Double progression: when should the weight go up?

Double progression means adding reps at a fixed weight until you reach the top of a rep range on every working set, then adding weight and starting the range again. Povver, a strength training app, runs it with one extra rule: the weight goes up once the last set reaches the top of the range at the effort you planned, and it stays put while the lift is sliding.

What is double progression?

Say you bench for three sets in a range of 8 to 12 reps. At 60 kg / 132 lb you log 8, 8 and 8. Next session you add reps where you can, and keep going until all three sets reach 12. Only then does the weight go up, to 62.5 kg / 138 lb. The reps usually drop back after the jump, and the climb starts over.

It is progressive overload in two steps: reps first, load second. Progression models compared sets it beside linear, percentage-based and RPE loading. The top of the range is the cue to add weight.

How much weight do you add, and after how many sessions?

Add the smallest step you can load. The classic version makes you hit the top of the range on every set, often for two sessions, before the weight moves. Povver can move it after a single session, on an effort check explained below.

Povver rounds to the step your log already shows. With no step in the log, it uses, in kilograms, 2.5 kg (1.25 kg at 40 kg and under), and in pounds, 5 lb. The more reps you had left, the bigger the jump, up to 10% of the load (5% on isolation lifts). On a light lift where a single step is too big, the reps keep climbing past the top of the range instead.

Does adding reps work as well as adding weight?

In an eight-week study (Plotkin 2022), 43 lifters, all with at least a year of lower-body training, were split into two groups. One group added load at fixed reps. The other kept the load and added reps. Muscle growth came out about even: most sites differed by under a millimetre, and the front-thigh muscle (rectus femoris) grew slightly more in the reps group. Smith machine squat 1RM rose about 2 kg / 4.4 lb more with load, a gap the authors called "of questionable practical significance".

A 2023 review pooled 15 studies on how close to failure a set has to go (Refalo 2023). Going to failure gave a trivial edge for growth, and when failure meant strictly the set where no more reps would come, the edge was not significant.

In that study adding reps worked about as well as adding load, and in the pooled data stopping short of failure cost little growth.

The reviews behind that, and how close to go, are in Training to failure.

RIR (reps in reserve) is how many more reps you had left at the end of a set, and RIR 0 is failure (Helms and colleagues, 2016). RPE vs RIR covers how well lifters judge it and how it maps to RPE.

Which rep range should you use?

A pooled analysis of 21 studies (Schoenfeld 2017) compared light loads (60% of 1RM or less) with heavier ones, every set taken to failure. Muscle growth was similar across the loads studied, and heavier loads gained more 1RM strength. Plans commonly use 5 to 8 reps on heavy compound lifts and 10 to 15 on isolation lifts. That is convention, not a finding. Every set in those studies went to failure, which is not how most plans are written, so Training to failure covers how close to go. How many reps build muscle has the full rep-range evidence.

How Povver runs double progression

Each exercise in your plan has a rep range, and each set has a target RIR. When a set hits its rep target at that effort or easier, next time's target goes up a rep. With two or more reps to spare beyond the target RIR, it goes up by that many, to a maximum of three.

The weight goes up after one session where every working set is finished and your last set would have reached the top of the range at your target RIR. With a range of 8 to 12 and a target of RIR 1, 11 reps at RIR 2 counts as 12. Log no RIR and it takes two sessions in a row at the top of the range.

Here the incline dumbbell press has a range that tops out at 10. On 12 September every set went 10 reps at RIR 1, the top of the range at the target effort. The next session was at 32 kg / 71 lb: 10 reps on the first set, 9 on the other two. That dip after a jump is normal, and the climb starts again from there.

The size of the jump depends on the reps you had left. It is at least one step of the weights you use, and otherwise no more than 10% of the load (5% on isolation lifts). Light isolation lifts are where a step can be too big: 2.5 kg / 5.5 lb on a 9 kg / 20 lb curl is 28%. There the reps keep climbing past the top of the range, up to 30, until the heavier weight still leaves you inside the range.

RIR sits in the grid next to weight and reps, pre-filled from your plan or your last session. Change it to match how the set felt.

A lift whose muscle is below the weekly sets it needs to grow (its minimum effective volume) still progresses as normal. The muscle just needs more sets.

When the weight stays

The weight stays put, or comes down, in these cases:

• A set harder than planned. A working set more than one rep under its target RIR keeps the weight where it is. If it happens again next session at that weight, the weight drops a step.
• Missed reps. A working set that ends two or more reps below the bottom of the range drops the next session a step. Reps are judged against your target RIR (an RIR above 4 counts as 4): spare reps beyond the target count as made, and each rep of RIR you went below it counts as one missed. One rep short only keeps the weight where it is. A set that reached the bottom of the range is never a miss, and if it was harder than planned, the rule above covers it.
• Light dumbbells. When the next dumbbell down is too big a drop for your rep range, neither rule above lowers the weight. It stays, and the rep targets come down to the reps you made.
• A slide. The lift's estimated 1RM (e1RM, from the best set each week) has trended down 2% or more over six weeks and has not levelled off.
• Fatigue. Lately you have trained the muscle the lift works most at more than 1.5 times its longer-term average load.
• Two missed changes in four weeks. A change counts as missed when you later lift under the new weight, or when half your sets or more fall short.
• A deload week. No increases that week.

The bent-over row here is stalling. Its estimated 1RM is down 14% in eight weeks, from 95 kg / 209 lb to 81.7 kg / 180 lb. The last three sessions were all at 70 kg / 154 lb, and the latest went 5, 5 and 4 reps with two sets at RIR 0. While a lift slides like this, the weight does not go up, whatever the reps.

A lift whose estimated 1RM has sat within about 4% for three weeks or more is marked maintaining (a plateau), even if it fell before that.

In Povver, a lift marked maintaining (a plateau) still gets its increase once it tops its range.

Come back to a lift after three weeks or more away and the first session starts 10% lighter (20% after 16 weeks). Two good sessions later you are back at the old weight. A break like that is not read as a stall.

Who decides

Every check above is just arithmetic on the sets you have logged. The coach in the app proposes the new weight, and a suggested change comes with the numbers behind it on screen. With auto-pilot on, the engine caps the increase, rounds it to a weight you can load and writes it into your plan, which then shows the before and after weights. Rounding uses your usual step when the log shows one, and otherwise, in kilograms, 2.5 kg (1.25 kg at 40 kg and under), and in pounds, 5 lb.

With auto-pilot off, the proposal waits for you.

Double progression: common questions

What is double progression?

Double progression adds overload in two steps. Reps go up at a fixed weight until your last set reaches the top of a rep range, then the weight goes up and the range starts again. You can use it on any lift with a rep range, such as 8 to 12 reps.

How does double progression work?

Pick a rep range and a weight you can lift for the bottom of it. Add reps each session until the last set reaches the top of the range, then raise the weight and let the reps dip. Povver, a strength training app, adds a check on effort: the last set has to reach the top of the range at the reps in reserve your plan targets.

When should you increase the weight?

When every working set has reached the top of the rep range, often for two sessions in a row. In Povver, one session is enough when every working set is finished and the last set reaches the top at your target effort, counting reps you had spare beyond the target. With no RIR logged it takes two sessions in a row.

Does adding reps build as much muscle as adding weight?

In one 8-week study of 43 trained lifters, it did about as well. Plotkin and colleagues (PeerJ, 2022) found muscle growth about even between a group that added reps and a group that added load, with most sites differing by under a millimetre.

How much should you increase the weight?

By the smallest step you can load. Povver rounds to the step your log already shows, or else, in kilograms, to 2.5 kg (1.25 kg at 40 kg and under), and in pounds, to 5 lb. It sizes the jump from the reps you had left, up to 10% of the load (5% on isolation lifts).

What rep range should you use?

A 2017 meta-analysis of 21 studies (Schoenfeld) found muscle growth similar between light and heavier loads taken to failure, and 1RM strength gains greater with heavier loads. Plans commonly use 5 to 8 reps on heavy compound lifts and 10 to 15 on isolation lifts, which is convention.

What if you miss the bottom of the range?

In Povver, a working set that ends two or more reps below the bottom of the range drops the next session a step. One rep short keeps the weight where it is. A set that was harder than planned also keeps the weight, and a second one in a row at that weight drops it a step.