Session RPE: definition, formula, example

Session RPE is the simplest training load measure that holds up in research: rate how hard the whole session was on a 0 to 10 scale, multiply by its duration in minutes, and you have a load in arbitrary units. Carl Foster and colleagues described it in the late 1990s and tested it against heart-rate methods in 2001. It needs no sensor, works for any sport, including strength and technical sessions, and captures things heart rate cannot see, like heavy legs on a downhill or the mental cost of a race.

How session RPE is calculated

The scale is a version of Borg’s CR-10: 0 is rest, 1 very easy, 3 moderate, 5 hard, 7 very hard, 10 maximal. The question is about the session as a whole, not the hardest interval or how you feel now. Foster’s protocol asks for the rating about 30 minutes after the end, once the last effort no longer dominates the answer. In practice, rating it when you upload the activity, the same day, works well enough if you stay consistent.

A worked example

A runner’s week, rated session by session:

• Tuesday, 60 min with 5 × 4' at threshold, RPE 7: 420 AU
• Wednesday, 45 min easy, RPE 3: 135 AU
• Thursday, 50 min steady, RPE 4: 200 AU
• Saturday, 100 min long run, RPE 5: 500 AU
• Sunday, 40 min recovery jog, RPE 2: 80 AU

The weekly load is 1,335 AU. The threshold session is 15% of the training time but almost a third of the load, which matches how most runners would describe their week. Foster (1998) also derived two weekly indicators from the daily values. Monotony is the mean daily load divided by its standard deviation: here 191 / 184 ≈ 1.0, with two rest days and real contrast between hard and easy days. Strain is weekly load × monotony, about 1,390. Training the same moderate amount every day pushes monotony up, and in Foster’s data illnesses tended to follow weeks of high strain.

How to use it in training

Rate every session, including the easy ones and the gym. Gaps make the weekly total meaningless. Then look at trends: a week 30% above your recent weeks is a jump, whatever the absolute number. Comparing the planned effort with the rated one is often more telling than the load itself. An easy run rated 5 instead of 3, or a threshold session that felt like a 9, says something about fatigue, heat or illness before any curve does.

Session RPE also works best next to an objective measure. When TRIMP stays flat while session RPE climbs for the same workouts, the body is paying more for the same output. Coaches use exactly that gap to adjust the next days.

In OpenAthlete, every activity has an RPE from 0 to 10, and a coach can require it before a session counts as done. The training load view shows session RPE (RPE × minutes) alongside heart-rate TRIMP. If the AI features are switched on, a few short questions after an activity can fill in the RPE from the athlete’s answer. More on the training load page.

Limits

It is subjective, which is both the point and the weakness. Ratings drift with mood, motivation and experience; beginners tend to compress everything into 4 to 7. The arbitrary units do not convert to TRIMP or TSS, so the numbers cannot be mixed in the same curve without rescaling. Duration weighs linearly: a 3-hour easy ride at RPE 3 (540 AU) outscores a hard 40-minute interval session at RPE 8 (320 AU), which is right for total load and wrong for the specific stress of intensity. And a rating says nothing about why the session was hard. The review by Haddad and colleagues (2017) finds session RPE valid across many sports, with correlations to heart-rate methods that are generally good but far from perfect. Use it as a second opinion that is always available, not as the only number.