Training to failure or two reps in reserve: how close to go

You hit the eighth rep on the bench and you can feel that two more are in there - just about. Do you rack it, or grind until your elbows fold? Whole training weeks turn on that decision, and it is a dosing question, not a test of character.

The shorthand you need is RIR, reps in reserve. RIR 2 means you stopped two reps before you would have been unable to complete another one in decent form. Failure is RIR 0. The real question is what those two banked reps actually cost you.

What we know with confidence

Proximity to failure drives growth, not failure itself. The evidence here is consistent: when the number of sets is matched, sets finished roughly one to three reps short of failure produce muscle growth very similar to sets taken all the way. Any difference is small and hard to detect with the tools we have.

It is equally clear that easy sets do not do the job. If you routinely stop five or more reps short, growth drops off noticeably. There is an effort threshold below which a set simply does not send a strong enough signal, and research agrees on that point.

The third solid finding is that failure is expensive. A set taken to the end leaves a bigger drop in strength, slows recovery and cuts the reps you manage in the sets that follow. The effect is largest on heavy multi-joint barbell lifts and smallest on machine isolation work.

For strength the picture is cleanest. When the goal is a bigger single, training to failure offers no advantage and sometimes lags behind, because it costs you quality sets and ruins bar speed. That is why serious strength programmes almost never call for failure on the main lifts, and it applies most of all to the deadlift, where a misjudged rep is the most costly.

Finally, load matters less than was long assumed. Lighter sets, somewhere around 30 to 50 per cent of your maximum, taken close to failure, build muscle comparably to heavy sets. That does not hold for maximal strength, but for muscle size the evidence is fairly firm - and it shows better than anything else that effort is what counts, not the number on the bar.

What is likely but not settled

The exact window is unresolved. Whether RIR 0-1 beats RIR 2-3 in trained lifters, and by how much, is poorly researched so far. Most comparisons run six to twelve weeks with small groups, so differences of a few per cent cannot be measured reliably.

It is plausible, though not proven, that experienced lifters need to work closer to failure than beginners to get the same signal. The reasoning is sound and the data lean that way, but there is no firm evidence.

The popular arrangement of taking only the last set of an exercise to failure while keeping earlier sets at two in reserve looks sensible and works well in practice. Still, it is an educated guess rather than a tested result.

The link between failure and injury is the weakest spot of all. It feels obvious that form breaks down when you run out of strength, yet direct evidence that training to failure causes more injuries is thin. Treat this as a cautious assumption rather than a fact.

The same applies to the long game. What happens after a year of training to failure on everything, and whether accumulated fatigue quietly eats your results, is largely an open question.

What gets passed on wrongly

The first error is the rule that a set does not count unless it ends in failure. The data do not support it. A set at two in reserve carries nearly full value at a much lower price.

The second is mixing up muscular and technical failure. Muscular failure is the moment another rep in good form is impossible. Technical failure is the moment form starts to break - the back rounds, the hips shoot up, the bar drifts. For most gym-goers, technical failure is the real limit and the place to stop.

The third is judging RIR by the burn. Burning arrives early and tells you very little. The more reliable cue is speed: when a rep that moved smoothly earlier suddenly takes noticeably longer, you have one or two left.

The fourth, and the most common, is that people badly overestimate how close they are. When lifters stop where they believe failure is, they typically still have three to five reps in the tank. For most amateurs the bigger problem is training too easy, not too hard.

The fifth is treating soreness as a scoreboard. Sore muscles the next day reflect novelty and loaded stretch, not how well the session was dosed.

What this means for your training

• The default: finish most working sets one to three reps short of failure. That band gives you nearly everything for very little cost.
• By exercise: on machines and isolation work, 0-1 in reserve is fine and the occasional set to failure is cheap. On heavy presses, squats and deadlifts, hold 2-3, because technique breaks first.
• If strength is the goal: on sets of one to five at 85 per cent and above, stop as soon as bar speed visibly drops. That usually lands around 2-4 in reserve.
• Dose of failure: one or two sets to failure per muscle group per week, placed at the end of a session, delivers whatever failure has to offer.
• A quick check: if reps fall by more than 25 to 30 per cent on the next set, the previous one went too deep or your rest was too short. Allow 2 to 3 minutes after sets near failure.
• Coming back after time off: keep 3-4 in reserve for the first few weeks back and rebuild technique before effort.

Nutrition does not change dramatically because of how close you train to failure, but recovery gets more expensive. Keep protein in the range growth needs anyway, roughly 1.6 to 2.2 grams per kilogram of body weight, and avoid aggressive calorie cuts during blocks where you often push to failure. Sleep matters more here than any supplement.

When to see a doctor

Sharp pain that shoots rather than burns, sudden weakness in one arm or leg, swelling or bruising after a set, pins and needles that do not settle once you leave the gym - these are reasons for a check-up, not for another set. The same goes for chest pain, dizziness or fainting under strain, and for dark urine with severe, unusual soreness a few days after a very hard session. If you have high blood pressure, a heart condition or a previous spinal injury, agree with your doctor how close to failure you should train before deciding it yourself.

In short

Proximity to failure is what builds muscle, and the last two reps add little while costing a lot in fatigue. Sit at one to three in reserve, save failure for isolation work and final sets, and stop on heavy lifts while your form is still clean. The most common mistake in an average gym is not too much failure - it is sets that end nowhere near it.