RPE and Microscopy: Adjust The Weight, Not Reps Or Sets

When using RPE, the variable to adjust is load

This is the 3rd or 4th time I’m writing about RPE, and I suspect there’ll be at least 3-4 more times, because it’s both a contentious and easily misunderstood topic. But it’s also a fun one, so let’s dig in. If you haven’t read any of my prior pieces on RPE, start here and then move onto this one, before reading further.

RPE-Based Programming vs Programming Modified by RPE: Course vs Fine Microscope Adjustments

Speaking very generally, there’s two basic types of ways that RPE is used by coaches to auto-regulate training.

RPE-based programming

RPE-based programmers base their whole session on RPE. They don’t go in with a specific goal weight to hit. They go in with a rep number and RPE target, and keep going till they hit the target # of reps at the target RPE. Sometimes even the amount of sets/total volume they do is regulated and adjusted by RPE, doing more on days they’re less fatigued (as indicated by RPE) and vice versa.

An example would be, the lifter is assigned to work up to 1 set of 1 rep at RPE 8, then back off to a set of 6 reps at RPE 7 and keep repeating sets of 6 at that weight, until it becomes RPE 8. No weight or % is assigned, and not even a specific number of sets - what determines when to stop is fatigue within the session, when the backoff sets go from RPE 7 to RPE 8.

If you remember your microscope from high school or college biology class, it had both a course and a fine focal adjustment. The coarse adjustment knob facilitates initial focusing by rapidly moving the objective lens closer to or farther away from the specimen, while the fine adjustment knob serves a more refined purpose. Its primary function is to fine-tune the focus of the microscope to achieve optimal clarity and detail.

Using RPE this way is like using it for BOTH the course and fine adjustment knobs. There’s no specific weight or number of sets to guide you. It’s all down to RPE.

Taking you back to the good old days

Programming modified by RPE

In contrast, the way I and many others use RPE is via more classic programming in advance, using RPE as a tool to modify it if necessary. We give a pre-assigned weight (or % or range), rep, and set assignment in advance, and then we give the lifter an RPE target. The RPE target serves as an optional modifier: IF, and only IF, the weight/range/percentage assigned is outside the target RPE, then adjust the weight of the next set accordingly. Don’t adjust sets. Don’t adjust reps. Adjust weight. And only on the next set, don't make any changes to the current set.

An example here would be: “3x3 at 405 lbs, RPE 8-8.5.”

If the first set at 405x3 is anywhere in the RPE 8-8.5 zone, stay put for the subsequent set. If it’s clearly harder or easier than RPE 8-8.5, adjust the weight slightly up or down for the next set following the guidelines below.

Outside of the rare exception when that first set is so easy that it's RPE 5 or below, you should count that first set as the first work set, but go up in weight (not adding sets or reps), the next set, as follows: If the weight was just a little off the RPE target, like 1 point, adjust the weight by 2-3%. If it was 1.5-2 points off, adjust weight by 4-6%. If it was 2.5-3 points off, adjust 5-8%.  IN WEIGHT ONLY.

Using RPE this way is akin to using it only to make fine adjustments to the microscope. The course adjustments are the assigned weight/sets/reps, and the RPE is for fine adjustments only, and only if necessary. If the initial course adjustments gets you close enough, no need to adjust further. If some adjustment is needed, it will be fine, not course, and only needs to be done in the realm of load/weight, not sets or reps, outside of exceptional circumstances.

The course knob is larger and moves the objective lens in larger increments, making it suitable for quickly bringing a specimen into view and starting the focusing process. This is why it’s akin to the pre-planned weight/set/rep assignment. A lifter or coach should know roughly what they can do on a given day. Of course both random daily fluctuations in readiness and a lifter’s fatigue/recovery status play a role, but outside of rare exceptions, the range isn’t unlimited. You should be able to predict and know within a fairly narrow band what should be done that day, most of the time.

Just as the coarse adjustment knob is particularly useful when initially locating the specimen, changing magnification levels, or shifting focus between different specimens, the planned weight/sets/reps assignment is useful when establishing baseline expectations for the day.

It’s like giving you a destination on a map: I’m 100% certain of the city, 99.9% certain of the neighborhood, 99% certain of the block, and 95% certain of the exact address. That’s your set/rep/weight assignment. RPE is there in case you need to make a fine adjustment to the directions in terms of address, if I got it slightly wrong.

Why Weight Only? What’s the Big Deal About Adding Reps?

To answer this Q, we need to talk about tonnage and its effects on the program. Tonnage is simply the total amount of weight lifted in a session, calculated for each lift. So if the assignment is 4 sets of 6 squats at 225 lbs, the equation is simply 4x6x225 = 5,400 lbs of squat tonnage for the day. We don’t typically count warm-up sets in the tonnage.

Let’s say in our example here, the 4x6 squat assignment was given along with an RPE target of 8.5, but the first set was only RPE 7, so the lifter immediately (instead of waiting till the next set) does 2 more reps for a total of 8 instead of 6 reps at 225. Then the next 3 sets are done for 8, 8, 7 reps, making a total of 31 reps instead of 24, all at 225 lbs.

This changes the tonnage from 5400 to 6975, a massive change of 1575 lbs, or 29% more tonnage, than the initial assignment.

Whereas if the lifter follows the guidelines above, he will stop at 6 reps on the first work set. Then he will add 5% WEIGHT to the next work set, which rounds to 235. Let’s say that’s adjustment was just right and he does 235x6 for the remaining 3 sets. The equation now is (225x6) + (235x18) = 5580. This is a change of only 180 total lbs or 3.3%, from the initial assignment. A much more reasonable and productive adjustment, vs a massive system shock of a 29% increase.

We can see then, why this fits our purpose better than adjusting reps. Rep adjustment makes for a massive change in tonnage, 29% in our example above, which is akin to using it for course adjustment instead of fine. Whereas using RPE exactly as described in the guidelines and example here, made for a 3% adjustment, a perfect analogy for the fine adjustment knob on the microscope that we want to use RPE for.

Does Using RPE for BOTH Course and Fine Adjustment Work for Anyone ?

Short answer is yes. There are some people who can utilize the first way of RPE, the “RPE-Based Programming Approach.” However in my experience, this is the exception rather than the rule, and is almost exclusively found among those who make lifting their main thing, either competitive lifters or the most serious of hobbyists who make it into a lifestyle. Only people with an extremely high level of awareness of their own bodies and the way they respond to training can do this successfully.

Even among people like this (such as myself), it doesn’t work best for everyone. Having tried both ways, I’ve had more personal success, even as an advanced lifter with a 600+ squat, 700+ deadlift, and nearly 500 lb bench, programming for myself the second way - using RPE only for fine adjustments, and classic traditional programming to set the expectation for the day as the course adjustment. I’ve also lost count of how many people have come to me either as paying clients or to ask a question, whose story was a variation of: “I did Starting Strength (linear progression) then switched to XYZ RPE-Based Programming afterwards, and now I’m weaker than I was at the end of LP.”

For most people, the second way, “Programming Modified by RPE,” wherein we use RPE only like the fine adjustment on a microscope, is a much more realistic way to auto-regulate training to optimize for unknowable-in-advance performance fluctuations, without asking way too much of the average person.