LIBRARY

RIR and Autoregulation

What RIR Is

RIR is a metric you self-assess immediately at the end of a set: how many more reps you could have lifted. It runs on an integer scale of 0-4, where 0 is the state of having reached failure, 1 means "I could have done one more," and 4 means "I had at least four more in reserve." Because reserve of 5 or more becomes hard to quantify, the common practice is to treat 4 as the ceiling value.

The scale itself was validated for resistance training by Zourdos et al. (2016), and Helms et al. (2016) organized how to implement it in an autoregulation context. Both papers arrive at the same conclusion: self-assessment accuracy rises with training experience (Zourdos et al., 2016).

Correspondence with RPE

RIR is a resistance-training-specific derivative of the 1-10 RPE (Rating of Perceived Exertion) scale. The correspondence is simple: RPE 10 = 0 RIR, RPE 9 = 1 RIR, shifting one step at a time.

Correspondence between RPE and RIR
RPERIRMeaning
100Reached failure
91Could have lifted one more
82Could have lifted two more
73Could have lifted three more
64At least four more in reserve
≤ 5—Grouped as "reserve to spare," not RIR

The reason RIR does not cover the region of RPE 5 and below corresponds to the findings of Hackett et al. (2012). Self-assessment of rep count is accurate near failure, but its reliability drops once there is reserve of 4 or more. This site treats RIR over the 0-4 range and groups anything beyond that as "reserve to spare."

As the Core of Autoregulation

A %1RM-based program (e.g., 70% 1RM × 8 reps) fixes the load. But daily variables such as fatigue, sleep, diet, and stress mean the same %1RM does not always translate into the same perceived effort on a given day.

A program that adopts RIR absorbs this fluctuation through a "target RIR." The instruction "raise the weight until 8 reps land at RIR 2" allows the same target to be met with a lighter load on low-condition days and a heavier load on high-condition days. This is the basic idea of autoregulation (Helms et al., 2016): taking the relative value of perceived effort, rather than the absolute value of the load, as the program's unit.

Contribution to Plateau Detection

RIR is a measure of effort within a single set, and at the same time a sign of progress or stagnation across multiple sessions. If RIR steadily drops at the same weight and the same rep count, neuromuscular adaptation is advancing. Conversely, if RIR climbs under the same conditions, recovery may not be keeping up, or the stimulus may be locally exhausted.

In the Renaissance Periodization autoregulation framework, weekly volume decisions combine multiple signs — failed lifts, persistent soreness, and an upward trend in RIR (Israetel et al., 2021). RIR is not meant to decide a deload on its own; cross-validation with other fatigue indicators is the premise.

Limits and Caveats

The greatest limit of RIR is the accuracy of self-assessment. The consecutive studies by Hackett et al. (2012, 2017) show that prediction accuracy near failure is at a practical level, but error widens in the region of 4+ RIR. Trainees with a short training history (under one year) also show greater variability in their predictions than experienced lifters (Zourdos et al., 2016).

For this reason, it is reasonable to operate RIR as a metric premised on (1) at least one year of continuous resistance training experience and (2) a habit of self-observation. For complete beginners, velocity-based training or fixed %1RM carries a lower incident rate. There is no rush to introduce RIR until a trainee can predict their own "one more rep" with a reasonable degree of accuracy.

The Semantic Distinction Between Unrecorded and 0

Once you start using RIR, there is a recording discipline that is often overlooked. A "set for which RIR was not recorded" and a "set recorded as RIR 0 (reaching failure)" carry entirely different meanings as data.

Treating the two together distorts weekly analysis. If you process a missing record as 0, every set that was not actually pushed to failure gets tallied as a set taken to failure, and you end up overestimating your average level of effort. Conversely, if you exclude missing records, the frequency of sets that genuinely reached failure disappears from the data.

Once you adopt RIR, "reaching failure" and "unrecorded" must be treated as distinct things. This is a precondition for accurately grasping your own training state while running autoregulation.

How DELT Handles It

In DELT, RIR is treated as an optional recording field. Enable it under Settings → Advanced Training Metrics, and a 0-4 RIR input appears in the record field of each set. A set that was not recorded and a set recorded as 0 are held as distinct things, and this distinction carries through directly to weekly summaries and plateau detection.

Putting It Into Practice

To use a subjective assessment in decision-making, you first need to verify whether your predictions match what actually happens.

  1. Record (2-4 weeks): Enable RIR in the settings above, and record 0-4 after each set. Do not change the program. Limit the goal to making "the distribution of your own predictions" visible.
  2. Calibrate and operate (ongoing): From your past reports, pull out sets that reached failure (RIR 0) and sets you recorded as "I had two more" (RIR 2), and check whether the lifting reserve in your next session was as predicted. Once your predictions match the reality, build in rules such as "raise the weight aiming for RIR 2" or "increase the weight by 2.5 kg (5.5 lb) if RIR is 4 or more" as the driver of Progressive Overload. You can maintain progression while absorbing daily condition fluctuation.

RIR is a recording field and at the same time a habit of observation. The accuracy of autoregulation rises through the practice of repeatedly translating your own bodily sensation onto the 0-4 scale.

Frequently Asked Questions

What is RIR (Reps in Reserve)?
RIR is a self-assessed metric of how many more reps you could have lifted to failure at the end of a set, rated on a 0-4 scale. 0 means you reached failure, 1 means you could have done one more, and 4 means you had at least four more in reserve. Zourdos et al. (2016) validated it for resistance training.
What is the difference between RIR and RPE?
RIR is a resistance-training-specific derivative of the 1-10 RPE scale, with a one-step shift: RPE 10 = 0 RIR, RPE 9 = 1 RIR, and so on. RPE 5 and below (RIR 5 and above) is grouped together as "reserve to spare" and is not assigned a number.
Is RIR appropriate for beginners?
Hackett et al. (2012, 2017) and Zourdos et al. (2016) show that self-assessment accuracy rises with training experience, and that trainees with a short training history (under one year) show greater variability in their predictions than experienced lifters. It is reasonable to operate RIR as a metric premised on (1) at least one year of continuous resistance training experience and (2) a habit of self-observation.
What is autoregulation?
It is a method of adjusting load intensity and frequency according to your condition on the day. Where a %1RM-based program fixes the load, the autoregulation framework organized by Helms et al. (2016) absorbs daily fluctuation through a "target RIR," taking the relative value of perceived effort, rather than the absolute value of the load, as the program's unit.
Why distinguish "unrecorded" from "RIR 0"?
Treating the two together distorts weekly analysis. If you process a missing record as 0, every set that was not actually pushed to failure gets tallied as a set taken to failure, and you overestimate your average level of effort. Conversely, if you exclude missing records, the frequency of sets that genuinely reached failure disappears from the data.
Can I decide on a deload from RIR alone?
No. In the Renaissance Periodization autoregulation framework, weekly volume decisions combine multiple signs — failed lifts, persistent soreness, and an upward trend in RIR (Israetel et al., 2021). RIR is not meant to decide a deload on its own; cross-validation with other fatigue indicators is the premise.

Related Articles

References