fitting lifting and cardio together
Not a seventh module but the intersection of the first six once both kinds of training exist in one week. The good news is that the cost is far smaller than the folklore: across 43 studies and 1,090 adults, neither muscle size (−0.01) nor maximal strength (−0.06) dropped significantly, and only explosive power was meaningfully blunted (−0.28) — mostly when the two shared a session. [9]
Which muscles are you actually training?
The third module is the easiest to miss and the easiest to fix, for one reason: you can feel volume and you can feel intensity, but you cannot feel coverage. Someone who only presses and never pulls can train extremely hard and give their back nothing, and nothing in the session itself will say so. It needs to be looked at, not felt.
The map below is that check. Go over the front and back views marking what this week actually loaded; whatever is left over is the gap. Tapping any region opens that muscle group’s guide, with plain-language anatomy and exercises that need no equipment.
Select a region on the body or use a named button.
If a list suits you better, the
12 muscle-group guides are sorted into upper body, trunk, and lower body; to see what a specific movement trains and how to do it properly, read the
exercise guides.
What this page does not cover
Exercise and health science is a good deal larger than training. Module 5 above touches only the sliver of nutrition that directly modifies training adaptation — whether you eat enough protein. The rest matters just as much and has no guide here, so this page names it rather than pretending to have covered it:
Total energy intake, fat loss and body composition — what mostly decides bodyweight and body fat is this, not training. It deserves its own body of content, not a paragraph inside a training guide.
The rest of the diet beyond protein — carbohydrate and performance, meal timing, micronutrients: none of it is covered here.
Injury and rehabilitation — needs an in-person assessment, not an article. See a professional for pain that persists.
Settle showing up and settle how much work you do, then start worrying about splits, frequency and rep speed. Doing it in the reverse order is where most people stall.
What Omini handles is the part of this list where execution slips: calling the intervals, counting the sets, and announcing every work and rest period out loud, so you never have to watch a screen mid-session.
Start your first session
Frequently asked questions
What actually matters most in training?
Roughly in this order: doing it at all and keeping it up, how much total work you do and whether it grows over time, which muscles or energy systems you actually load, how hard the work is, and how much you recover. Everything else — training split, weekly frequency, rep tempo, exercise order — is tuning. Meta-analyses that hold total volume constant repeatedly find no meaningful difference between splits and full-body routines, between one and three sessions per muscle per week, or between rep durations from 0.5 to 8 seconds.
How much exercise do I need before it counts for anything?
Far less than most people assume, and the first bit is worth the most — on both sides. For aerobic activity, a pooled analysis of 661,137 adults found 20% lower all-cause mortality even below the recommended weekly minimum, 31% at one to two times it, and a ceiling around three to five times. For lifting, 16 cohort studies put muscle-strengthening activity at 10–17% lower mortality with the curve peaking at roughly 30–60 minutes per week — two short sessions. Both curves have the same shape: the gap between zero and something is much larger than the gap between something and a lot.
Do I need to learn anatomy to train?
Not in detail, but you do need to know what you are and are not loading, because a muscle you never train cannot adapt no matter how good the rest of your programme is. The practical version is a coverage check rather than a course: look at a body map, find the regions nothing in your week trains, and add one movement for each. The muscle map on this page opens a plain-language guide for any region you tap.
Should I do cardio or lifting first?
Do the one that matters most to you right now first, and only worry about it when both land in the same session. An updated meta-analysis of 43 studies found concurrent training left muscle size and maximal strength essentially untouched; only explosive power was meaningfully blunted, and mostly within a single session. Spacing the two by a few hours removed most of that.
Is more training always better?
More volume does keep helping, but with clearly diminishing returns. A 2026 meta-regression over 67 studies and 2,058 participants found the probability that more weekly sets means more muscle and more strength was effectively 100%, while both curves flattened as volume rose — sharply so for strength. So added sets keep paying, just less each time, and the cost in recovery does not flatten with them.
How much does diet matter compared with training?
For bodyweight and body fat, diet is the bigger lever by far. For whether training itself produces adaptation, the piece that matters most is simply eating enough protein — and its effect is real but modest. Across 49 randomised trials and 1,863 participants, protein supplementation added 2.49 kg to one-rep max and 0.30 kg of fat-free mass, with no further gains once total intake passed about 1.62 g per kg of bodyweight. So protein is a variable you satisfy rather than optimise. Beyond that, energy intake, fat loss, meal timing and micronutrients are not covered by these guides.
References