It is unclear whether physiological measures monitored in an incremental test during a training season provide useful diagnostic information about changes in distance running performance. PURPOSE: To quantify the relationship between changes in physiological measures and performance (peak running speed) over a winter training season. METHODS: Well-trained male distance runners (n=34, age 35.2 ± 1.1 y, VO2max 64.5 ± 6.2 ml.kg−1.min−1, training for 9.5 ± 0.9 h.wk−1, mean ± SD) completed four incremental treadmill tests over 17 wk. The test provided values of peak running speed, VO2max, running economy, and 4 mM lactate threshold (as speed and fraction of VO2max). The physiological measures were included in simple and multiple linear regression mixed models to quantify the effect of changes in these measures on changes in peak speed. RESULTS: The within-subject typical variation in peak speed from test to test was 1.9%, whereas those for physiological measures were: VO2max, 2.5%; economy, 2.6%; lactate threshold speed, 6.2%; lactate threshold fraction, 6.5%; and body mass, 1.6%. In simple models these typical changes in physiological variables predicted the following changes in performance: VO2max, 1.2% (90% confidence limits ±0.2%); economy, 0.5% (±0.3%); lactate threshold speed, 0.5% (±0.3%); lactate threshold fraction,−0.3% (±0.3%); and body mass,−0.4% (±0.3%). After adjustment for other predictors in a multiple linear regression, the effect of VO2max on peak speed was not diminished, but economy became a marginal predictor, while lactate threshold and body mass became ineffective. CONCLUSION: Change in VO2max in an incremental test during a running season is a good physiological predictor of change in peak running speed. Changes in running economy, lactate threshold and body mass provided little additional information about changes in performance.