Background: Athletes strive for optimal performance in sport, and this is no different for high performance netball umpires. Performance of referees/umpires in elite sport (i.e. soccer, rugby union, rugby league, field hockey and basketball) has been widely investigated to assess match-play demands, and in some cases fitness testing performance has been compared to match day performance measures, including speed, distance covered and physiological demands. To our knowledge, no data has been collected on Netball umpires over the past 20 years and findings within other high performance sports indicate that the current fitness testing utilized for high performance netball umpires may also not correlate to actual match day demands. Purpose: The aim of this study was to assess the match-play physical, physiological and movement demands for three levels of elite umpires during 2012 New Zealand (NZ) netball season and to ascertain if the current Netball New Zealand (NNZ) fitness tests correlated to these match-play demands. Methods: There were two stages of the data collection for this study, 1) match day observations and 2) the completion of the NNZ battery of fitness tests. A sample of convenience was recruited from the NNZ High Performance Netball Umpire Squads, and they were ranked from 1 to 22 (with 1 as the best) and categorised into three performance level; Group One (N = 9, ranking 1 - 9) Group Two (N = 6, ranking 10 - 15) and Group Three (N = 7, ranking 16 - 22). Catapult mini-max S4 Global Positioning System (Catapult Sports, Victoria, Australia) units and Polar Team2 heart rate monitors (Polar Electro, Oy, Kempele, Finland) were worn by participating umpires (N = 22) for the duration of each match and each fitness test. All matches and fitness tests were filmed using a Canon LEGRIA HV40 video camera, and the footage was uploaded to Sportscode Elite Version 10 (Sportstec, NSW, Australia) for analysis. The fitness tests used were the current NNZ protocols for the Yo Yo Intermittent Recovery Level 1 Test, the anaerobic Octorepeater Test and the 5, 10, 20m Speed Test. Results: Umpires travelled a distance of 3.84km, worked at an average intensity of 80% of Heart Rate peak and spent the most time (35%) in HR Zone 3 (75 – 85% of HRpeak). Work-rate (Player Load and Heart Rate), frequency and percentage of time performing work movements followed a trend of decreasing across the duration of a match for all umpires. Umpires spent 49% of a match standing, 23% of time working and 10% of time sprinting. Group One umpires overall worked at higher intensities (HR and movement demands) than both Group Two and Three and also chose to use movements that orientated their bodies and heads more towards the ball. Correlations were found between the Yo Yo fitness test distance and the percentage of time jogging during match-play (r = .51, p = .03), Mean HR in both the Yo Yo (r = .59, p =.01) and Octorepeater (r = .50, p = .03) was correlated with the percentage of time working during match-play. The frequency of turns in the Octorepeater correlated with the amount of turns during match play (r = .48, p = .05). Conclusion: The correlations between the amount of jogging during match play and during the Yo Yo Test and the frequency of turning during match-play and the Octorepeater both suggest some validity of current testing protocols. However as all other correlations were poor the overall results indicate that the current fitness tests are not a valid indicator of match-performance though may be used as normative tests for base line fitness.