How long should a training block last before you change it?

A training block is a run of weeks with one plan for effort and volume, closed by a deload. So how long should a training block last? Four to six weeks for most lifters. A review of periodization describes this middle cycle as lasting about a month [1], and 246 competitive strength and physique athletes reported a deload every 5.6 weeks on average [2]. In the app, any working block of four weeks or more ends on a deload week, so a five-week block is four hard weeks and one easy one.

The length is the easy part. What you change at the end depends on what the last hard week shows, and that is a number you can read off a logged set.

What a longer block buys you, in kilograms

A longer block does not end harder. It takes smaller steps to the same place. The app's hypertrophy block starts every lift at 3 reps in reserve (RIR, the reps you could still have done) and finishes at 1, whatever the length. Six weeks only spreads that ramp over five working weeks instead of three.

Take a 90 kg lifter who benches 80 kg for 8 at RPE 7. The model behind the one-rep max calculator reads that set as an estimated 1RM of 113.2 kg. Here is the load for 8 reps each week, from the same model, then rounded to a 2.5 kg jump. The model's unrounded figure is in brackets.

Both blocks reach 87.5 kg for 8 in their last hard week. The six-week version gets there with two extra weeks of practice at loads you can handle, and weeks 2 and 3 land on the same bar. With 2.5 kg plates, a ramp finer than that cannot show up on the bar at all, so a longer block mostly adds sets and reps at the same weight. That suits a lifter who recovers slowly or wants more volume before the hard weeks. It wastes time for a lifter who was ready for 87.5 kg in week three.

Change the block when the hard week says so

Read the last hard set of the block before you plan the next one. The same 87.5 kg for 8 means three different things depending on the RPE you logged.

Logged at RPE 8, that set is an estimated 118.4 kg, well above the 113.2 kg you started from. The block worked. Run the same block again, starting from the new number, and change nothing except the loads.

If the set was RPE 9, the target for that week, the estimate is 114.8 kg. That is progress, but small. Keep the block type and consider one change for the next run: a set or two more per week, or a different rep range for the main lift.

RPE 10 is the bad case: 111.3 kg, below where the block began. You finished weaker than you started, which usually means fatigue built faster than the plan expected. Take the deload, then shorten the next block by a week or change its focus, for example from a high-volume hypertrophy block to a strength block that holds volume steady.

Two signals override the calendar. Three sessions without progress on a lift flags a stall by default, and the guide to stalled lifts covers what to do then. Joints that ache at loads that used to feel fine are the other, and that is the moment to bring the deload forward. The guide on how often to deload sets out those signs in full.

What you should not change mid-block is the exercise list. A block compares a lift to itself across weeks. Swap a squat for a front squat in week three and the comparison starts over. If a lift bothers you, swap it at the start of the next block.

What the research can and cannot say

No trial has compared four-week blocks against eight-week blocks in trained lifters. The evidence is indirect. A meta-analysis of 18 studies found that periodized training produced greater 1RM gains than non-periodized training, and longer studies showed larger effects [3]. Most of the studies behind that work lasted under 16 weeks, which may undersell what a planned year of blocks does [1].

Programs that work in trials also change things inside the block. In one eight-week study of trained men, the rep targets dropped every two weeks while load climbed, and the last week was a taper [4]. Both the RPE group and the percentage group gained strength and muscle. The block held the plan steady while the targets inside it moved.

So the research supports planning in blocks and is quiet on their exact length. Four to six weeks is a convention that fits what competitive lifters already do, held loosely and checked against your log. If you want those weeks written into your plan for you, that is what Top Sets does.

Questions

Is a 3-week training block too short?

Not on its own. In the app, a three-week hypertrophy block runs RIR 3, 2 and 1 with no deload, so it suits a short phase that leads straight into a block that has one. Run several three-week blocks back to back and you never get an easy week. If you train hard every week, pair a short block with a longer one, or add a deload week yourself.

Can a training block last 8 weeks or longer?

Yes, if you keep progressing. A longer block spreads the same effort ramp over more weeks, so each week is a smaller step. Watch for the stall flag and for effort creeping up at a held load. In the app, a block of eight weeks or more gets a note to rotate some accessories every four to six weeks, because joints and small lifts tend to tire of the same movement first.

Should I stop training completely between blocks?

Usually not. Lifters in the 2024 survey kept training through their deloads, with fewer sets, lighter loads and more reps in reserve [2]. In one trial of trained lifters, a full week off in the middle of a nine-week program cost some lower-body strength compared with training straight through, and made no difference to muscle size [5]. A lighter week keeps the habit and the movement patterns.

Do I need to change exercises every block?

No. A new block can keep every exercise and change only the effort ramp, the set count or the rep range. Keeping your main lifts lets you compare the estimated 1RM at the end of one block with the start of the next. Change an exercise when it hurts, when it has stalled through a full block, or when your goal has moved. Change it at a block boundary, not halfway through.