Count the hard sets one muscle gets in a week, direct and indirect, and see how they are spread across your sessions.
Estimates are for general guidance only and are not medical advice. For a personalised plan, book a free guided tour at RPM Gym.
The headline figure is your direct sets plus half your indirect ones. Twelve direct and six indirect reads as 15 — not 18, and not 12. It describes work arriving at one muscle across a week, which is the quantity the growth research is modelled on.
Everything on this page hangs on one word in both input boxes: hard. A hard set is one taken close enough to failure to count, where another rep at the same technique was genuinely in doubt. A set finished with five in reserve is not the same input, and feeding it in is the quickest route to a large number attached to a muscle that has not changed in months.
The half is a counting convention, and calling it anything more would be a claim nobody has the evidence for. When Pelland and colleagues fitted their dose-response meta-regressions, they tested three ways of handling indirect sets: count each one fully, count it as half, or ignore it. The fractional version fitted the data best, so that is what this page reports. A compound lift trains several muscles at once, and that fraction describes how the sets were counted in a model — not how much stimulus a bench press delivered to your triceps.
What the page will not do is tell you the number ought to be higher. More weekly hard sets is associated with more hypertrophy on average, with diminishing returns as the count climbs, and there is no published figure at which it stops working and no published optimum. This tool used to name one anyway: 10-12 weekly sets for beginners, 12-18 for intermediates, 16-22 for advanced lifters. None of the three had a traceable source, and they overlapped at 16-18, so a beginner and an advanced lifter could be handed the same prescription by a tool whose entire premise was that experience changes the dose. Those bands are gone, and no other invented range has taken their place.
Sessions and the rough per-session figure are there so you can see the shape of the week rather than score it. Sixteen hard sets in one session and sixteen across three sessions are the same weekly total and rarely the same training, because the last sets of a long session are the first ones to become hard in name only.
Research-weighted weekly sets = direct hard sets + (0.5 × indirect hard sets) Roughly per session = (direct + indirect) ÷ sessions The 0.5 is the fractional quantification the dose-response modelling fitted best. It is a rule for counting, not a conversion between two kinds of set.
Triceps across a pressing-heavy week: 6 direct sets of pushdowns and overhead extensions, 10 indirect sets from bench and overhead pressing, spread over 3 sessions.
Eleven weighted sets, and most of the work came from pressing rather than from anything chosen for the triceps. Count only the isolation work and they look neglected at 6. Count everything at face value and they look hammered at 16. The middle figure is the one the dose-response models were fitted on, and the gap between 6 and 16 is exactly why the direct and indirect counts stay on screen next to it.
There is no published optimum, and any page handing you one has invented it. What the dose-response evidence supports is an association: more weekly hard sets goes with more growth on average, with diminishing returns as the count rises. The 2017 meta-analysis that made this a mainstream question sorted its studies into fewer than five, five to nine, and ten or more weekly sets — categories for grouping research, not doses to pick from. Count what you are doing, change it deliberately, and judge it against your own last block.
It used to, and the answer it produced was made up. The old version prescribed 10-12 weekly sets for beginners, 12-18 for intermediates and 16-22 for advanced lifters, with no traceable source behind any of them, and with two of the bands overlapping at 16-18 — so the tool disagreed with its own premise that experience should change the dose. In the meta-regressions, training status is a covariate the models adjust for, not a basis for prescribing a different number of sets. The field was removed rather than re-tuned.
One where the next rep at the same technique was genuinely in doubt. Research comparing sets taken to failure with sets stopped a rep or two short finds the difference between them small; sets stopped five or six short are a different input altogether, and they are not what the volume research counted. If you cannot tell which kind of set you did, leave it out — an inflated count is worse than a cautious one, because it makes an under-trained muscle look well served.
No, and nothing here says so. Counting it as half is what the dose-response modelling did, tested against counting it in full and against ignoring it, and it fitted the data best. That is a finding about a model's bookkeeping. How much your triceps actually get from a heavy bench press compared with a pushdown is not a question these numbers answer.
Ask what the exercise was in the session to do. If you chose it to train that muscle, the set is direct; if the muscle worked hard while something else was the target, it is indirect. Pressing is direct chest work and indirect triceps work. A row is direct back work and indirect biceps work. Compound lifts spread work across several muscles, which is the whole reason the two columns exist — keep your rule consistent and the figure stays comparable from week to week.
There is no published exchange rate, so this page will not invent one. Count the set you took to the point of real difficulty as one hard set, and leave the extension out of the total rather than assigning it a fraction of your own design. Your count is only comparable to your last count while the counting rule holds still.
Frequency is modelled as its own question in the same dose-response work rather than as a multiplier on weekly volume, and this page counts the total rather than scoring the split. What the per-session figure is good for is spotting a session that has quietly become long: when one muscle is taking eight or ten hard sets in a single session, the sets at the end are the ones worth checking before the weekly total is.
Nobody is a reliable judge of their own last two reps, which is precisely where a coach standing behind the rack earns their money. Book a free guided tour at RPM Gym in Al Manhal.