The Validity and Reliability of a Global Navigation Satellite System in Canoe Slalom

This study investigates the usefulness of a 10 Hz GPS device for tracking scalar performance in canoe slalom through assessing the validity of automated-informed-aerial video tracking (30 fps and 10 fps) and GPS capability in relation to a known track. Additionally, a real-world (canoe-slalom). A comparison between manual-aerial video tracking (10 fps) and the 10 Hz GPS was performed. All three methods of tracking used during the dry-land test (30 fps or 10 fps video and GPS) reported significantly lower distances (-3.2, -5.1 and -8.5%, p < 0.0001) but were deemed useful based on sample rate and body positioning difference. Intra-method reliability was good (CV = 2.5-2.6%) but requires visual inspection for dataset errors. Informed-colour filtered automated tracking on-water was not possible, but manual tracking provided fewer dataset errors than dry-land automated tracking. GPS significantly (p < 0.0001) under reports distance travelled at key moments during real-world slalom with a bias +/- SD of 2.26 +/- 2.07 m compared to 10 fps manual-aerial video tracking. The aerial video combined with manual tracking proved most suitable for tracking canoe slalom athlete trajectory in a real-world setting but needs to be automated into an application-based package to make it useable for coaches. GPS, as presented, provides insight but does not accurately quantify movements critical in determining the performance of canoe slalom.