Double progression: reps first, then weight

Double progression means you raise two numbers in turn. First reps: keep the weight and add one rep each session until you reach the top of a range, say 6–8. Then weight: add the smallest step your plates allow and start again at the bottom of the range. The Plate Calculator shows the smallest step your plates can make. The Next-Set Calculator runs the rule on your last session.

The rule in one table

Four cases cover everything that can happen. The rule reads one set — your best working set from the last session on that exercise — and ignores warm-up sets. Below the top of the range, you add a rep. At the top with something left, you add weight. At the top with nothing left, you repeat the session.

Masheen is a workout log for people who lift, and it is coming soon to iPhone and Android. Its coach uses double progression, exactly as in the table above, and shows the reason for each suggestion in plain words. You can change any suggestion. The coach is part of Masheen Pro. You can see it at work in the coach on the Masheen home page: "Last time you hit 8 reps at 100 kg — top of your 6–8 range with reps to spare. Add 2.5 kg and reset to 6 reps."

A worked example: six sessions on the bench press

Here are six sessions on the bench press with a 6–8 rep range and a 2.5 kg step. The weight stays at 100 kg for four sessions while the reps go up, then moves once. Each line shows the best set, the reps left over, and what the rule asks for next time.

Session 3 is the one to look at. You reach 8 reps, the top of the range, but you have nothing left, so the weight holds. Session 4 repeats that weight and those reps, and this time two reps are left over. That earns the step: 102.5 kg, back to 6 reps.

Why reps go first

Because a rep is a smaller step than a plate, at least on a light lift. Going from 8 reps to 9 at the same weight raises your estimated one-rep max by about 2.6%, whatever the weight. A 2.5 kg jump is 6.3% of a 40 kg press, more than twice as big. The lighter the lift, the more the reps have to carry.

The arithmetic uses the Epley formula: estimated max (the most you could lift once) = weight × (1 + reps ÷ 30). At 8 reps that is weight × 1.267, at 9 reps weight × 1.300, a rise of about 2.6% at any weight. At 40 kg, 8 reps gives 50.7 kg and 9 reps gives 52.0 kg. It is an estimate: how accurate one-rep max formulas are covers the limits.

The plate side has no such ratio. The smallest jump a barbell can make is twice your lightest plate, because one goes on each side: 1.25 kg plates step 2.5 kg, 0.5 kg plates step 1 kg. As a share of the bar, the same step shrinks as the bar gets heavier.

Read the 40 kg row: the plate step is 6.3% against a rep step of 2.6%. At 100 kg the two are about the same size, 2.5% against 2.6%, and the order matters less. For scale, the American College of Sports Medicine's 2009 position stand on progression recommends a 2–10% increase in load once you can do one to two repetitions over the target number on two sessions in a row, with the lower end for small-muscle exercises and the higher end for large ones. Reps are what carries progressive overload between those jumps.

Picking the range and the step

Two choices set the pace: how wide the rep range is, and how small the weight step is. Masheen uses 8–12 reps when a routine sets none. The example above uses 6–8. A narrow range reaches the next weight sooner; a wide range gives more sessions between jumps. The step is twice your lightest plate.

Count the rep steps. A 6–8 range has two, 6 to 7 and 7 to 8. An 8–12 range has four. So the same lift on the wider range has twice as many rep steps to climb at each weight before it moves.

Masheen's default step is 2.5 kg for upper-body lifts and 5 kg for lower-body lifts. You can pick a different step for any exercise, from 0.5 kg or 1 lb up. For barbell lifts it rounds the suggestion to a weight you can build from the bar and plates you have set under Plates and bar. The bar counts in that total as well as the plates: how much each bar weighs lists the usual ones, and the Plate Calculator loads a bar from the plates you have.

"A rep to spare": how sure can you be?

Reps in reserve is what you log after a set: how many more reps you could have done. The rule asks for one, not three. A 2022 review found lifters under-predicted their reps left by 0.95 of a rep on average: they thought they were nearer the end than they were.

That figure comes from Halperin and colleagues in Sports Medicine, pooling 12 studies and 414 participants. Predictions were a little better closer to failure and in sets of 12 reps or fewer. Training experience made no clear difference.

When the guess is wrong, the cost is one session at the same weight. Session 3 in the table above is that cost, and it is small.

Does it work?

Two studies have compared adding reps with adding load. Neither tested double progression as such — each ran one method against the other for the whole study — and both found similar results from the two. That is the honest answer: adding reps and adding weight gave similar results, within the lifts and the weeks those studies covered.

Plotkin and colleagues, in a 2022 study in PeerJ, put 43 trained lifters on eight weeks of lower-body training, twice a week. One group added load and kept reps the same; the other added reps and kept load the same. Muscle growth and strength gains were similar, with small differences either way. The authors concluded that both "appear to be viable strategies" over eight weeks.

Chaves and colleagues, in a 2024 study in the International Journal of Sports Medicine, trained 39 previously untrained young adults for ten weeks, one leg progressing by load and the other by reps on the leg extension. Strength and muscle size rose on both legs, with no difference between them.

Keep the limits in view: trained lifters, lower body, eight weeks in the first; beginners, one machine exercise, ten weeks in the second. Neither study tested double progression, and neither says it beats anything else.

When it stops working

When a lift stops answering the rule, the reps stop going up and the weight stays put. Masheen's coach watches for that. When your estimated one-rep max on an exercise has not improved across your last three sessions, it marks the lift as stalled and suggests dropping about 10% next time and building back up, or swapping in a close variation for a few weeks.

From the lighter weight, the same rule builds back up. A cut like that is part of progressive overload, not a break from it.