Answers for when your plan changes.

A lower prescription is a calculation, not a verdict. The engine reduces load when logged reps fall below target, when recovery signals are elevated, or during deload and recovery sessions, because the target reps at the target RIR are only reachable at the adjusted weight. Lower weight at the prescribed RIR delivers the stimulus the plan calls for; grinding the heavier weight would not. Each exercise shows a one-line basis for this week’s prescription on the workout screen.

The engine sets next week’s volume per muscle from your soreness, pump, workload, and performance data. A reduction means the data showed recovery lagging behind the current dose, so the dose was lowered to keep every set productive. The workout screen shows which rating caused the change. Fewer high-quality sets can be more productive than adding sets when recovery is already lagging.

No. A single low session moves the estimates less than it moves your mood. The engine weighs trends across sets and sessions; one underperformance is recorded as a data point, not read as a trend. If the next session lands on target, the plan continues unchanged. Sustained signals across multiple sessions are what trigger adjustments. Log the session honestly and train the next one as prescribed.

The deload is scheduled because accumulated fatigue masks the fitness you built. A week at reduced load lets the fatigue resolve while the adaptation remains; published research on mid-block deloads shows no loss of muscle size or strength from the reduction. Skipping it carries the fatigue into the next block and lowers the quality of every set there. The week is light by design, not by error. Train it as prescribed and the next block starts from a measurable baseline.

No. A recovery session is a planned reduction, not a failure state. When fatigue signals cross a defined threshold, the engine prescribes reduced load and volume so the muscle returns to a productive baseline. Logged history is unchanged and progression resumes from your measured values. Completing the lighter session is what restores the conditions for progress.

One to three days late: resume at the next uncompleted workout, unchanged. Four to ten days: repeat the last completed week’s prescriptions. More than ten days: start a reduced-volume week or a fresh block; your data is preserved either way. A missed session is an expected input, not a failure state.

A week off is unlikely to erase muscle tissue. After a week, muscles can look smaller because stored glycogen and water drop, not because tissue is gone; both return within days of resuming. The first session back often feels heavier than the logs say it should. Strength readings are typically normal again by the second or third session, and the engine adjusts the resumption targets for the gap. Resume at the next uncompleted workout.

Yes. Soreness tracks novelty, and it is expected to fade as your body adapts to a repeated movement. It is one input among several; the engine reads performance, pump, and workload alongside it. Rising reps and load at the same RIR are the evidence that the stimulus is landing. Report soreness as you actually feel it: a string of zeros is valid data, not a problem to correct.

Single misjudged sets do not distort the plan. Research on RIR estimation shows trained lifters are typically within about one rep of their true reserve, and accuracy improves with practice. The engine reads trends across sets and sessions, so individual errors average out. When unsure, report the higher number; the next prescription stays conservative and your calibration corrects itself over the following sessions.

Yes. Tap Undo on the set row, re-enter the correct values, and log again. The engine reads final logs; corrections within the session are clean.

Ramp to your first working set: about 12 reps at a light weight, 8 reps at a medium weight, then 4 reps near your working weight, resting about a minute between steps. Warmup sets are not logged. For a second exercise on the same muscle, one short feeder set is enough.

Two to four minutes on compound lifts, 90 seconds to two minutes on isolation work. Longer rest is never penalized. Resting too little reduces rep output, which the engine reads as fatigue.

Complete exercises in the listed order and stop when you are out of time. Order encodes priority. Unperformed sets are treated as skipped, not failed, and reduce that muscle’s counted volume for the week. If the prescription is consistently too large for your schedule, rate workload Too much and the engine reduces next week’s sets.

Yes. Replace when equipment is unavailable, joint pain recurs across two sessions, or the movement cannot be loaded near target. Keeping the same exercises across a block compounds progress, so replace with a reason. The incoming exercise starts from a new baseline; the muscle’s volume plan is unaffected.

MR marks a myo-rep set: one activation set taken close to failure, then short 5-15 second rests with small clusters of additional reps until the target total is reached. MM marks a myo-rep match set: match the rep total of the first myo-rep set at the same weight, resting as needed between clusters. The engine prescribes them on isolation exercises where they deliver equivalent stimulus in less time. The set row shows the procedure, and the glossary covers execution step by step.

MGFI is the Muscle Group Fatigue Index, a per-muscle model built from performance change, soreness, set completion, and workload. Pump is recorded as context, not a causal input. The Readiness Index measures short-window readiness. The Accumulation Index measures decayed multi-session training burden. Together they tell the algorithm when to push and when to deload.

No. The engine builds a baseline from your first session and refines it across the mesocycle using Epley-derived e1RM. Estimates improve as you log more data.

No. Every block builds muscle; calorie intake changes the rate, not the structure. Volume tiers are the in-app lever. Rate sessions honestly and the engine calibrates volume to your recovery in either phase.

Precision periodization engine. Calyber logs your sets, detects RIR drift, and recalibrates the next session’s prescription. Every weight and rep target shows the basis it was computed from.

Those apps log workouts. Calyber prescribes them and adapts based on what you logged. Periodization plus autoregulation, not just history.

Calyber is built on published hypertrophy research: volume-landmark periodization (MEV/MAV/MRV) and RIR-based autoregulation. The difference is transparency. Every prescription shows the inputs and formula it was computed from, and the full algorithm is documented on the algorithm page.

Most percentage-based programs assume your 1RM is fixed and accurate. Calyber treats e1RM as a moving estimate and uses RIR-based daily autoregulation, which can better reflect daily readiness than fixed percentages, especially when your true strength fluctuates.

The algorithm is deterministic and auditable. Every prescription shows the math it was computed from. "AI" often implies a black box; Calyber is the opposite.

Yes. The autoregulation runs on your inputs (RIR, soreness, performance), not on your demographics. 84 templates cover Upper/Lower, Push/Pull/Legs, Whole-Body, and 8 split categories.

Calyber is a Progressive Web App. Install it from your browser to your home screen on iOS or Android. The full feature set is available on mobile.

Not at launch. Import from other logging apps is on the post-launch roadmap.

After the 28-day trial, your past sessions, full history, and e1RM trend stay visible. Logging new workouts, the adaptive engine, recovery protocol, and deload prescription require Founding Member or Standard.

Logging new workouts needs an active plan after the trial. Nothing is deleted: your mesocycles, logs, and calibration remain visible in a read-only account, and subscribing restores full access whenever you decide. There is no penalty for deciding later.

Open Settings, then Plan, then Manage billing, to change plan or cancel. Cancellation takes effect at the end of the paid period; there are no further charges and no cancellation steps beyond that screen. Your data remains readable and exportable afterward (see "What happens to my data if I cancel?").

Your data is retained in read-only form. You can export logs as CSV or JSON (post-launch). Re-subscribing restores the ability to log new workouts at the Founding rate while founding seats remain open (see "What is Founding pricing?").

The 28-day trial collects no card. After your first paid charge, you have a 7-day good faith refund window for expectation mismatch, billing error, or sustained service unavailability. A refund ends the subscription. Full terms are on the refund policy page.

Founding pricing is $19/month or $199/year, locked for the lifetime of an active subscription: future price increases do not apply to an active Founding account. It is offered while founding seats remain open. If you cancel and return while seats are still open, you re-subscribe at the Founding rate. Once founding seats close, the Founding rate is no longer offered to new or returning subscribers and Standard pricing applies.