5/3/1 for hypertrophy: how to run it in a training log

Most 5/3/1 stalls are set on day one. The training max comes from a lift the lifter remembers rather than a set they logged, and every percentage for the next four weeks inherits the error. Wendler's own fix is to start from 90 per cent of a realistic one-rep max and to err light [1]. A training log that estimates your 1RM from each set gives you that realistic number without a test day, and it tells you each cycle whether the training max is still honest.

This is a guide to running 5/3/1 for size in a log that tracks set roles. It is not a review of whether 5/3/1 beats anything else. It is the method as Wendler wrote it, with the numbers worked out in kilos for one lifter.

What 5/3/1 prescribes

The program is built around four lifts: the parallel squat, bench press, deadlift and standing press, one per session, in a four-week cycle [1]. Every load is a percentage of a training max, and the training max is 90 per cent of the lifter's 1RM, so a 315 lb (143 kg) bench is trained off 285 lb (129 kg) [1]. Wendler restates the rule in later templates as 90 per cent of your best gym lift, never a competition lift [2], and drops it to 85 per cent for the hardgainer template [3].

The weekly table is fixed [1]:

The plus means as many reps as you can manage with that weight, with the goal of a rep record each session. The deload week is the exception, where no records are attempted [1]. After each cycle the upper-body training maxes go up 5 lb (2.3 kg) and the squat and deadlift go up 10 lb (4.5 kg), then every percentage is recalculated [1]. In a kilo gym those steps become 2.5 kg and 5 kg, because the smallest change a pair of 1.25 kg plates allows is 2.5 kg, which the plate calculator will confirm for your gym's plates.

Wendler treats the training max as a tool rather than a trophy. He republished a forum member's explanation of it, with the line that a higher training max does not equal more strength and that the more progress you can get out of a low one the better, and said he could not imagine it being explained more simply [4]. Keep that in mind for the rest of the guide.

Set the training max from a logged set

Take a 75 kg lifter whose last squat top set was 100 kg for five at RPE 8. Run that set through the app's model, the same calculation as the one-rep max calculator, and the estimated 1RM is 123.3 kg. Epley would say 116.7 kg, because it assumes the fifth rep was the last. Ninety per cent of 123.3 kg is 111 kg. Round down to 110 kg, since Wendler's advice when in doubt is too light [5], and every load for the cycle follows.

Those are Wendler's percentages of 110 kg rounded to the nearest 2.5 kg. On a 20 kg bar, 92.5 kg is a 25, a 10 and a 1.25 per side, and 105 kg is a 25, a 15 and a 2.5 per side. Enter the three main sets as normal sets with the third tagged as the top set, and write the training max down where you will see it next cycle, so the recalculation starts from a number and not a memory.

Run the main sets under the Manual scheme. The app never autoregulates Manual, so the loads stay exactly what the training max says, and a good week does not quietly add 2.5 kg the program did not ask for. The six autoregulated schemes belong to lifts where you want the log to pick the next weight. 5/3/1 already has a rule for that.

Log the plus set with RPE and read it, do not act on it

The plus set is where 5/3/1 does its work, and it is the one set that tells you whether the training max is right. Log it with reps and RPE like any other top set, and the model turns it into an estimated 1RM you can compare with the one you started from.

Suppose the lifter's three plus sets go 92.5 kg for eight at RPE 9, 100 kg for five at RPE 9, and 105 kg for three at RPE 9. The model reads them as 121.4 kg, 119.5 kg and 117.7 kg. All three sit within a few kilos of the 123.3 kg the cycle was built from, so the training max was honest. Had the week-one set come out at 135 kg or above, the training max would have been set from an estimate that was too low, and the next cycle's bump would still be too small. Had it come out at 110 kg, the lifter was grinding sets that were meant to be fast.

Wendler's own check is simpler and worth logging against: if you cannot hit five strong reps at 95 per cent, the training max is wrong [3]. The model lets you test that without a separate set. From 105 kg for three at RPE 9, the rep-max ladder puts the lifter's five-rep max at 101.6 kg and makes 105 kg a four-rep load. Five clean reps at 105 kg are not there yet, so this lifter holds the training max for another cycle rather than adding 5 kg.

What you do not do is set next cycle's training max from the plus set's estimate. Wendler's own calculator carries the warning that a strong set means the plan is working and that you should not automatically increase a training max from an estimate [6]. The estimate is a check on the training max. The 2.5 kg and 5 kg bumps are the progression, and they only happen when the reps were clean.

Boring But Big as back-off sets

Boring But Big is Wendler's hypertrophy supplement: after the three main sets, five sets of ten with a lighter weight, at 50 to 60 per cent of the training max, or around 30 per cent for a lifter new to it [7]. You can pair each lift with itself or with its opposite, so squat day finishes with five tens of deadlift and deadlift day with five tens of squat. Wendler calls neither better [7]. In the deload week the five sets of ten stay or drop to three [7].

For the 110 kg training max that is 55 kg to 65 kg: a 15 and a 2.5 per side, or a 20 and a 2.5. Log the five sets with the back-off role. Back-offs count toward weekly sets per muscle, so the squat day puts eight working sets on the quads and glutes, three main and five back-off, and the warm-up ramp adds nothing. If the opposite-lift pairing is used, the quads see five more squat sets on deadlift day, which is where the size in the template comes from.

The effort on those sets matters less than their presence. In an eight-week study where lifters trained three times a week with sets and reps matched, loading by RPE and loading by percentage of 1RM produced the same muscle thickness changes, and most of the training sat three to four reps short of failure [8]. Wendler's own instruction for the back-offs is in the same spirit: do not try to add more to this program, which he calls the biggest mistake lifters make with it [7]. Start the tens light and let the training max carry them up. Each cycle's bump raises the five-by-ten loads with it [7].

Where do the cycle bumps stall?

A 5 kg bump every four weeks is 65 kg a year on the squat training max, and no lifter sustains that. Wendler says so himself: once strength reaches a certain point, progress takes a lot longer than three or four weeks, and the lifter who stalls is usually the one who stayed at too heavy a training max, repeated cycles and ended up run down [9]. His remedy is to take the training max back and stay at the lighter number one cycle longer than you think you need [9].

In the log, the stall shows up before the miss does. Watch the plus set's estimated 1RM across cycles. If the training max has climbed 10 kg over two cycles while the estimate from the plus set has not moved, the training max is now above 90 per cent of what the lifter can do, and the next 95 per cent single will be a grinder. The week-three check, five strong reps at 95 per cent, is the time to decide [3]. Fail it and the next cycle repeats the same training max or takes it back, and the Boring But Big sets keep the volume up while it does.

The companion guide on progressive overload for hypertrophy covers what to do when a lift stalls outside a fixed-percentage program. Inside 5/3/1, the answer is almost always the training max.

Questions

What training max should I use for 5/3/1?

Ninety per cent of a realistic one-rep max, which is Wendler's own rule, and an estimated 1RM from a recent logged set with RPE is the most realistic number most lifters have. Round down to the nearest 2.5 kg. If the week-three set does not give five strong reps at 95 per cent, the training max is too high.

Should I add weight to the training max every cycle?

Only when the reps were clean. The written rule is 5 lb (2.3 kg) for the presses and 10 lb (4.5 kg) for squat and deadlift after each cycle, 2.5 kg and 5 kg in a kilo gym. A strong plus set is the plan working, not a reason to jump the training max to the new estimate.

Can I run 5/3/1 for hypertrophy without Boring But Big?

Yes. Wendler's beginner template uses First Set Last instead: after the three main sets, five sets of five at the first set's weight. Log them as back-off sets so they count toward weekly sets per muscle. It is less volume than five tens, so it suits a lifter who is also squatting and pressing on the same day.