Purpose: With knowledge of an individual’s critical power and W′, the SKIBA 2 model provides a framework with which to track W′ balance during intermittent high-intensity work bouts. There are fears that the time constant controlling the recovery rate of W′ (τW) may require refinement to enable effective use in an elite population. Methods: Four elite endurance cyclists completed an array of intermittent exercise protocols to volitional exhaustion. Each protocol lasted approximately 3.5–6 min and featured a range of recovery intensities, set in relation to the athlete’s critical power (DCP). Using the framework of the SKIBA 2 model, the τW values were modified for each protocol to achieve an accurate W′ at volitional exhaustion. Modified τW values were compared with equivalent SKIBA 2 τW values to assess the difference in recovery rates for this population. Plotting modified τW values against DCP showed the adjusted relationship between work rate and recovery rate. Results: Comparing modified τW values against the SKIBA 2 τW values showed a negative bias of 112 (46) s (mean ± 95% confidence limits), suggesting that athletes recovered W′ faster than predicted by SKIBA 2 (P = .0001). The modified τWDCP relationship was best described by a power function: τW = 2287.2 × DCP–0.688 (R2 = .433). Conclusions: The current SKIBA 2 model is not appropriate for use in elite cyclists, as it underpredicts the recovery rate of W′. The modified τW equation presented will require validation but appears more appropriate for high-performance athletes. Individual τW relationships may be necessary to maximize the model’s validity.