Plotkin and colleagues gave 43 lifters with at least a year of lower body training eight weeks of squats, leg extensions and calf raises, twice a week. Half of them added load and kept their reps fixed. The other half kept the load fixed and added reps. Rectus femoris growth leaned slightly toward the rep group and Smith machine squat strength leaned toward the load group by 2.0 kg, and the 90 percent confidence intervals for both differences crossed zero. Every other muscle thickness difference was under a millimetre (Plotkin, 2022). Both ways of adding stress worked.
That result settles less than it seems to. Eight weeks in lifters with a year of experience says nothing about month nine, when the easy gains are spent and the log shows the same bench press for the fourth week running. At that point the question is which way of adding stress fits the rate at which you still adapt.
This article sets the four common progression schemes side by side and gives a rule for choosing among them by training age. It also gives one deload rule with its sources, a 12 week table built on both, and what to log so each next step comes from your own numbers. The glossary entry on progressive overload covers the definition.
01Why lifts stall before the program runs out
Take a lifter who benches 80 kg for 3 sets of 8 in January and adds 2.5 kg every week. By the end of February the bar reads 95 kg, the third set stops at 6 reps, and the next week it stops at 5. She takes 10 kg off and climbs again. Six weeks later she misses at 95 kg a second time.
Her effort was fine and so was her program, for the first eight weeks. What changed is the ratio between the size of her step and the rate at which she adapts. A 2.5 kg weekly step on an 80 kg bench is a 3 percent increase every seven days. Early on her strength rose faster than that. By March it rose more slowly, so each week asked for more than the last week had built.
Every progression scheme answers the same two questions, how big is the step and how often do you take it. Linear progression takes a large step every session. Double progression takes a small step in reps every week and a load step only when the reps allow it. RPE based loading lets the day's performance set the load. Wave loading spreads the step across a sequence of heavy sets. The right scheme is the one whose step matches the rate you actually adapt at now, and that rate falls with training age.
02Four ways to add stress, and how to pick one
The table sorts the schemes by the signal that tells you which one you need. Training age in months is a weak guide because people adapt at different rates. The signal column is the rule.
| Scheme | Signal you need it | Unit of progress | Typical step | Signal to move on |
|---|---|---|---|---|
| Linear | Lifts go up every session, usually the first 6 to 9 months of consistent training | Load, every session | 2.5 to 5 kg lower body, 1 to 2.5 kg upper body | Missing target reps at the same load after one reset |
| Double progression | Linear progression stalls twice at the same load | Reps first, then load | One rep per set, then the smallest load increment | Reaching the top of the rep range on every set takes more than 4 to 6 weeks |
| RPE and RIR | Performance at a fixed load varies day to day, and you can rate reps in reserve within about one rep | Load chosen daily against an effort target | Top set load rises when it rates below target | Top sets flat for a full block at the same RPE |
| Wave loading | Heavy lifts improve over months, and a single session cannot hold enough heavy work | Load across a sequence of heavy sets | 2.5 kg per set from wave to wave | Second wave fails to beat the first on consecutive sessions |
Linear progression while form and recovery allow big jumps
Linear progression adds load every session at fixed sets and reps. A novice squatting three days a week might start at 60 kg for 3 sets of 5 and add 2.5 kg each session, which is 7.5 kg a week. Much of a novice's early strength comes from learning the movement, so a step that size holds for a few months.
The first warning is missing the target reps at a load on two sessions in a row. The usual fix is to take 10 percent off and climb back once. The reset buys a few weeks because the first sessions after it are easy. If you stall again at the same load, the step is now bigger than your adaptation rate and resetting a third time only repeats the loop. Move to double progression.
Double progression to bank reps before adding load
Double progression fixes a rep range and a load. You add reps across sessions until every set reaches the top of the range at your effort target, then add the smallest available load and drop back to the bottom of the range. A dumbbell bench press over five weeks, in a 3 × 8 to 12 range at about 2 reps in reserve, looks like this.
| Week | Load per hand | Reps per set | Total reps |
|---|---|---|---|
| 1 | 26 kg | 10, 9, 8 | 27 |
| 2 | 26 kg | 11, 10, 9 | 30 |
| 3 | 26 kg | 12, 11, 10 | 33 |
| 4 | 26 kg | 12, 12, 12 | 36 |
| 5 | 28 kg | 9, 8, 8 | 25 |
The reps carry the progress between load jumps. That matters most where the smallest increment is large relative to the load. Going from 26 kg to 28 kg dumbbells is a 7.7 percent jump, which no intermediate lifter can take every week. Four weeks of added reps bridges it. The Plotkin trial is the direct evidence that adding reps at a fixed load is a valid form of overload in trained lifters, with growth and strength outcomes close to load progression over eight weeks (Plotkin, 2022).
Double progression suits accessories and machine work for as long as you train. On the main barbell lifts it usually holds for a year or two, until performance at a fixed load starts to swing enough from day to day that a planned rep target is wrong half the time.
RPE and RIR based progression once performance varies day to day
The resistance training RPE scale ties each rating to reps in reserve. Zourdos and colleagues defined it so that RPE 10 means no reps left, RPE 9 means one rep left, RPE 8 two, and so on down (Zourdos, 2016). The RPE glossary entry covers the older Borg scale it replaced.
| RPE | Reps in reserve | What the last rep looks like |
|---|---|---|
| 10 | 0 | Maximal, another rep would fail |
| 9 | 1 | Slow, one more rep was possible |
| 8 | 2 | Clearly slowing, two more reps were possible |
| 7 | 3 | Bar speed still fast, several reps left |
| 6 | 4 or more | Warm-up speed |
The scale's accuracy depends on experience. In the same Zourdos study, 15 experienced squatters with an average training age of 5.2 years rated their true one rep max at 9.80 on average. Fourteen novices averaging 0.4 years rated theirs at 8.96, which means they believed a rep remained when none did. Helms and colleagues drew the practical conclusion that novices should record RIR alongside their normal training data and should not base load or progression on it until their ratings become accurate (Helms, 2016).
For lifters past that stage, RPE loading performs at least as well as percentages. Helms and colleagues ran 21 trained men with at least two years of experience through eight weeks of the same periodized program, three sessions a week. One group lifted fixed percentages of their pre-test max, and the other chose loads to hit RPE targets. Squat 1RM rose 17.05 kg in the RPE group and 13.91 kg in the percentage group, and bench press rose 10.70 kg against 9.64 kg. Neither difference was statistically significant, and muscle thickness gains were similar (Helms, 2018). A meta-analysis of six studies comparing autoregulated and fixed load prescription found a 2.07 kg difference in 1RM strength favouring autoregulation, with a confidence interval that crossed zero (Hickmott, 2022). The case for RPE rests on what it does on the days when the plan is wrong.
A worked example. The plan calls for a squat top set of 5 at RPE 8, then back-off sets at 90 percent of whatever the top set turns out to be. Last week that top set was 145 kg. Today, after two short nights, 135 kg for 5 already rates RPE 8, so 135 kg is today's top set and the back-off sets go at 121.5 kg, rounded to 120 kg. A percentage program would have put 145 kg on the bar and she would have ground out a set at RPE 9.5 or missed it. On a better day, if 145 kg for 5 rated RPE 7, she would add 2.5 to 5 kg the following week. The load moves with the day and the effort stays where the program intended.
Wave loading for lifters close to their ceiling
Wave loading runs heavy sets in repeating sequences, most often triple, double, single, with the load rising inside each wave and again from one wave to the next. For a bench press with a 150 kg max, the first wave is 127.5 kg for 3, 135 kg for 2 and 142.5 kg for 1, which is 85, 90 and 95 percent. The second wave adds 2.5 kg to each set, 130, 137.5 and 145 kg. If the second wave beats the first, the next session starts from it.
The structure packs more work above 85 percent into one session than a single top set can, and it lets an advanced lifter progress in 2.5 kg steps between waves when a weekly step on the top single would be too large. Its support comes from coaching practice and the idea that a heavy single primes the next set. None of the trials cited here tests wave loading against another scheme, so treat it as a tool for a specific problem.
Beginners should skip it. The work sits at 85 to 95 percent with very few total reps, where a novice gains more from volume at moderate loads. It also depends on rating near-maximal sets, the exact range where Zourdos found novices misjudge reps in reserve by close to a full rep.
03Autoregulation in plain terms, matching the plan to the day
Autoregulation means adjusting load or volume to the day's actual performance. RPE targets do this inside the session, after your warm-ups tell you how the bar is moving. A readiness signal from the night before tells you earlier, before you pick the first plate.
Titan's Recovery score is that earlier signal. It compares last night's HRV and sleeping heart rate with the median of your measured nights in the Recovery window, 60 nights by default, measured in units of your median absolute deviation, and reports a score from 0 to 100. Low is 0 to 39, Moderate is 40 to 74 and High is 75 to 100. The app describes the Low band as a day to favor easy work, sleep and nutrition. Personal baselines explains how the window is built.
For a lifter, the rule of thumb outside the app is this. A morning HRV more than one median absolute deviation below your baseline median for two or more consecutive days, with resting heart rate above baseline, is a signal to reduce training intensity. Under linear or double progression, that means repeating last week's loads instead of taking the next step. Under RPE loading, it means keeping the top set at its target RPE and accepting the lighter load it produces. On a wave day, it means stopping after the first wave. Titan does not change your program. It reports the numbers, and the decision stays with you. The HRV training readiness guide covers the rule in more depth.
04Deloads, how often and how deep
A deload is a short planned drop in training stress that lets fatigue clear before the next block. Two studies from the same research group describe what experienced strength athletes and coaches actually do.
Rogerson and colleagues surveyed 246 competitive strength and physique athletes with an average of 8.2 years of resistance training. They deloaded every 5.6 ± 2.3 weeks, for 6.4 ± 1.7 days. During a deload, 78.9 percent cut weekly sets, 83.7 percent lowered the load on multi-joint lifts, 84.9 percent trained further from failure, and 63.0 percent kept the same number of sessions (Rogerson, 2024). Bell and colleagues interviewed 18 national level coaches in weightlifting, powerlifting and bodybuilding. They typically deloaded for 5 to 7 days every 4 to 6 weeks. They cut volume by roughly 30 to 50 percent through fewer sets or fewer reps, and reduced effort by adding one to three reps in reserve or by lowering load about 10 percent at the same reps. Coaches scheduled deloads in advance for novice and intermediate athletes and used reactive deloads more with advanced ones (Bell, 2022).
This article uses those reported practices as a starting point, not a proven optimal interval. After three loading weeks, take a deload week of five to seven days. Keep the same sessions and exercises, halve the working sets, keep the reps, and use 90 percent of the previous week's loads. The four week cycle is a practical template; the coach interviews and athlete survey did not test which interval works best. Halving the sets sits at the top of the coaches' 30 to 50 percent volume cut, and the 10 percent load drop is the one they described. The adaptive sports coaching guide covers when to bring a deload forward.
If a deload leaves your next heavy session no better than the one before it, check sleep and life stress before you extend the deload. A week of halved volume clears ordinary training fatigue. It does not fix five short nights.
05The twelve week progression table
This table is written for an intermediate lifter training three or four days a week, with two or three main barbell lifts such as the squat, bench press, deadlift or overhead press. It uses the deload rule above, so deloads fall in weeks 4, 8 and 12. The first block uses double progression. The second and third use RPE top sets with back-off sets. Every loading week has four working sets per main lift and every deload week has two.
| Week | Block | Main lift work per session | Effort target | Progression rule |
|---|---|---|---|---|
| 1 | Accumulation | 4 × 6 | RPE 7 on the first set | Choose the load that rates RPE 7 for 6 reps. It stays fixed through week 3. |
| 2 | Accumulation | 4 × 6 to 8, week 1 load | RPE 7 to 8 | Add a rep to any set that allows it at RPE 8 or below |
| 3 | Accumulation | 4 × 6 to 8, week 1 load | RPE 8 | Keep adding reps. Stop at 8. |
| 4 | Deload | 2 × week 3 reps at 90 percent of the week 3 load | RPE 6 or lower | Half the sets, same reps, load down 10 percent |
| 5 | Intensification | 1 × 5 top set, then 3 × 5 at 90 percent of the top set | Top set RPE 8 | Pick the top set load from warm-ups. Back-off sets follow it. |
| 6 | Intensification | 1 × 4 top set, then 3 × 4 at 90 percent | Top set RPE 8 | If last week's top set rated 7, go 2.5 to 5 kg heavier. If it rated 9, stay put. |
| 7 | Intensification | 1 × 3 top set, then 3 × 3 at 90 percent | Top set RPE 8 | Same as week 6 |
| 8 | Deload | 2 × 3 at 90 percent of the week 7 back-off load | RPE 6 or lower | Half the sets, same reps, load down 10 percent |
| 9 | Realization | 1 × 3 top set, then 3 × 3 at 90 percent | Top set RPE 8.5 | The top set load should exceed the week 7 top set load |
| 10 | Realization | 1 × 2 top set, then 3 × 2 at 90 percent | Top set RPE 9 | Same as week 6 |
| 11 | Realization | 1 × 1 top set, then 3 × 2 at 90 percent | Top set RPE 9 | This single is your best estimate of current strength |
| 12 | Deload | 2 × 2 at 90 percent of the week 11 back-off load | RPE 6 or lower | Half the sets, same reps, load down 10 percent. Then start the next cycle at week 1. |
Accessories run double progression in a 4 × 10 to 15 range at one to two reps in reserve for all twelve weeks and follow the same deload rule, so they drop to two sets at 90 percent of the previous week's load in weeks 4, 8 and 12.
The blocks trade volume for load. Main lift reps per session fall from 32 in week 3 to 20 in week 5 and 7 in week 11, while the loads rise. Weeks 9 to 11 use RPE 8.5 and 9 on the top set only, which keeps the heaviest work to one set per lift per session. The week 11 single replaces a formal max test. An RPE 9 single sits close enough to a true max to show whether the cycle worked, without the fatigue of a failed attempt.
The next cycle starts again at week 1, with the load chosen by the same RPE 7 rule. That load should be heavier than the week 1 load of the cycle before. If it is not, the cycle did not produce progress and the next section is where to look.
An advanced lifter can replace the realization block with wave loading. Weeks 9 to 11 become two waves of 3, 2 and 1 per session, with the second wave 2.5 kg heavier per set. Week 12 stays a deload. For how the same structure applies to a hypertrophy goal, with set targets per muscle group, see the muscle building guide.
06Diagnosing a stall before touching the program
A stall is three weeks or more without progress at the same effort. It has three common causes, and each needs a different fix. Changing the program when the cause was fatigue, or deloading when the cause was too little training volume, costs a month.
Technical stalls. One lift stalls while the others progress, and the missed rep fails at the same point each time. The bar drifts forward out of the hole, or the elbows flare at the same height on the press. More load will not help. Drop 10 percent, add paused reps or tempo work at the sticking point for two or three weeks, and film a set each session.
Fatigue stalls. Several lifts stall at once and the same loads rate a full RPE point higher than they did two weeks ago. The HRV readings meet the rule above, or Recovery has spent most of the week in the Low band. Sleep is often short. Take the deload now rather than at the scheduled week, then restart the block. Pushing through this pattern for several more weeks is how functional fatigue turns into overreaching that takes longer to clear.
Programming stalls. Loads feel fine at the target RPE, recovery signals are stable, and the numbers still do not move. Either the step is too big for the scheme, or the volume is too low. The first calls for moving one row down the scheme table. For the second, Schoenfeld and colleagues pooled 15 studies and found a graded dose response, with each extra weekly set per muscle associated with about 0.37 percent more growth (Schoenfeld, 2017). Adding two sets per muscle per week is a reasonable first step.
Pushing every set to failure is rarely the fix for a programming stall. A meta-analysis of 15 studies by Refalo and colleagues found a trivial advantage for training to set failure over stopping short, with an effect size of 0.19, and no advantage at all when failure was defined strictly as momentary muscular failure (Refalo, 2023). On heavy compound lifts, failure adds a lot of fatigue for that small return.
07Logging sets so progress is visible instead of assumed
Every scheme above needs the same four things recorded for each set, the exercise, the load, the reps completed and the date. RPE loading and wave loading also need reps in reserve for each working set. Without the reps in reserve column you cannot tell whether the same load and reps required the same effort or more effort. The training log glossary entry covers formats.
A log also turns training volume into a number you can follow. Volume load for a set is reps multiplied by load, so 4 sets of 8 at 100 kg is 3,200 kg. Two patterns in the 12 week table show up clearly in that number. Each deload week lands at roughly half the week before, because half the sets at 90 percent of the load gives about 45 percent of the volume. Volume also falls from block 1 to block 3 even when every top set is a personal best, because reps drop faster than loads rise. Expect the volume line to fall in weeks 5 to 11, and judge those weeks by the top sets.
Keep a separate note of the top set of each main lift for each week, with its RPE. That single column is the most direct record of strength across a cycle, and it is the one that tells you whether the next cycle's week 1 should be heavier.
08Record each session on Apple Watch
Titan does not log sets, reps or weight, so the per-set record and the top set note described above live in your own log. Record each lifting session as a strength workout on Apple Watch as well, as strength workouts in Titan explains. Titan reads the workout from Apple Health, lists it on the Workouts screen and scores it from the heart rate recorded during it, so it adds to Exertion and training load. A strength workout without heart rate counts at 0.9 load per minute when no other workout that day has heart rate.
Heart rate cannot measure muscular fatigue, so Titan's load follows session time and heart rate, not the volume load in the 12 week table. Judge progression from your log, and use Titan's Recovery score and training load for the systemic picture.
09References
- Bell L et al. (2022). "You can't shoot another bullet until you've reloaded the gun," coaches' perceptions, practices and experiences of deloading in strength and physique sports. Frontiers in Sports and Active Living. https://doi.org/10.3389/fspor.2022.1073223
- Helms ER et al. (2016). Application of the repetitions in reserve-based rating of perceived exertion scale for resistance training. Strength and Conditioning Journal. https://doi.org/10.1519/SSC.0000000000000218
- Helms ER et al. (2018). RPE vs. percentage 1RM loading in periodized programs matched for sets and repetitions. Frontiers in Physiology. https://doi.org/10.3389/fphys.2018.00247
- Hickmott LM et al. (2022). The effect of load and volume autoregulation on muscular strength and hypertrophy, a systematic review and meta-analysis. Sports Medicine Open. https://doi.org/10.1186/s40798-021-00404-9
- Plotkin D et al. (2022). Progressive overload without progressing load? The effects of load or repetition progression on muscular adaptations. PeerJ. https://doi.org/10.7717/peerj.14142
- Refalo MC et al. (2023). Influence of resistance training proximity-to-failure on skeletal muscle hypertrophy, a systematic review with meta-analysis. Sports Medicine. https://doi.org/10.1007/s40279-022-01784-y
- Rogerson D et al. (2024). Deloading practices in strength and physique sports, a cross-sectional survey. Sports Medicine Open. https://doi.org/10.1186/s40798-024-00691-y
- Schoenfeld BJ et al. (2017). Dose-response relationship between weekly resistance training volume and increases in muscle mass, a systematic review and meta-analysis. Journal of Sports Sciences. https://doi.org/10.1080/02640414.2016.1210197
- Zourdos MC et al. (2016). Novel resistance training-specific rating of perceived exertion scale measuring repetitions in reserve. Journal of Strength and Conditioning Research. https://doi.org/10.1519/JSC.0000000000001049
