RPE is a 1 to 10 rating of how hard an effort feels. In strength training it pairs with RIR, reps in reserve: RPE 10 means 0 reps left, RPE 8 means 2 in reserve (Zourdos et al. 2016). Autoregulation means matching today’s load to today’s readiness instead of following a fixed percentage. For fingers, where readiness swings daily and injury risk is real, that matters most.
Key takeaways
- The scale is anchored by reps, not adjectives: RPE 10 = 0 reps left, RPE 8 = 2 in reserve, RPE 7 ≈ 3 in reserve (Zourdos et al. 2016).
- Helms et al. (2018, Frontiers in Physiology) matched sets and reps and found RPE-based loading produced a small 1RM advantage over fixed %1RM.
- A 2025 network meta-analysis ranked load-prescription methods by SUCRA: APRE 93.0%, RPE 66.8%, velocity-based 27.0%, percentage-based 13.2%.
- For unfamiliar grip positions, Lattice recommends reverting to RPE with a target of 5 to 7, well below the 7 to 8 band used for finger strength work.
What RPE and RIR mean
RPE rates how hard an effort felt from 1 to 10, and RIR counts the reps you could still have done. They are the same judgment read from opposite ends. RPE 10 means 0 reps left, RPE 8 means 2 in reserve, and RPE 7 is roughly 3 in reserve (Zourdos et al. 2016). Saying a set was RPE 8 and saying you had two reps left in it are not two pieces of information. They are one.
| RPE | Reps in reserve | What that means on the wall |
|---|---|---|
| 10 | 0 | Nothing left. The set ended because you failed, not because you chose to stop. |
| 8 | 2 | Two more were available and you left them. Hard, controlled, repeatable. |
| 7 | ≈ 3 | Clearly working, clearly not close to the limit. |
The RIR column is what makes the scale usable. “How hard did that feel” is a question with no wrong answer, which is another way of saying it has no answer at all. “Could I have done two more” is specific, and you can be wrong about it, and being wrong about it is information. That is the whole reason RPE beats a vague sense of having trained hard.
Finger work has no reps, so the translation for isometrics is seconds or effort left in the tank. A near-limit hang you could have held about two seconds longer is an RPE 8 hang. A hang that ended because your fingers opened, whether you agreed with the decision or not, was RPE 10. The same logic covers a limit boulder: RPE 8 is the go you could have repeated straight away, RPE 10 is the one after which the next attempt is visibly worse.
Rate the set in the few seconds after it ends and write the number down before you rest. This is the single habit that decides whether RPE is useful to you or just a mood you record. The common mistake is rating the session instead of the set. Fatigue arrives late, so a set that was genuinely RPE 7 while you were doing it can feel like RPE 9 twenty minutes later, and a log full of those inflated numbers will quietly talk you into training too light for a month. Rate the effort, not the aftermath.
What you are rating is a specific piece of work, which means the protocol has to be defined before the number means anything. Max hangs, repeaters and density hangs are three different jobs at three different intensities, and the three main hangboard methods sets out what each one is asking for.
Autoregulation vs fixed percentages
A fixed percentage assumes you are identical every day, and you are not. Sleep, stress, skin condition and accumulated fatigue all move your true max, so a percentage of a number you tested three weeks ago can land anywhere from a warm-up to a set you had no business taking. The percentage is precise about the load and completely silent about you.
Helms et al. (2018, Frontiers in Physiology 9:247) put the two approaches against each other with sets and reps matched, and RPE-based loading came out with a small advantage in 1RM. Small is the honest description, and the mechanism behind it is unglamorous: on a good day the fixed percentage under-loads you, on a bad day it over-loads you, and across a training block those two errors do not cancel out. One of them costs you an adaptation you could have had. The other costs you a week.
A 2025 network meta-analysis ranked four ways of prescribing load for maximal strength and produced this order: APRE 93.0%, RPE 66.8%, velocity-based 27.0%, percentage-based 13.2%. Read those as SUCRA ranking probabilities rather than as amounts of strength gained. They describe how consistently each method came out on top across the pooled comparisons, not how much stronger anyone got. The ordering is the finding worth carrying away, and the ordering puts the two effort-anchored methods above the fixed-percentage scheme that most published training programs are still written in.
The practical reading for a climber is narrower than the study’s own framing. Percentage-based prescription depends on a max test, and a max test is a snapshot with an expiry date. The longer it has been since you tested, the more of your programme is arithmetic performed on a number that has stopped being true. Effort ratings do not expire, because you generate a fresh one every set.
None of this is an argument for training without a plan. Structure decides what you train, in what order, and for how long, and it has to be set weeks in advance because that is what a training block is. RPE only decides how heavy today’s version of that structure turns out to be. If you want the block shapes themselves, how to structure a training block covers the periodization models; autoregulation is what you pour into them.
Applying RPE to fingers
Finger tissue is unforgiving, so autoregulating it doubles as injury insurance. On a strong day, add weight or move to a smaller edge until you reach the target RPE. On a flat day, take weight off. The prescription is the effort, not the number on the loading pin, and matching yesterday’s kilos on a day your fingers are not there is how a planned RPE 8 turns into an actual RPE 10.
Roughly RPE 7 to 8 covers most finger strength work, and 5 to 7 is the band for new grip positions and for rehab. Lattice explicitly advises reverting to RPE with a 5 to 7 target when you introduce an unfamiliar hang, and the reasoning generalizes well: on a grip you have never trained, your load history is worthless as a guide, because it was built on a different position with different tissue doing the work. The first few sessions in a new position are calibration, and calibration at RPE 9 is an injury with extra steps.
Here is the same session on two different days. The plan says a hard set at RPE 8, two efforts in reserve. On the good day you find that at a load you have never held, and the log records a personal best you did not plan. On the flat day the same effort arrives several kilos lower, and the log records a completed session at the prescribed intensity. Both days did the job. Only one of them would have survived a fixed percentage.
The common mistake is treating autoregulation as a one-way permission slip: a green light when you feel strong, never a brake when you do not. Cutting the load on a bad day is the half that keeps you climbing. If your logged loads only ever move upward, you are not autoregulating, you have just found a slower way to write percentages.
Two things RPE will not do for you. It does not replace planned recovery, because a planned deload week is scheduled in advance and happens whether or not today felt fine. And it does not manage frequency: an effort rating is a single-session judgment, while finger load accumulates across the whole week, so how often you can actually hangboard is a separate calculation that RPE does not make for you.
Sources
- Zourdos, M. C. et al. (2016). RIR-based RPE scale for resistance training.
- Helms, E. R. et al. (2018). Frontiers in Physiology, 9, 247.
- Network meta-analysis of load-prescription methods (2025). SUCRA rankings: APRE, RPE, velocity-based, percentage-based. ScienceDirect S1728869X25000590.
Common questions
What's a good training RPE for fingers?
Roughly RPE 7 to 8 for finger strength work, and RPE 5 to 7 when you are introducing a new grip position or working back from injury. Lattice Training recommends the lower band for unfamiliar hangs. RPE 9 to 10 on fingers is rarely worth the risk, especially when you are cold or tired.
Is RPE better than percentages?
For most climbers, yes, by a small margin. Helms et al. (2018) matched sets and reps and found RPE-based loading gave a slight one-rep-max advantage over fixed percentages, and a 2025 network meta-analysis ranked autoregulated methods (APRE 93.0 percent, RPE 66.8 percent) well above percentage-based loading (13.2 percent).
Can beginners use RPE?
Yes, with practice. Beginners typically over- or under-rate effort by about one point at first, and accuracy improves over a few weeks of deliberate rating, especially near failure. Rate every set immediately after it ends, write the number down, and treat your first month of RPE numbers as calibration rather than data.
Jan Luther is a climber and the developer behind Boulder Trainer, the hangboard app with guided workouts, voice coaching and a habit tracker.
Keep reading
Hangboard Training: The Complete Guide for Boulderers
Everything a boulderer needs to train fingers safely: the three main hangboard methods, the research behind them, load and edge guidelines, and a beginner-safe first protocol.
Deload Weeks: Why Training Less Builds More Progress
A deload isn't lost time — it's when adaptation happens. Here's how to program a deload week, what to reduce, and the science of why less becomes more.