A Fractional Set Is Half a Set. Count Yours Right.

A fractional set is an indirect set counted as half instead of counted as one. Your bench press pays your triceps something, and the fractional method says the honest figure to write down is 0.5.

The Short Answer

Count a set at one when the muscle you are counting drives the lift, and at a half when it only helps. Six sets of triceps pushdowns in a week are six, and the fourteen sets of pressing on top of them are seven more, not fourteen and not zero. The 0.5 is a modelling choice that fit the research better than the alternatives, not a number anybody measured, and that distinction is the point of this page.

An Indirect Set Is One Where the Muscle Helps and Does Not Lead

Direct sets count as one. Indirect sets count as half. That is the entire convention, and everything after it is deciding which is which.

The Sports Medicine paper that introduced the method defines an indirect set for muscle growth as one in which the measured muscle "was likely to be meaningfully trained but not the primary force generator of the exercise (i.e., synergist)". Your triceps on a bench press. Your biceps on a lat pulldown. Your front delts on almost any press you did not choose for your front delts. For strength the paper tightens the definition further, counting only the exact lift being tested as direct, so leg press sets count as half each towards a squat one-rep max.

Those labels were applied by hand by the review team, who write that the process "was not wholly objective". If you cannot decide whether a muscle is leading or helping on some exercise, you are in the position the authors were.

The 0.5 Came From One Paper That Counted the Same Studies Three Ways

Sixty-seven resistance training studies, 2,058 participants. Every set in every study was classified as direct or indirect for the muscle being measured. Then the same body of research was counted three separate times, once with indirect sets weighted at 1, once at 0.5, once at 0.

The half-set version predicted both muscle growth and strength better than either alternative, on all four model comparisons the authors ran. Their own reading of that result is that on average an indirect set delivers a stimulus close to half that of a direct set.

The paper's own worked example: a study measures biceps growth, participants do five sets of curls in one session and five sets of rows in another, and that week counts as 10 total, 7.5 fractional, and 5 direct.

Now the part most pages repeating this number leave out. The paper's Key Points call it "The novel fractional quantification method". In the discussion the authors write that assigning indirect sets a weight of 0.5 "is still an assumption", and that the method "should be regarded as a heuristic to improve the accuracy of dose-response modeling, rather than a definitive standard for practical application across all contexts". They name four things that should move the right weight: the outcome being measured, the exercise, the rep range, and the training status of the lifter.

So the support for 0.5 is model fit, not experiment. Nobody has run a trial pitting 0.5 against 0.25 or 0.75 as a training prescription, and until somebody does it is the best supported of three crude options rather than a constant. Use it because it beats the two things lifters actually do, which are counting a bench press as a full triceps set or not counting it at all.

Yes, a Bench Press Counts for Your Triceps, and Half Beats Both One and Zero

Yes, it counts, and half is a better guess than either whole number. The trial that pulls the two apart most cleanly ran ten weeks with 43 men and measured cross-sectional area by MRI, in the Journal of Strength and Conditioning Research.

Bench press on its own moved whole triceps cross-sectional area 4.8 percent, which did not reach significance (p = 0.20). Lying barbell triceps press on its own moved it 9.5 percent (p < 0.0001). Split the muscle by head and the picture stops being tidy. The bench built the lateral head (7.2 percent, p = 0.02) where the triceps press did nothing measurable (0.6 percent). The triceps press built the long head (17.5 percent) where the bench did nothing measurable (2.1 percent).

Worth one sentence and no more: 4.8 against 9.5 is almost exactly half, but the 4.8 was not statistically significant, the groups held only 10 to 12 men each, and the between-group comparison found no significant difference in whole triceps growth between any of the four arms. The participants had also done no regular lifting for at least six months beforehand, so this is a detrained result rather than a trained one.

Point the same question at the biceps and the answer flips. Twenty-nine untrained men did lat pulldowns and bench presses for ten weeks, half of them with curls and elbow extensions added on top. Elbow flexor thickness went up 6.5 percent without the isolation work and 7.04 percent with it. A null result in a small trial of untrained men is not proof that the two are equal, but it is a second data point against any flat rule.

A helper muscle gets something. It does not get everything. How much depends on the muscle, the exercise and who is lifting, which is the honest case for a fraction rather than a whole number or a zero.

The Same Training Week Is 6 Sets, 13 Sets or 20 Sets

Take a session you may already be running. Bench press, 4 sets. Barbell row, 4 sets. Overhead press, 3 sets. Lat pulldown, 3 sets. Triceps pushdown, 3 sets. Dumbbell curl, 3 sets. Twenty sets, run twice a week.

Count your triceps the way most lifters do, which is to count the pushdowns and nothing else, and you get 3 sets a session and 6 a week.

Count every set that touches the triceps at full value and you get 10 a session and 20 a week.

Count fractionally and you get 3 direct plus 7 indirect at a half each, which is 6.5 a session and 13 a week.

Biceps come out identically: three curl sets direct, seven sets of rowing and pulling at a half, 13 a week.

Same twenty sets. Same Tuesday. Three answers that differ by more than a factor of three, and nothing biological separates them.

Every weekly set target you have read is denominated in somebody's counting convention, and the target rarely says which one. In the Journal of Human Kinetics, a systematic review of seven randomised trials in 241 trained men named 12 to 20 weekly sets per muscle group as the standard recommendation for growth. Triceps were the only muscle where going above 20 sets beat the moderate range, at a standardised mean difference of -0.5 and p = 0.01, while the quadriceps (p = 0.19) and the biceps (p = 0.59) showed no advantage.

That last line is the one that gives the game away. A research team hit the counting problem mid-paper and translated their own answer by hand. Two careful reviews can read the same evidence and print different set totals for the same training week purely because of how each one counts.

Count One Week Fractionally and You Will Probably Find You Are Already Ahead

Most lifters undercount, and they undercount in one direction. They tally the isolation work, ignore everything else, and conclude their arms and shoulders are under-trained. Correcting it takes about ten minutes.

Write out this week's sessions by exercise and sets. Pick one muscle. Mark every set where it drove the lift, then every set where it helped. Direct at one, helper at a half, add them up. Compare the total against whatever target you use, and ask which convention that target was written in. Do the triceps and biceps first, where the gap is widest.

Two rules keep this from becoming a spreadsheet hobby. Pick a convention and stay inside it, because a number counted one way this month and another way next month tells you nothing. And do not treat the higher count as permission to cut isolation work, since the trials above disagree about how much a compound lift covers.

There is a payoff on the weeks that fall apart. When a meeting runs long and the accessories disappear, the pressing and pulling you did still bought something, and the week reads as a smaller loss than it looks. The next question is usually how many sets you actually need, and if the answer is more than your calendar allows, pairing sets back to back buys some of it back.

Whatever does the counting has to keep counting when you are busy, because nobody sustains a manual tally past week three.

About Be Fit and Strong Evidence-based fitness coaching for busy professionals in Dubai. Built by Mirza.

Want your weekly volume counted properly instead of guessed at? See how online coaching works.