Estimating the Cost of Lifting Weights

Cardio is straightforward to price: steady effort, a known MET value (a multiple of what you burn lying still), measured duration. Resistance training isn't: it's short bursts against varying loads with rest in between, and a single MET number for "weight training" ignores whether you squatted 140kg or curled 8kg. Kriterion estimates it from the physics instead: the mass moved, the distance it travelled, and the fact that muscle converts only around a fifth of the metabolic energy it consumes into actual mechanical work. Two things get added on top of the lift itself.

Lowering the weight costs real energy too, though less than raising it. The literature puts eccentric work, meaning the lowering half of a rep, at roughly a quarter to a half of the cost of the concentric or lifting half at the same load. And oxygen uptake stays elevated after you rack the bar; Scott's work found recovery uptake following a set can exceed uptake during the set. Neither is huge on its own, and together they're the difference between an estimate that's roughly right and one that's a third too low.