Women's lacrosse has grown more than 2-fold in the past 25 years, with approximately 6,000 female athletes participating in this sport at the intercollegiate level (11). Although the popularity of lacrosse is growing at great rates, the understanding of the physiological requirements of this sport and the physical characteristics of the athletes participating in intercollegiate competition is quite limited. Success in lacrosse has been suggested to be dependent on skill, speed, agility, strength, flexibility, and both aerobic and anaerobic capacities (14,17). These recommendations though have been based primarily on empirical evidence. Video analysis of work/rest intervals in National Collegiate Athletic Association (NCAA) Division I Men's Lacrosse reported that performance relied primarily on anaerobic metabolism (12), but no direct metabolic measurements were calculated. Subsequent research on male club team lacrosse players and NCAA Division I female lacrosse players has indicated that aerobic capacity levels in these athletes are higher than the 90th percentile in age-matched individuals but are similar to values seen in college basketball, team handball, and ice hockey athletes but less than that seen in soccer players (9,15,16). Similarly, anaerobic power outputs have also been shown to exceed the 90th percentile for age-matched individuals, but much lower than that seen for other anaerobic athletes (i.e., football and basketball players) (9,15).
There are 4 primary positions in lacrosse: attack, midfield, defense, and goalie. Each of these positions has overlapping responsibilities; however, there are also distinct roles for each position. The attacker's responsibility is to score goals and play primarily in the opponents end. Midfielders cover the entire field, playing both offense and defense, while defenders primarily remain on the defensive side of the field and protecting the goal. Goalies are responsible for defending the goal. Although midfielders cover the greatest distance during competition, the work to rest ratios were similar between the positions (12). Performance characteristics of these positions also appear to be similar. Both Steinhagen et al. (15) and Vescovi et al. (16) reported no significant differences between positions in a variety of performance-related variables (e.g., agility, endurance, power, speed, and vertical jump height) in male college-club lacrosse athletes and female NCAA Division I lacrosse players, respectively. To date, only male subjects participating in club lacrosse have been examined during a season. The importance of examining athletes during the competitive season is necessary to get a clear understanding of the physical characteristics that help determine team success. Therefore, the purpose of this study was to examine an elite team (two-time defending national champions) of NCAA Division III female lacrosse players during their competitive season and specifically examine performance differences between starters and nonstarters and to provide additional insight on physical performance characteristics between attackers, midfielders, and defenders. This information will assist strength and conditioning professionals in designing optimal training program and assisting lacrosse athletes with developing sport-specific training goals.
Experimental Approach to the Problem
All subjects were either returning members or freshman players on a nationally ranked NCAA Division III women's lacrosse team and were preparing to defend their national champion status for a third consecutive year. Subjects were examined prior to the onset of the regular season (during preseason training). Testing occurred in both laboratory and field settings. Subjects participated in a total of 5 testing sessions, separated by at least 72 hours. All testing sessions were supervised by certified strength and conditioning specialists.
Twenty-two female intercollegiate athletes (mean ± SD: age, 19.2 ± 1.0 years; height, 163.5 ± 5.1 cm; body mass, 61.1 ± 5.9 kg) of a NCAA Division III lacrosse team were examined during the 2007 preseason training period. These athletes were returning two-time national champions from the 2005 and 2006 seasons. The athletes were not involved in any supervised team-directed in-season or off-season resistance training program. All subjects provided their informed consent as part of their sport requirements consistent with the institution's policies of our Institutional Review Board for use of human subjects in research.
Subjects performed both upper (1 repetition maximum [1RM] bench press)-and lower (1RM squat)-body maximal strength tests. Anaerobic power was assessed with both field (vertical jump) and laboratory measures (Wingate anaerobic power test [WAnT] and a 30-second sprint test using a nonmotorized treadmill test). Additional laboratory measures included assessment of maximal aerobic capacity. Both speed and agility were assessed with standardized field tests (40-yd sprint T-drill and pro-agility test). Test-retest reliabilities for all assessments were R >0.90.
Subjects participated in a total of 5 separate testing sessions, separated by at least 72 hours. All testing sessions were supervised by certified strength and conditioning specialists. Anthropometric (height and body mass) and strength measures were performed initially. All strength and anthropometric measures were performed in the Human Performance Laboratory (HPL). During the second testing session, subjects returned to the HPL for vertical jump and WAnT measures. Subjects again returned to the HPL for the third testing session and performed the 30-second anaerobic sprint test. Maximal aerobic capacity was assessed during the fourth testing session in the HPL. The final testing session assessed speed and agility performance and occurred on the stadium's Astroturf field. The Figure illustrates the performance testing protocol.
Maximal Strength Testing
During each testing session, subjects performed a 1RM strength test on the bench press and squat exercises to measure upper- and lower-body strength, respectively. The 1RM tests were conducted as previously described (9). Each subject performed a warm-up set using a resistance that was approximately 40-60% of their perceived maximum and then performed 3 to 4 subsequent trials to determine the 1RM. A 3- to 5-minute rest period was provided between each trial. The squat exercise required the subject to place an Olympic bar across the trapezius muscle at a self-chosen location. Each subject descended to the parallel position, which was attained when the greater trochanter of the femur reached the same level as the knee. The subject then ascended until full knee extension. Bench press testing was performed in the standard supine position; the subject lowered a strength training bar to sternum level and then pressed the weight until her arms were fully extended. Trials not meeting the range of motion criteria were discarded.
Anaerobic Power Measures
To quantify anaerobic power performance, all subjects performed the WAnT (Lode Excalibur, Groningen, The Netherlands). After a warm-up period of 5 minutes of pedaling at 60 rpm interspersed with 4 all-out sprints lasting 5 seconds, the subjects pedaled for 30 seconds at maximal speed against a constant force (1.0 N m·kg−1). Peak power, mean power, and fatigue rate were determined. Peak power was defined as the highest mechanical power output elicited during the test. Mean power was defined as the average mechanical power during the 30-second test. Fatigue rate was determined by dividing the highest power output from the lowest power output.
To quantify anaerobic sprint power performance, all subjects also performed an anaerobic power test on a nonmotorized treadmill (Woodway, USA, Waukesha, WI). The treadmill uses a user-driven loading applied via a magnetic braking system interfaced with a computer. Four load cells were located underneath the running surface to record force and power data. Subjects performed one 30-second sprint against a resistance of 20% of body mass. Peak power, mean power, total work, and a fatigue rate were calculated for each sprint.
Countermovement vertical jump height was measured using a Vertec (Sports Imports, Columbus, OH). Prior to testing, each athlete's standing vertical reach height was determined. Vertical jump height was calculated by subtracting the standing reach height from the jump height. Subjects performed 3 attempts. The highest vertical jump height achieved was recorded.
All o2peak tests were performed on a motor-driven treadmill (TrackMaster 210; JAS MFG, Co., Carrollton, TX) using the test protocol described by Åstrand et al. (1). All subjects were familiarized with the equipment and test protocol. Following a 4- to 5-minute warm-up, which consisted of the subject beginning in a walk and ending in a run, the subject began the exercise protocol. Subjects were required to run between 5 and 8 mph, but the precise pace was self-chosen. The initial stage of the protocol was performed at 0% grade and lasted for 3 minutes. Each subsequent stage was performed at the same velocity for 2 minutes. Increases in workload were achieved with a 2% increase in grade for each stage. Subjects were instructed not to use handrails until the completion of the test.
For each test, the subject was verbally encouraged to continue exercise until volitional exhaustion. Both o2 and heart rate (HR) were recorded during the last 15 seconds of every minute throughout the test. o2max and HRmax were determined by averaging the 2 consecutive highest measures of each variable. To ensure that a true o2max had been attained, at least 2 of the following 3 criteria were met: an increase in o2 of less than 100 ml·min−1 despite an increase in work rate, an HR within ± 5 b·min−1 of age-predicted maximum, or a resting expiratory rate (RER) greater than 1.10.
o2 was determined using a two-way T-shaped breathing valve (Hans Rudolph 2700; Hans Rudolph, Inc., Kansas City, MO) and an open-circuit respiratory-metabolic system (Metabolic Measurement Cart 2900; SensorMedics, Inc., Yorba Linda, CA). Heart rate was measured using a wireless HR monitor (Pacer, Polar CIC, Inc., Port Washington, NY).
Speed and Agility Assessments
Speed was determined by a timed 40-yd (37-m) sprint. Sprint times were determined using handheld stopwatches and performed on an Astroturf field. Timing began on the subject's movement out of a three-point stance. The best of 3 attempts was recorded as the subject's best time. The same investigator timed all 40-yd sprint tests.
Agility was determined by both a T-test and a pro-agility test on an Astroturf field. The protocols were conducted as previously described (9). The T-test required the subject to sprint in a straight line from a two-point stance to a cone 9 m away. The subject touched the cone and then side shuffled to either her left or right without crossing her feet to another cone 4.5 m away. She touched that cone and then side shuffled to the opposite side to a third cone that was 9 m away. The subject then side shuffled back to the middle cone and back pedaled to the starting position. The pro-agility test was performed using the markings on the football field. The subject straddled the 5 yd line and sprinted to the goal line (4.5 m away) and touched the line. She then changed direction and sprinted to the 10 yd line (9 m away), touched the line with the same hand used to touch the goal line, reversed direction, and returned to the starting point. Subjects were instructed to sprint through the 5 yd line. The timer began upon the subject's initial movement and stopped as the athlete crossed the 5 yd line. The same investigator conducted all agility tests. Each subject performed 3 maximal attempts for each drill, and the fastest time for each drill was recorded.
Statistical comparisons between starters and nonstarters were accomplished using an independent t-test. Comparisons between positions (attackers, defenders, and midfielders) were accomplished with a 1-way analysis of variance. In the event of a significant F ratio, least significant differences (LSD) post hoc tests were used for pairwise comparisons. A criterion alpha level of p ≤ 0.05 was used to determine statistical significance. All data are reported as mean ± SD.
Anthropometric, strength, power, speed, and agility comparisons between starters and nonstarters are depicted in Table 1. No significant differences were observed between starters and nonstarters in any of these variables.
Comparisons between positions are shown in Table 2. Anthropometric analysis revealed that attackers were 15.7% (p < 0.05) heavier than midfielders. However, no other statistical differences in height or body mass were seen. Strength comparisons revealed a significant difference (10.3%) between defenders and midfielders in 1RM squat, but no other differences in 1RM squat were noted between positions, and 1RM bench press performance was similar between all positions. Power analyses indicated that attackers were more powerful in the WAnT (both peak and mean power) than both defenders (19.6 and 13.4%, respectively) and midfielders (21.2 and 13.4%, respectively). No other differences were seen between the groups in power performance. In addition, no significant differences were noted between the groups in any speed or agility measure.
Results of this study indicated that measurement of physical performance was unable to differentiate starters from nonstarters in an elite NCAA Division III female lacrosse team. This is consistent with a previous study examining men's club lacrosse that was also unable to differentiate skill level (i.e., first and second team) based on physical performance (15). However, in the study by Steinhagen et al. (15), no field tests were used and only measures of aerobic capacity and anaerobic power were examined. This present study, using both laboratory and sport-specific field assessments, was still unable to differentiate starters from nonstarters based on any physical or performance attribute.
Previous studies on anaerobic athletes have shown that physical ability (e.g., strength, speed, and power) can be an effective predictor for success in both basketball (10) and football (2,3,5,7). Many of these studies have compared physical performance characteristics between different divisions of play or in game outcomes (e.g., national rankings). Hoffman et al. (10), in a 4-year study of an elite NCAA Division I college basketball team using playing time as the dependent variable, demonstrated that lower-body strength, speed, and power contribute to greater playing time, while Black and Roundy (3) showed that strength, speed, and power measures can separate starters from nonstarters in NCAA Division I football. However, Black and Roundy (3) combined results of 11 teams. In studies examining a single team, the ability to differentiate starters from nonstarters based on physical attributes appears to be limited. Schmidt (13) showed that strength and power performance was able to differentiate starters from nonstarters in Division III football players. However, the total number of players on most athletic teams, outside American football, likely does not provide the statistical power necessary to see statistical differences in physical characteristics between starters and nonstarters. This is especially relevant, in that coaches will recruit a certain type of player who fits to the needs and coaching style being employed by that specific team.
Another aspect of examining predictors for success in sport is the importance of the athlete's sport-specific skill. Although strength, power, and speed are all desirable factors, the primary issue in sport is related to the athlete's ability to play that sport. Hoffman et al. (10) showed that the coach's perception of the athlete's sport-specific skill is the most critical component for determining playing time. Once this variable is factored out, then the relative importance of physical factors relating to athleticism can be assessed. In a sport such as lacrosse, in which skill plays such an important role, the relative importance of these physical factors may be secondary to the ability of the athlete to play the game, especially in a group of athletes who appear to be quite homogenous in regard to their physical performance characteristics.
Comparisons between positions revealed several physical characteristics that were significantly different. This is in contrast with the recent study by Vescovi et al. (16) who reported no differences between positions except for height in NCAA Division I female lacrosse players. Results of this study indicated that attackers were significantly heavier than midfielders and that they were more powerful than both defenders and midfielders. The enhanced power for attackers appears to provide a significant advantage for these athletes whose primary responsibility is to score. Interestingly, midfielders had less lower-body strength than both defenders (10.3%, p < 0.05) and attackers (7.6%, p > 0.05). This may reflect the difference between positions in the total distance run during a game. Midfielders play in both the offensive and the defensive zones, whereas attackers and defenders generally play on one side of the field. Because of a greater coverage of the field, this position has been considered the most arduous (12,14). It is likely that the greater running required of these athletes during competition has a cumulative effect on decreasing lower-body strength performance. This is supportive by several studies examining the suppressive effect that concurrent training has been shown to have on strength performance (4,8).
In comparison to other intercollegiate female athletes, the results of this study indicated that NCAA Division III lacrosse players were weaker and slower than Division III female soccer players (9). However, these athletes were faster than Division I female volleyball players in the 40-yd sprint but not as quick as the volleyball players in the T-drill (6).
Although physical performance characteristics were not different between starters and nonstarters, results do suggest that attackers are heavier and more powerful than the other positions. Midfielders appear to be weaker in lower-body strength relative to the other positions in lacrosse. This may be related to the greater distance covered during competition relative to the other positions. The findings of this study do provide coaches and athletes an idea as to how training can be planned differently in order to maximize performance of each respective position. For example, training that emphasizes power development may be more preferable for attackers or defenders.
1. Åstrand, PO, Rodahl, K, Dahl, HA, and Strømme, SB. Textbook of Work Physiology. Physiological Bases of Exercise
(4th ed). Champaign, IL: Human Kinetics, 2003. pp. 277-279.
2. Berg K, Latin, RW, and Baechle, T. Physical and performance characteristics of NCAA Division I football players. Res Quart
61: 395-401, 1990.
3. Black, W and Roundy, E. Comparisons of size, strength
in NCAA Division I-A football players. J Strength Cond Res
8: 80-85, 1994.
4. Dolezal, BA and Potteiger, JA. Concurrent resistance and endurance training influence basal metabolic rate in nondieting individuals. J Appl Phys
85: 695-700, 1998.
5. Fry, AC and Kraemer, WJ. Physical performance characteristics of American collegiate football players. J Appl Sport Sci Res
5: 126-138, 1991.
6. Fry, AC, Kraemer, WJ, Weseman, CA, Conroy, BP, Gordon, SE, Hoffman, JR, and Maresh, CM. The effects of an off-season strength
and conditioning program on starters and non-starters in women's intercollegiate volleyball. J Appl Sport Sci Res
5: 174-181, 1991.
7. Garstecki, MA, Latin, RW, and Cuppett, MM. Comparison of selected physical fitness and performance variables between NCAA Division I and II football players. J Strength Cond Res
18: 292-297, 2004.
8. Hennessy, LC and Watson, AWS. The interference effects of training for strength
and endurance simultaneously. J Strength Cond Res
8: 12-19, 1994.
9. Hoffman, JR. Norms for Fitness, Performance, and Health
. Champaign, IL: Human Kinetics, 2006. pp. 27-39.
10. Hoffman, JR, Tennenbaum, G, Maresh, CM, and Kraemer, WJ. Relationship between athletic performance tests and playing time in elite college basketball players. J Strength Cond Res
10: 67-71, 1996.
11. National Collegiate Athletic Association. NCAA Sports Sponsorship and Participation Rates. 1981-82-2005-2006
. Indianapolis, IN: National Collegiate Athletic Association, 2007.
12. Plisk, SS. Regression analysis of NCAA Division I final four men's lacrosse competition. J Strength Cond Res
8: 28-42, 1994.
13. Schmidt, WD. Strength
and physiological characteristics of NCAA Division III American football players. J Strength Cond Res
13: 210-213, 1999.
14. Scott, B. Lacrosse: Techniques & Tradition
. Baltimore, MD: John Hopkins University Press, 1976.
15. Steinhagen, MR, Meyers, MC, Erickson, HH, Noble, L, and Richardson, MT. Physiological profile of college club-sport lacrosse athletes. J Strength Cond Res
12: 226-231, 1998.
16. Vescovi, JD, Brown, TD, and Murray, TM. Descriptive characteristics of NCAA Division I women lacrosse players. J Sci Med Sport
10: 334-340, 2007.
17. Walker, J. How to Play Winning Lacrosse
. New York, NY: Cornerstone Library Publication, 1980.