Reps in reserve — how many more repetitions you could have completed before failing. Zero RIR is failure. Two RIR means you stopped with two good reps left. It is the standard way of recording effort on a set, and it is more reliable than a percentage of one-rep max because it accounts for how you actually feel that day.

Why does Base need an RIR rating?

Two things depend on it. Your next prescription for that lift is scaled by how far the last set stopped short — twelve reps at four RIR means the bar was never the limiting factor, and waiting a week to discover that is a week spent not growing. And weekly volume only counts sets close enough to failure to have caused growth.

What happens if I do not rate a set?

The set is still logged and still shown, but it is reported separately from your counted volume rather than being assumed to have counted. Base will not guess, because a volume number you cannot trust is worse than no volume number at all.

Why are the RIR targets different for compounds and isolation?

A squat taken to failure costs systemic fatigue that shows up in the next three sessions; a lateral raise taken to failure costs almost nothing. So compounds are held further from failure — roughly two to three reps — and isolation work is taken closer, around zero to one, which is where the stimulus is cheapest to buy.

Is training to failure better?

The evidence says no. Refalo and colleagues in 2023, pooling fifteen studies, found training to failure was not superior to stopping short of it, and that the relationship is non-linear — inside roughly zero to five RIR closer to failure is modestly better, and above about five reps in reserve hypertrophy falls off steeply. So the job is staying inside the band, not maximising effort.

Rate Your Sets With RIR, and Why It Matters

What the number means

Zero RIR is failure. Two RIR means you stopped with two clean reps left in the tank.

Rate it honestly and slightly conservatively. Most people under-report — a set that felt like one rep left was frequently two or three — and the correction is easy once you notice that your prescriptions keep jumping.

The two things it drives

Your next prescription

The progression engine reads the last working set on a lift. The rule set is double progression with an RIR clause:

What happened	What happens next
Every set at the top of the rep range	Add load, scaled by the RIR, reset reps to the bottom
RIR above the stimulus ceiling	The bar was far too light — jump it now
Any set below the bottom of the rep range	Deload about 10% and rebuild
Anything else	Hold the load, add a rep

That RIR clause is the whole difference between this and rep-only progression. Twelve reps at zero RIR is a genuine ceiling and earns one increment. Twelve reps at four RIR means the load was never the limiting factor, and a rep-only scheme would spend another week discovering that.

The load jump is scaled by the rating rather than being a fixed step, so a set that stopped a long way short moves further.

Whether the set counts

Weekly volume only counts sets taken close enough to failure to have caused growth. A set above the stimulus ceiling happened — it is in your log, it produced fatigue — but it did not buy adaptation, and rolling it into "sixteen sets of chest" is how a programme reads as compliant while producing nothing.

Warm-ups never count either. Details in checking your weekly volume.

The target bands

Compounds sit further from failure than isolation, deliberately.

Role	Target RIR	Why
Compound	~2–3	A squat at 0 RIR costs systemic fatigue that shows up in the next three sessions
Isolation	~0–1	A lateral raise at 0 RIR costs almost nothing, and that is where the stimulus is

The role is derived from the lift. Seeded exercises carry it; a custom lift you typed in has its role inferred from its name, falling back to compound — the more conservative band and the smaller load step — when the name carries no signal.

What the evidence actually says

Refalo et al. 2023 pooled fifteen studies and found training to failure is not superior to stopping short of it, with a non-linear relationship: inside roughly 0–5 RIR, closer to failure is modestly better; above about 5 RIR hypertrophy falls off steeply.

So the job is not maximising effort. It is keeping every working set inside the band where it counts, at the lowest fatigue cost that achieves it — which is exactly why the compound and isolation targets differ.

Where to rate a set

Two places:

In the live session, after logging.
In the edit grid afterwards, from fixing a logged workout — the RIR bar sits under each lift and a mis-tap can be cleared back to unrated.

Retroactive rating is genuinely useful. Rating in the moment is more accurate, but a session rated the same evening is far better than one never rated.

Common mistakes

Not rating at all. The most common, and it silently guts the volume screen. An unrated week looks like a light week.

Rating everything 0. If every set is at failure, either the ratings are optimistic or the programme is producing more fatigue than it needs to. Both are worth knowing.

Rating warm-ups. They are excluded from volume and from progression regardless. Rate working sets.

Rating from how it felt afterwards rather than how many reps were left. "That was hard" is not RIR. The question is strictly: how many more could you have completed.

Expecting the prescription to move without ratings. With no RIR the engine falls back to rep progression alone, which is slower and less responsive.

Treating a deload prescription as a failure. A set below the bottom of the range triggers a 10% deload because rebuilding from a load you can actually complete is faster than grinding a load you cannot.

What to do next

Rate every working set in your next session, including the ones that felt easy. Then open the weekly volume screen and compare the counted sets against the sets you performed — the gap is the work that is not currently buying you anything.