TRIMP vs Power vs RPE: Three Ways to Score Cardio Load (and When Each Wins)

There are three serious ways to turn a training session into a number, and people argue about which is best as though there were one answer. There isn't. Each wins in a different situation, and for anyone training more than one sport the deciding factor is usually not the one that gets discussed.

Power: the most accurate, in the narrowest window

Power meters measure external work directly, in watts. That is a genuine measurement rather than an inference, and it has properties the others would love to have. It does not drift over a long session. It is unaffected by heat, caffeine, sleep or how stressed you are. Two sessions with the same power profile represent the same mechanical work, always.

Load scores built on it, TSS and its relatives, take your intensity relative to threshold, weight the harder efforts disproportionately, and multiply by time.

The catch is coverage. Power works beautifully on a bike. It works well on rowing and ski ergs. Running power is a different animal — it is a model built from accelerometer data rather than a direct measurement of work done, and different devices produce meaningfully different numbers for the same run, so it isn't comparable across brands the way cycling watts are. In the water there is essentially nothing.

Be clear about the conclusion this leads to: if you ride a bike and only ride a bike, and you own a power meter, use power. Nothing below is more accurate for that case.

Heart rate and TRIMP: the universal translator, with a lag

TRIMP — training impulse, from Eric Banister's work in the 1970s and 80s — takes duration and multiplies it by how hard your heart was working, using heart rate reserve with a weighting that penalises higher intensities. Several variants exist: Banister's exponential version, Edwards' zone-based one, Lucia's. They differ in the weighting, not the idea.

The important difference from power is conceptual. Power measures what you did. Heart rate measures what it cost you. Three hundred watts on a hot afternoon at the end of a heavy week is not the same physiological event as three hundred watts fresh on a cool morning, and your heart rate knows the difference while your power meter does not.

It also works anywhere. Running, riding, rowing, skiing, hiking uphill with a pack: the same calculation produces numbers in the same units across all of them.

The weaknesses are real and worth stating:

• Cardiac drift. Heart rate climbs during long sessions at constant effort, so very long efforts can score higher than they truly cost.
• It is sensitive to things that aren't training. Heat, dehydration, caffeine, illness, poor sleep and altitude all shift it.
• It under-reads short, maximal efforts. A thirty-second all-out effort is over before heart rate has finished responding. The work was enormous; the number is small.
• It barely sees lifting at all. This is the big one for hybrid athletes, and it is the same blind spot that makes sled pushes and carries look like recovery.

Session RPE: the only one that covers everything

Carl Foster's session RPE is the simplest of the three: rate the whole session from 1 to 10 once it's done, multiply by duration in minutes.

It is subjective and it is coarse — one number for an entire session cannot distinguish a steady hour from an easy hour with a brutal ten minutes inside it, and it leans on your judgement on exactly the days your judgement is least reliable.

But it has one property neither of the others can match: it works for absolutely everything. Lifting, cardio, a climbing session, a football match. It is the only method on this list that gives you one currency for your entire week, which is why it's the method I'd recommend to anyone doing this by hand. It also validates reasonably well against heart-rate and power-based measures in the literature, which is more than you might expect from a number you make up at the end of a session.

Choosing

Honestly, and without pretending there's one winner:

• Only cycling, with a power meter — use power. It is more accurate and there's no reason to give that up.
• Several cardio sports — heart rate. It's the only method that produces one comparable unit across all of them.
• Lifting in the mix — none of the cardio methods covers it. You need tonnage or RPE for that half of your week.
• You want one simple thing covering literally everything — session RPE, in a spreadsheet.

The question people skip

Almost every comparison of these methods argues about accuracy, which quietly assumes you're measuring one sport. If you run, ride and lift, you have a different problem, and it's a worse one.

You cannot add TSS from your bike to TRIMP from your runs to a tonnage figure from the gym. They are three unrelated scales. You can have a beautifully precise number for one third of your training and no honest way to combine it with the rest — which leaves you back at the thing that started all this, several charts each describing part of a week and none describing the week.

Put plainly: a slightly noisier number that covers your whole week beats a very precise number that covers a third of it. Once you accept that, the ranking above changes shape. The question stops being "which method is most accurate" and becomes "which method can I apply consistently to everything I do".

That's also why the ramp rate you calculate is only as trustworthy as the coverage underneath it. A ratio computed from a load figure that omits two of your four training modes is arithmetic, not information.

If you go the heart rate route, calibrate properly

TRIMP is only as good as the zones underneath it, and the age-based formula is a rough population average that can be off by a considerable margin for any individual. Use a threshold test, a recent hard race, or a properly executed field test, and update it once or twice a year. This is the single highest-return thing you can do for the quality of a heart-rate-based number, and most people never do it.

Disclosure

I built LOAD, which uses heart-rate TRIMP for cardio and tonnage for lifting, fused into one number. That is a deliberate trade and worth naming: a cyclist with a power meter gives up some precision on the bike compared to a power-based score. What they get back is a figure that also contains their gym work, their runs and everything else, on one scale.

If you only ride, that trade is a bad deal and you should use power. If your week has a barbell in it, I think it's the right one. It's free for individual athletes, and connecting Strava backfills 12 weeks so the picture is real from the start.

The methodology and citations are public at loadtraining.app/faq.