TL;DR
Autoregulation adjusts today’s load based on the lifter’s daily readiness rather than a pre-fixed percentage of 1RM. Three working approaches: RPE-capped sets (most flexible, our default), velocity-based training (most precise, requires equipment), and daily-max-driven percentages (simplest mental model). For most online coaching clients, RPE-capped is the right call.
Why autoregulate
A pre-fixed percentage program assumes yesterday’s max equals today’s max. They don’t. Sleep, nutrition, life stress, accumulated fatigue from training move the lifter’s daily ceiling 3-7%. A program written at 80% of an old 1RM might be 76% on a tired day or 84% on a fresh day. Either way, the load doesn’t match today’s body.
Autoregulation closes the gap. The program prescribes “5 reps at RPE 8” instead of “5 reps at 80 kg”. The lifter picks today’s load to hit RPE 8. Which might be 78 kg today and 82 kg next session. Same training stimulus, adjusted for daily readiness.
The three approaches
RPE-capped sets
Pattern: prescribe target reps and target RPE. Lifter picks the load.
Example: “Top set: 3 reps RPE 8. Backoff: 4 sets of 5 reps at 80% of top set.”
Pros: No special equipment. Self-adjusts for readiness. Works for any rep range. Cons: Requires the lifter to accurately self-rate (4-8 weeks to learn). Bias-susceptible.
This is Coach Sheet’s default. The RPE calculator gives starting load suggestions; the 3-test PR protocol calibrates the table to the specific lifter over time.
Velocity-based training (VBT)
Pattern: attach a velocity tracker to the bar. Stop the set when bar velocity drops below threshold.
Example: “Squat 5 sets, stop set when bar velocity < 0.55 m/s.”
Pros: Most objective form of autoregulation. No rating bias. Excellent for peaking blocks. Cons: Requires equipment ($150-1000). Velocity targets need per-lifter calibration. Bad fit for online coaching workflows where the coach can’t watch the meter live.
VBT is the right call for advanced lifters with a coach who has equipment. Most online coaching clients aren’t there.
Daily-max-driven percentages
Pattern: lifter works up to a “daily max” each session (a top single at RPE 8-9). All other sets are percentages of that daily max.
Example: “Work up to a daily max single at RPE 8. Backoff: 5 sets of 3 at 80% of daily max.”
Pros: Simpler than RPE rating. Self-adjusts via the “today’s max” anchor. Cons: Higher injury risk if form degrades during the daily max. Mental load of taking near-max singles every session.
This is the conjugate-method approach. Works for competition athletes; overkill for most online coaching clients.
How to pick
Default: RPE-capped, Tuchscherer-style table for percentage suggestions, three-test PR protocol every 6-9 weeks to calibrate.
Switch to VBT when you have equipment and your client is peaking for competition.
Switch to daily-max when training for raw power and the lifter handles near-max singles well mentally.
Common mistakes
Switching approaches mid-program. Pick one for the entire mesocycle (4-12 weeks). Switching mid-cycle disrupts the lifter’s sense of “what RPE 8 feels like.”
Using RPE without calibration. A new client’s RPE 8 isn’t the same as a 5-year client’s RPE 8. Run a 4-8 week calibration period; watch logs against the table and note systematic drift.
Hard-RPE-capping during deload weeks. Deloads are intentionally easy. Prescribing “RPE 6 for all sets” lets the lifter pick too-light loads; you want the deload as fixed percentages of recent loads.
Treating novice lifters’ RPE feedback as gospel. They don’t know the scale yet. For the first 4-6 weeks, ask in RIR terms (“how many reps could you have done”) and translate yourself.
How it works in Coach Sheet
The Workout Plan tab has separate columns for Prescribed RPE, Achieved RPE, Load, and Reps. Save Week reads the achieved RPE alongside logged loads to calculate the next week’s prescribed load using the progression rules from the Save Week guide. The calibration loop closes inside the sheet without manual recompute: prescription → execution → feedback → adjustment → next prescription.
Sources
- [1]
- [2]
- [3]