Original W'-balance models do not transfer to high-intensity interval training in trained adult runners: Proposal of a single-coefficient linear alternative.

This study evaluated the off-the-shelf predictive accuracy of four original W'-balance recovery models (Morton and Billat linear, W'BAL-INT, W'BAL-ODE and W'BAL-CAEN) and proposed a single-coefficient modified linear formulation (Linear_mod) for predicting high-intensity interval training (HIIT) performance with passive recovery in trained runners. Twenty-seven trained runners completed a 9/3-min running critical power (CP) test and two open-ended HIIT sessions to volitional exhaustion at 108% CP with a 2:1 work: rest ratio. Work intervals were individualized so that energy expenditure above CP equalled 40% (HIIT40%) or 60% (HIIT60%) of W'. The original models were applied using their published parameters. All substantially over-predicted the number of intervals completed (pooled MAE 26.4-72.4 intervals; bias +20.1 to +72.4). Linear_mod with its default recovery coefficient (R = 1) achieved a MAE of 4.85 intervals. Optimizing R at the population level (R = 1.63) reduced MAE to 2.85 intervals, while athlete-specific optimisation (R = 1.54 ± 0.29) reduced MAE to 0.55 intervals. Individual R values were stable between protocols (within-subject CV = 9.9%). Original W'-balance models did not accurately predict running HIIT performance, whereas Linear_mod provided high predictive accuracy using a single athlete-specific recovery coefficient.