PURPOSE: There is little published data to guide coaches and sport scientists about important variables for measuring game and training load in sports outside of soccer and rugby. The purpose of this study was to statistically evaluate the relationship of internal and external load metrics in women’s collegiate lacrosse games. METHODS: Twelve Division I collegiate female lacrosse players wore a heart rate (HR) monitor and global positioning system (GPS) during 17 intercollegiate games. Seven measures determined training load: two internal measures [mean HR and training impulse (TRIMP)] and five external measures [total distance, high-intensity distance (HID), distance rate, accelerations, and decelerations]. The training load measures were analyzed for the whole game and by first and second halves. Principal component analysis (PCA) was used to determine which internal and external load variables were most associated with each portion of the game. A paired samples t-test was used to compare differences in first and second half metrics. RESULTS: The whole game and each half extracted only one principal component. For the whole game, HID, decelerations, accelerations, TRIMP, and total distance explained 58% of the variance (p < .001). The same metrics explained 55% of the variance for the first half (p < .001). For the second half, the same metrics with the addition of distance rate explained 57% of the variance (p < .001). Interestingly, the distance rate measure was only significant in the second half, although total distance contributed the most to the component, as it did in the first half and whole game data. The paired samples t-test showed differences between first and second halves for HID (p < .001), accelerations (p < .001), decelerations (p < .001). In all cases, there was greater distance and more intense efforts in the first half than the second. CONCLUSIONS: These results show that a combination of internal and external load measures should be used to determine load during games. The loaded metrics should be compared to a complimentary analysis for drills to ensure that training load metrics are similar. These data also support the concept of reduced high-intensity performance in the second half. This information should be used to bolster appropriate training methods to improve second half fitness.