How the calculator works
Every number on this page comes from one idea. A set of 5 at RPE 8 stopped two reps short of failure, so it was as hard as a set of 7 taken to failure. The RPE chart turns that count, reps to failure, into a percentage of the lifter's max: 7 reps to failure is 81.1 percent. Divide the weight by the percentage and you have an estimated one rep max, usually written e1RM.
The second half runs the same arithmetic forwards. Pick the reps and the RPE you want next, read their percentage off the same chart and multiply it by the e1RM. At the default input, 100 kg for 5 at RPE 8 is an e1RM of 123.3 kg, and a set of 3 at RPE 8 is 86.3 percent of that, which the tool rounds to 107.5 kg because that is a load a gym can put on a bar.
The two tables under the result answer the questions coaches ask next: what the same rep count weighs at every other RPE, and what every rep count weighs at the same RPE. The full chart sits under them with both sets marked, so you can see where the numbers came from.
The RPE chart, and where it comes from
The chart is Mike Tuchscherer's, from Reactive Training Systems, and it is the one most RPE calculators are built on. It covers 1 to 12 reps at RPE 6 to 10 and reads the same way in every cell: add the reps done to the reps left in reserve, and every combination with the same total sits at the same percentage. That is why 1 rep at RPE 8, 2 at RPE 9 and 3 at RPE 10 all read 92.2 percent.
The top of the chart, up to ten reps to failure, is the part every published version agrees on, to within rounding. Past that the values continue in a straight line, and calculators extend the line differently: at 12 reps and RPE 8 one widely copied version reads 65.6 percent and this one reads 62.6. The tool flags any set that lands out there. Treat those loads as a starting point and let the next set correct them.
A set of 5 at RPE 10 reads 86.3 percent, which puts 100 kg for 5 at an e1RM of 115.9 kg. The one rep max calculator's seven formulas put the same set between 112.5 and 119 kg, so the two methods agree when the set really went to failure. The RPE chart earns its place on every set that did not.
RPE and RIR
RIR, reps in reserve, is the same scale counted from the other end: RPE is 10 minus RIR. Some clients answer "how many more could you have done?" more honestly than "how hard was that out of ten?", and the chart works identically with either. Half points mean the lifter was between two answers: RPE 8.5 is definitely one rep left, maybe two.
| RPE | RIR | A set of 5 is |
|---|---|---|
| 10 | 0, nothing left | 86.3% of 1RM |
| 9.5 | 0, but a little more weight was possible | 85.0% of 1RM |
| 9 | 1 | 83.7% of 1RM |
| 8.5 | 1, maybe 2 | 82.4% of 1RM |
| 8 | 2 | 81.1% of 1RM |
| 7.5 | 2, maybe 3 | 79.9% of 1RM |
| 7 | 3 | 78.6% of 1RM |
| 6.5 | 3, maybe 4 | 77.4% of 1RM |
| 6 | 4 | 76.2% of 1RM |
Programming with it: a top set, then back-off sets
The most common way to use RPE is a top set followed by back-off work. The client works up to one set at a prescribed RPE, say 3 reps at RPE 8, and whatever that weighs on the day sets the rest of the session. Put the top set into the calculator and read the back-offs from the tables.
A worked example. A client squats 140 kg for 3 at RPE 8. That is 5 reps to failure, 86.3 percent, so the e1RM is 162.2 kg. The back-offs are 4 sets of 5 at RPE 7, which is 8 reps to failure and 78.6 percent: 127.5 kg once rounded to the bar. On a flat day the same top set only reaches 130 kg at RPE 8, and every back-off drops with it, to 117.5 kg.
That is the reason to program by RPE at all. A percentage written last month assumes the client is the same person every day. An RPE target lets the load follow how they actually showed up, which is what autoregulation means in practice.
Teaching clients to rate RPE
The method rests on the client's rating, and new lifters rate badly. They usually say 9 when they had three reps left, because an unfamiliar heavy set feels like a maximal one.
Calibrate before you rely on it. For the first few weeks, ask after each working set how many more reps they could have done and write that down instead of a number out of ten. Now and then, on an exercise where failure is safe, such as a machine press or a leg extension, let them take a set they rated at RPE 8 all the way to failure. If they get five more reps instead of two, both of you have learned what an 8 feels like.
Until the ratings match the results, treat the tool's loads as a ceiling rather than a target, and prescribe from the lower end.
Where it breaks
The chart describes barbell strength work in sets of roughly 1 to 10. On isolation exercises, machines and high-rep sets the percentages drift, and on dumbbells the jumps are too coarse to hit them anyway. Use RPE there as an effort target and let the load be whatever produces it.
Fatigue inside a session moves the numbers too. The fourth set at the same weight is harder than the first, so an e1RM from a late set reads low. Use the first hard set of the day, not the last.
And an e1RM from RPE is only as good as the rating. Two estimates from the same client a week apart that differ by 10 kg usually mean the rating changed, not the client.
Keeping RPE in the log
A calculator handles one set on one day. The useful version is every set, for every client, over months, so you can watch a client's RPE 8 get heavier.
In Onyx Coach every set has an optional RPE, both in the prescription you write and in what the client logs from their phone, beside the weight, reps and rest. Onyx also stores a 1RM per exercise per client and lets you write sets as a percentage of it, so the e1RM you get here can go into the client's profile and every percentage in their block moves with it.
