Race and Final-600 m Speed Response to Distance and Track Condition in Thoroughbred Flat Racing

The temporal and spatial distribution of speed varies with race distance and track condition, influencing where mechanical demand peaks within a race. This study aimed to describe variation in race-day speed with race distance and track condition. Race-day data from 18 flat racing seasons on turf and synthetic surfaces (2007/8-2024/25, n = 200,601 starts in 3389 races) were used to examine initial speed (from barriers until the final 600 m), final-600 m speed, race distance (m), turf track condition score (TCS, 1-10), horse rating and carried weight (kg) using a general linear model. The pattern of racing was consistent for races ≥ 1000 m. Final-600 m speed at all distances was 0.41 (95% CI 0.41-0.42) m/s faster than initial speed (p < 0.001). Both speeds decreased with increasing race distance, but this decrease was smaller for final-600 m speed (β = -0.15 [95% CI -0.16--0.15] m/s per 200 m) than for initial speed (β = -0.17 [95% CI -0.18--0.17] m/s per 200 m, p < 0.001). Speeds had a negative quadratic relationship with TCS (differing in magnitude and intercept, p < 0.001), with a rapid decline at TCS 9. Higher horse rating increased speed, while greater carried weight reduced speed (both p < 0.001). The higher race final-600 m speeds were less affected by distance, indicating that mechanical demand is higher late in the race and load accumulation is not uniform.