This article was downloaded by: [McMaster University] On: 23 December 2014, At: 02:59 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Sports Biomechanics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rspb20
Is starting with the feet out of the water faster in backstroke swimming? ab
b
a
Cecilia Nguyen , Elizabeth J. Bradshaw , David Pease & Cameron Wilson
b
a
Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Canberra, Australia b
School of Exercise Science, Australian Catholic University, Melbourne, Australia Published online: 24 Feb 2014.
Click for updates To cite this article: Cecilia Nguyen, Elizabeth J. Bradshaw, David Pease & Cameron Wilson (2014) Is starting with the feet out of the water faster in backstroke swimming?, Sports Biomechanics, 13:2, 154-165, DOI: 10.1080/14763141.2014.885072 To link to this article: http://dx.doi.org/10.1080/14763141.2014.885072
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [McMaster University] at 02:59 23 December 2014
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
Sports Biomechanics, 2014 Vol. 13, No. 2, 154–165, http://dx.doi.org/10.1080/14763141.2014.885072
Is starting with the feet out of the water faster in backstroke swimming? CECILIA NGUYEN1,2, ELIZABETH J. BRADSHAW2, DAVID PEASE1, & CAMERON WILSON2 Downloaded by [McMaster University] at 02:59 23 December 2014
1
Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Canberra, Australia, and 2School of Exercise Science, Australian Catholic University, Melbourne, Australia
(Received 16 December 2012; accepted 14 January 2014)
Abstract This study aimed to determine if starting with the feet above the water (FAW) in male backstroke swimming resulted in faster start times (15-m time) than when the feet were underwater (FUW). It was hypothesised that setting higher on the wall would generate increased horizontal force and velocity, resulting in quicker starts. Twelve high-level male backstrokers performed three trials of the FAW and FUW techniques. A biomechanical swimming testing system comprising one force plate (1,000 Hz), four lateral-view (100 Hz), and five overhead (50 Hz) video cameras captured the swimmers’ performance. Data for each participant’s fastest trial for each technique were collated, grouped, and statistically analysed. Analysis included Wilcoxon, Spearman Rho correlation, and regression analysis. Wilcoxon results revealed a significantly faster start time for the FAW technique ( p , 0.01). Peak horizontal force was significantly smaller for FAW ( p ¼ 0.02), while take-off horizontal velocity was significantly greater ( p ¼ 0.01). Regression analysis indicated take-off horizontal velocity to be a good predictor of start time for both techniques, and the horizontal displacement of the centre of mass for the FAW start.
Keywords: Biomechanics, force, velocity, start
Introduction The start phase is defined by many as the time it takes the middle of the swimmers head to reach the 15-m marker and is considered a critical phase of a swimming race (e.g. Cossor & Mason, 2001; Jorgic´ et al., 2010; Mason & Cosser, 2000; Ruschel, Araujo, Pereira, & Roesler, 2007; Vantorre, Seifert, Fernandes, Vilas-Boas, & Chollet, 2010). The start represents between 0.8% and 26.1% of the final race time with the influence being greater for shorter events such as the 100 m (Mason & Cosser, 2000). In World Championship and Olympic competitions, races have been won or lost by a fraction of a second. For example, at the 2012 Summer Olympics in London, American Nathan Adrian edged out Australian James Magnussen in the 100-m freestyle event by 0.01 s to win gold. Also at the 2008 Correspondence: Cecilia Nguyen, Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Leverrier Street, Bruce, ACT 2617, Australia, E-mail: [email protected] q 2014 Taylor & Francis
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
155
Olympics, German Britta Steffen won the 50-m freestyle gold by the same margin over American Dana Torres. Swimmers are continually developing their start technique with the goal of gaining a faster start time and therefore an early edge over their competitors. For freestyle, breaststroke, and butterfly, the ventral dive start is used. This is where the swimmer stands on top of the starting block and dives into the water once the trigger sounds. The backstroke start requires different skills to the ventral start as it begins with the swimmer already in the water, holding onto the starting block and diving backwards. In July 2005, the Federation Internationale de Natation (FINA, 2011) modified the backstroke start rule to allow competitors to set their feet above the water (FAW) line prior to the take-off signal of a race. Since this rule change, many competitive backstrokers have altered their starting technique to place their feet completely out of the water. With advances in knowledge and technology, biomechanical parameters of the swimming start have been broadly investigated to determine if any significantly affect start time. Research is abundant comparing the grab and track starts, which are the two most popular ventral starts (e.g. Blanksby, Nicholson, & Elliott, 2002; Breed & McElroy, 2000; Hohmann, Fehr, Kirsten, & Krueger, 2008). However, those studying the differences in backstroke start techniques are scarce, particularly since the 2005 rule change. Kruger, Hohmann, Kirsten, and Wick (2006) noted that there were definite kinematic and dynamic influences of the overwater (take-off and flight) phases of the backstroke start. Comparable to ventral start studies, strong correlations were identified between resultant take-off force and start time (Kruger et al., 2006). The higher the impulse or resultant force on the wall, the greater the take-off velocity and quicker time to 7.5 m (r ¼ 0.83, p , 0.01). In a comparison study of backstroke start techniques, de Jesus et al. (2011) proposed that shorter start times would be attributed to the mechanical characteristics at the wall set position. It was suggested that setting the feet high on the wall and above the water at the start may produce a larger horizontal force off the wall (de Jesus et al., 2011). This was because the centre of mass (COM) was higher; hence, the swimmer was not required to project their body as much vertically (de Jesus et al., 2010; Hooper, 1981). The purpose of this study was to compare the biomechanics of the FAW and feet underwater (FUW) starting techniques for backstroke to determine if starting with the FAW led to a faster start time. It was hypothesised that placing the feet high on the wall would produce faster start times (15-m time) due to greater generation of horizontal force and velocity from a higher COM in the set up position.
Methods Participants Twelve Australian male backstroke swimmers aged 20.5 (SD: 1.6 years) (M ^ SD height ¼ 187.0 ^ 6.0 cm, body mass ¼ 80.7 ^ 6.0 kg) were recruited for this study. All participants were healthy (no serious injury or illness in the last 6 months), able-bodied, and were, at the time of this study, racing in state and national competitions with backstroke as their main event. Only males were recruited for this study as very few female backstrokers use the FAW starting technique. The FINA point score system was used to quantify skill level. The minimum scores required for participation in this study were 770 FINA points in the 50-m backstroke event and 602 FINA points in the 100 m. These scores corresponded to the qualifying times of the 2012 Energy Australia Swimming Championships (Swimming Australia, 2012). Both the Australian Institute of Sport (AIS) and Australian Catholic University ethics committees approved the study before testing commenced.
156
C. Nguyen et al.
Downloaded by [McMaster University] at 02:59 23 December 2014
Procedure The participants attended one testing session of approximately 1 h at the AIS technology pool in Canberra. The height (cm) and body mass (kg) of the participants were measured at the start of the session. Each swimmer was then asked what starting technique they would normally use in a race (feet completely underwater, partially underwater, or their whole FAW) and this was recorded. The participants then completed a warm up that involved swimming 400 m of freestyle at their own pace. The testing required participants to perform both variants of the backstroke start. These were starting backstroke with the feet fully above the water (FAW) and then completely underwater (FUW). For the FAW technique, the feet were set so that the heels were just above the water line (Figure 1a). For the FUW, the toes were placed just below the water’s surface (Figure 1b). The technique variants were randomised with the order determined before testing commenced. The two backstroke techniques were verbally described as well as visually depicted by video recordings to each participant. The swimmers were provided with familiarisation time of each technique before data collection commenced. This involved participants performing a number of practice trials of both techniques on the testing wall. Verbal instruction and feedback were given during familiarisation to ensure that the starts were performed correctly. Each participant performed three trials of the FAW and FUW variants, with 2 min rest in between each trial. Participants wore their standard training suit during testing.
Data collection The biomechanical systems embedded in the AIS technology pool have previously been described by Mason, Mackintosh, and Pease (2012). Four high-speed digital gigabit Ethernet cameras (TMC6740GE, 100 Hz, Pulnixw, San Jose, CA, USA) recorded the swimmer’s performance from a lateral view for the whole 15 m distance (Mason et al., 2012). Camera one viewed the above water left lateral side at initial set up position, take-off, and water entry; cameras 2–4 recorded a continuous view of the underwater left lateral side. The feeds from all cameras were integrated so that one complete video recording of the trial was produced. An overhead video system captured timing information. This overhead video system utilized five overhead cameras (SCC-C4301, 50 Hz, Samsung, Suwon, South Korea) positioned normal to the lane in which the athletes were performing at key distances (5, 7.5, 10, 15, and 20 m).
Figure 1. (a) FAW starting technique and (b) FUW starting technique. The heels of the feet are above the water’s surface for FAW and the toes of the feet are below the water’s surface for FUW.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
157
From these cameras, it was possible to obtain the time when the centre of the athlete’s head passed the given distance. The cameras were genlocked to provide a synchronised image that ran through in-house developed time-coding computer software. When watching the playback of the trials, a fixed line on the screen at each segment was used as a distance marker. Using this system, the time it took for the middle of the athlete’s head to reach each distance marker was determined. The middle of the head was used as the reference point because it is also used as the marker for ‘breakout distance’ during a race. Breakout distance is measured during the start phase and after a turn. It is the distance between the starting or turning wall and the point at which the swimmer’s head breaks the surface of the water. One force plate (Z20313, 600 mm £ 900 mm £ 40 mm, 100 Hz, Kistler, Winterthur, Switzerland) was inserted into the starting end of the pool, flush with the wall to measure the three-dimensional forces when the swimmer pushed off after the starting signal. All force measures were transformed from analogue to digital signals via amplifiers (Type 5070 Multichannel Charge Amplifier, Kistler, Winterthur, Switzerland) and a converter board and then relayed via cables to a personal computer that ran customised software (Mason et al., 2012). An auditory buzzer signal, similar to what is used in competition, was used to simultaneously start the swimmer and the data collection systems. Data analysis The customised computer program enabled semi-automatic analysis of force platform data and video images collected from the video cameras (Mason et al., 2012). Video footage from the four high-speed cameras (100 Hz) were manually digitised by finding the centre and top of the swimmer’s head and approximate COM during the start; specifically in the initial set up position, at the point when the feet left the wall and when the head entered the water. This was estimated using the grid markings from the computer software and the swimmer’s left greater trochanter (left hip). The hip is the closest point to the COM and therefore provides an easily digitisable, body landmark during analysis (Getchell & Robertson, 1989; Ranavolo et al., 2008). The left hip was therefore used to estimate the COM position in this study. From finding these points, and using a 100-Hz video camera that captured a frame every hundredth of a second, the computer program was able to calculate the biomechanical variables investigated. Feet position on the wall at the start was the independent variable (the feet position was either FAW or FUW). The main dependent performance variable was start time (time from the start signal to 15 m). Other variables examined were wall time (time from start signal until time the swimmer’s feet left the wall); 5-, 7.5-, 10-m time (all to when the middle of the swimmer’s head reached the distances); peak vertical, horizontal, and resultant forces (peak forces over the whole take-off phase); take-off vertical and horizontal velocities (at the instant of take-off); entry velocity (calculated from horizontal and vertical take-off velocities); entry distance (wall to head water contact); and horizontal and vertical displacement of the hip at the start and hip angle of entry (calculated using take-off horizontal and vertical velocities and the water line as the horizontal). Dependent measures for each trial and swimmer were collated in an Excel spreadsheet (Microsoft, Redmond, WA, USA). The trial with the fastest start time for each swimmer for the two start techniques was identified. The data were then grouped and descriptive statistical analysis was performed using Statistical Packages for the Social Sciences software (version 19.0.0, IBM Corporation, New York, NY, USA). An a level of 0.05 was set as the significance level for all statistical tests. Normality of the grouped data was tested using the Shapiro – Wilk test. The data was normally distributed with the exception of age, entry
Downloaded by [McMaster University] at 02:59 23 December 2014
158
C. Nguyen et al.
velocity for the feet above condition, time to 5 m for both conditions, and peak vertical force for both conditions. Due to the inconsistency in the normality of the data-set and the small sample size, non-parametric statistical methods were subsequently employed unless otherwise stated. A Wilcoxon test was used to identify any statistical differences between the biomechanics of the two starting techniques. Spearman Rho (r) correlation analysis was used to determine if there were any significant relationships between time variables and velocity, force, or start hip position components of the backstroke start for both FUW and FAW. Significant correlations were identified if jrsj . 0.59, p , 0.05 (Zar, 1972). Following the correlation tests, a forward stepwise regression analysis was performed on the data. The two starting conditions were analysed separately, with start time as the dependent variable and variables that had significant correlations set as the independent variables. Using this method, one variable at a time was added to the regression model until the entered variable’s p value was higher than the p value cutoff ( p . 0.05). From this forward stepwise regression, the best predictors of start time for the FUW and FAW techniques were determined. Results Backstroke start performance On average, the FAW technique resulted in a 0.08-s (1.1%) faster starting performance than the FUW technique ( p , 0.01). Out of the 12 swimmers, 4 participants preferred the FAW technique, 3 preferred the FUW technique, and 5 participants normally adopted the feet partially underwater technique. When the participants’ results were divided into groups based on their technique preference, descriptive analysis showed that FAW start time was faster than the FUW start time regardless of the swimmers technique preference (Table I). Furthermore, an examination of the individual data revealed that 10 out of the 12 participants were able to perform a quicker backstroke start to the 15-m lane marker when using the FAW technique. The largest improvement observed was 0.24 s for participant 3 which represented a 2.3% faster start time. The smallest improvement observed was 0.03 s for participant 8, which represented a 0.4% faster start time. Both swimmers (participants 6 and 10) who had performance decrements with the FAW technique had a 0.02-s (participant 6 –0.2%, participant 10 – 0.3%) slower start time. Kinematic and kinetic differences between start techniques The results for the kinematic and kinetic measures of start performance are summarised in Table II. Small technical differences were identified in the swimmers start displacement (start hip position), and the vertical and horizontal forces developed ( p , 0.05). The swimmers had a higher set position (start vertical hip displacement) for FAW which resulted in the generation of higher vertical but less horizontal force. This slight variation between the FUW and FAW techniques had no effect on resultant force generated but did result in a significantly faster take-off horizontal velocity ( p ¼ 0.010) and a steeper dive entry ( p , 0.01) for FAW. The timing results showed that the FAW start lead to faster times at the 5- and 10-m lane markers. Biomechanical attributes of a faster start The results of the Spearman r correlation analysis between biomechanical measures of the FUW and FAW techniques are summarised in Table III. The generation of a high peak
Foot position in backstroke start
159
Table I. Best start times
Downloaded by [McMaster University] at 02:59 23 December 2014
Start time (s)
Participant
Technique preference
1 2 3 4 5 6 7 8 9 10 11 12 Technique preference group Overall group
Group ranking
FUW
FAW
Difference FAW– FUW
Above Partial Above Above Above Under Under Partial Partial Partial Partial Under
8 3 2 4 1 11 10 12 5 7 6 9
7.72 7.2 7.02 7.22 6.88 8.42 7.88 8.61 7.28 7.6 7.56 7.74
7.54 7.06 6.78 7.06 6.78 8.44 7.78 8.58 7.14 7.62 7.34 7.7
-0.18 -0.14 -0.24 -0.16 -0.1 -0.02 -0.1 -0.03 -0.14 -0.02 -0.22 -0.04
Under Partial Above
Mean (SD) Mean (SD) Mean (SD)
8.01 (0.36) 7.65 (0.56) 7.21 (0.37)
7.97 (0.41) 7.55 (0.55) 7.04 (0.36)
-0.04 (0.06) -0.10 (0.10) -0.17 (0.06)
Mean (SD)
7.62 (0.40)
7.52 (0.47)
-0.10 (0.07)
Note: FUW, feet underwater; FAW, feet above water.
horizontal and resultant force off the starting block had moderate to strong correlations with 7.5 – 15 m times for both techniques. Similarly take-off horizontal velocity was moderately correlated with 7.5 – 15 m times for both techniques as well as block time for the FAW technique. On the contrary, peak vertical force and take-off vertical velocity were not related to any of the time measures. Dive entry velocity had a relationship with the later phase (10 – 15 m) of the start for the FUW technique (entry velocity and 15 m, rs ¼ -0.67, p ¼ 0.02) and the wall and middle phases (7.5 – 10 m) of the FAW technique (wall time and entry velocity, rs ¼ -0.81, p , 0.01). Biomechanical predictors of a faster start Forward stepwise regression modelling identified the strongest predictors of 15 m start time for the two backstroke start techniques. For the FUW technique, take-off horizontal velocity was the best predictor of start time: Start time ðsÞ ¼ 13:0721:56v ðr ¼ 0:84; r 2 ¼ 0:74; SEE ¼ 0:30; SEE% ¼ 3:94Þ; where v is the take-off horizontal velocity. For the FAW start, take-off horizontal velocity and the swimmer’s start horizontal displacement of the hip were the best predictors of start time as summarised below: Start time ðsÞ ¼ 19:5222:86v22:98d ðr ¼ 0:94; r 2 ¼ 0:88; SEE ¼ 0:22; SEE% ¼ 2:93Þ; where v is the take-off horizontal velocity and d is the start horizontal displacement of the hip. Both linear regression analyses reveal that faster start times using either techniques required larger take-off horizontal velocity from the starting wall. In addition, for the FAW
0.68 (0.06) 1.86 (0.10) 3.22 (0.18) 4.63 (0.29) 7.59 (0.53) 0.50 (0.11) 0.07 (0.09) 34.66 (3.14) -0.08 (0.62) 3.51 (0.25) 2.34 (0.16) 3.56 (0.28) 1.33 (0.11) 1.47 (0.16) 1.99 (0.14)
Wall time (s) 5 m time (s) 7.5 m time (s) 10 m time (s) 15 m time (s) Start hip horizontal displacement (m) Start hip vertical displacement (m) Hip angle of entry (8) Take-off vertical velocity (m/s) Take-off horizontal velocity (m/s) Dive entry distance (m) Dive entry velocity (m/s) Peak vertical force (BW) Peak horizontal force (BW) Peak resultant force (BW)
0.67 (0.08) 1.87 (0.09) 3.22 (0.19) 4.63 (0.30) 7.58 (0.64) 0.48 (0.20) 0.04 (0.17) 34.00 (5.35) -0.04 (0.95) 3.50 (0.25) 2.39 (0.27) 3.75 (0.19) 1.33 (0.07) 1.47 (0.07) 1.96 (0.19)
Median (IQR) 0.66 (0.03) 1.72 (0.25) 3.13 (0.23) 4.54 (0.30) 7.51 (0.89) 0.54 (0.10) 0.18 (0.12) 38.19 (3.92) -0.70 (0.54) 3.65 (0.19) 2.39 (0.27) 3.75 (0.19) 1.40 (0.11) 1.42 (0.16) 1.99 (0.16)
Mean (SD)
Note: SD, standard deviation; IQR, inter quartile range; BW, body weight. *Significant at the 0.05 level.
Mean (SD)
Kinematic measure
FUW
FAW
0.66 (0.04) 1.77 (0.13) 3.07 (0.26) 4.46 (0.35) 7.44 (0.70) 0.53 (0.20) 0.20 (0.27) 38.20 (6.75) -0.72 (0.98) 3.66 (0.31) 2.40 (0.41) 3.77 (0.20) 1.37 (0.11) 1.38 (0.25) 1.97 (0.41)
Median (IQR)
Wilcoxon 0.090 0.004* 0.080 0.020* 0.010* 0.020* 0.010* 0.005* 0.060 0.010* 0.170 0.100 0.003* 0.020* 0.720
Difference (%) -3.30 ^ 2.80 -3.30 ^ 1.60 -2.19 ^ 1.40 -2.00 ^ 1.10 -1.40 ^ 0.70 9.20 ^ 5.00 100 ^ 98.10 10.10 ^ 4.60 -0.60 ^ 0.30 4.10 ^ 3.00 1.70 ^ 3.10 5.80 ^ 3.20 5.80 ^ 2.90 -4.10 ^ 2.50 0.30 ^ 1.60
Table II. Descriptive statistics for the kinematic and kinetic measures of start performance for the FUW and FAW techniques
Downloaded by [McMaster University] at 02:59 23 December 2014
160 C. Nguyen et al.
0.06 (0.87) -0.03 (0.94) -0.25 (0.44) -0.28 (0.38) -0.45 (0.15) 0.10 (0.75) -0.37 (0.24)
-0.16 (0.62) -0.07 (0.82) 0.14 (0.68) -0.48 (0.11) -0.56 (0.06) 0.12 (0.71) -0.46 (0.13)
5 m time (s) 0.08 (0.81) 0.07 (0.82) 0.24 (0.45) -0.68 (0.02)* -0.60 (0.04)* 0.26 (0.41) -0.54 (0.07)
7.5 m time (s) -0.33 (0.29) -0.15 (0.65) 0.25 (0.43) -0.77 (,0.01)* -0.66 (0.02)* 0.31 (0.34) -0.61 (0.04)*
10 m time (s) -0.40 (0.20) -0.18 (0.58) 0.23 (0.48) -0.78 (,0.01)* -0.68 (0.02)* 0.37 (0.24) -0.67 (0.02)*
15 m time (s) 0.21 (0.52) 0.06 (0.85) -0.50 (0.10) -0.33 (0.28) -0.40 (0.19) 0.15 (0.65) -0.81 (,0.01)*
Block time (s) 0.12 (0.72) 0.54 (0.63) -0.24 (0.46) -0.20 (0.54) -0.28 (0.38) -0.08 (0.81) -0.17 (0.60)
5 m time (s) -0.08 (0.80) -0.20 (0.53) -0.38 (0.22) -0.74 (0.01)* -0.73 (0.01)* -0.17 (0.60) -0.65 (0.02)*
7.5 m time (s)
FAW
-0.15 (0.65) -0.19 (0.55) -0.36 (0.25) -0.75 (0.01)* -0.73 (0.01)* -0.16 (0.62) -0.62 (0.03)*
10 m time (s)
-0.28 (0.40) -0.33 (0.30) -0.36 (0.26) -0.73 (0.01)* -0.71 (0.01)* -0.18 (0.58) -0.53 (0.07)
15 m time (s)
Notes: Results are reported as rs ( p); rs, Spearman r correlation coefficient; COM, centre of mass; BW, body weights. *Correlation significant at the 0.05 level.
Resultant entry velocity (m/s)
Take-off vertical velocity (m/s)
Peak resultant force (BW)
Peak horizontal force (BW)
Peak vertical force (BW)
Start hip horizontal displacement (m) Start hip vertical displacement (m)
Block time (s)
FUW
Table III. Results of Spearman r correlation analysis between kinematic and kinetic measures of the backstroke start.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start 161
162
C. Nguyen et al.
start, the horizontal displacement of the swimmer’s hip in the set up position was critical in predicting start time. The greater the horizontal distance of the hip from the wall, the quicker the start time.
Downloaded by [McMaster University] at 02:59 23 December 2014
Discussion and implications Adopting the FAW technique resulted in faster 15-m start times over the FUW start, with take-off horizontal velocity being significantly higher for this technique ( p , 0.01). Regression analysis indicated that higher take-off horizontal velocity highly correlated with faster start times, agreeing with the initial hypothesis and previous research (Kruger et al., 2006). Conversely, peak horizontal force was less for the FAW technique compared to the FUW start. On investigation of take-off velocity, the change in the swimmer’s set up position was found to contribute to the significant difference in take-off horizontal velocity between the FAW and FUW starts. The swimmer sat considerably higher on the wall when using the FAW technique; the average start vertical displacement of the hip being 0.18 m compared to 0.07 m for the FUW start ( p ¼ 0.01). Less of the body was in contact with the water; hence, there was less resistance or drag from the water that the body needed to move through to push off the wall (Hooper, 1981). The higher horizontal velocity for FAW may also be attributed to better force-time distribution during the take-off phase of the start. The area under a force-time curve represents impulse of the movement. When comparing the forcetime curves of the two techniques (Figure 2), the FAW start has a much greater area under the curve and hence greater impulse and take-off horizontal velocity. Take-off vertical velocity was on average larger for the FAW start. Again, this was due to the higher positioning of the body on the wall. The vertical distance that the swimmer’s COM (estimated from the hip position) had to travel before water entry was greater, explaining the larger negative inclination of take-off vertical velocity. This was further validated by the greater hip angle of entry for FAW ( p ¼ 0.01). Altering starting feet positions resulted in a significant change in peak horizontal ( p ¼ 0.02) and peak vertical force ( p ¼ 0.003). However, no significant differences were identified for both peak resultant force and dive entry distance. This suggests that the total force that each swimmer produced from the start was more or less the same regardless of
Figure 2. Example of the difference in the FUW and FAW techniques force-time curves
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
163
where they set their feet on the wall; the difference was in the direction of force production. To compensate for the lower peak horizontal force production during FAW, peak vertical force was larger (Table III). The greater peak horizontal force produced using the FUW may be due to a number of reasons. According to Lyttle and Benjanuvatra (2005), a longer time on the block or more correctly the wall for backstroke, allows for increased generation of force due to the relationship between time and force production. The results of this current study are supportive of their concept. The FUW start had a longer wall time then the FAW start (0.68 s vs. 0.66 s, respectively) and peak horizontal force was significantly higher for FUW ( p ¼ 0.02). De Jesus et al. (2010) suggested that the horizontal displacement of the COM at the start may affect force production. Having a COM horizontal start displacement closer to the wall would lead to greater generation of horizontal force due to the tighter coiling of major leg and trunk muscles involved in driving the body backwards. This study identified a considerable difference in horizontal displacement of the hip ( p ¼ 0.02) during the start phase and peak horizontal force ( p ¼ 0.02) between FAW and FUW. The hip horizontal displacement for FUW at the start was closer to the wall (0.50 m vs. 0.54 m), and FUW also had a larger peak horizontal force consistent with the idea that greater pre-stretching of the muscles (like for vertical jumping) produces more force. A major finding of this study was that the initial time variables after water entry (5, 7.5, and 10 m) were faster for the FAW start. At these distances, the influence of the wall, flight, and entry phases of the start were greater than after the 10-m mark when swimmers were well into executing underwater work. Past studies have revealed the importance of the underwater phase of the swimming start. Ruschel et al. (2007) observed that as underwater average velocity increased, the swimmers’ start time became faster (Ruschel et al., 2007). Pereira Ruschel, and Araujo (2006) also found a similar significant negative correlation between underwater velocity and 15 m time ( p , 0.01). Although underwater distance and time were determinants of underwater velocity, the two variables by themselves did not present significant correlations to start time (r ¼ 0.11 and r ¼ 0.38, respectively). This demonstrates that although distance and time may be factors in the underwater phase, the most critical component was the swimmer’s ability to minimise drag and optimise propulsion underwater. Cosser and Mason (2001) found that for ventral or backstroke starts, underwater distance and underwater time together had the greatest affect on time to 15 m. Due to the time constraints of the current study, only variables until water entry and then discrete time points from water entry to 15 m were investigated. Underwater parameters (including total time underwater, glide velocity, and distance underwater) were not examined. In an attempt to reduce the impact of differing underwater skill level, swimmers were instructed to perform both start variations and intra-subject comparisons made. This was performed with the idea that differences seen between variations would be due to takeoff, flight, and entry characteristics as each swimmers’ underwater skill would be the same for each start variation (Holthe & McLean, 2001). Correlation analysis revealed significant relationships between start time and take-off horizontal velocity, but only weak correlations with take-off vertical velocity for both techniques. As take-off horizontal velocity increased, start time was faster. Forward stepwise regression analysis with start time as the outcome indicated that take-off horizontal velocity as a predictor was significant. This demonstrates how take-off horizontal velocity is a critical component for an effective backstroke start regardless of feet setting on the wall. All correlations between entry velocity and the time points were negative, signifying that when entry velocity was higher, swim times were lower. Each participant’s technique preference was a critical consideration for this study. The effect of how much training a participant had
Downloaded by [McMaster University] at 02:59 23 December 2014
164
C. Nguyen et al.
done on the techniques prior to testing would have major influences on results. In a study by Vantorre et al. (2010) on front crawl (freestyle) swimming, it was found that athletes were more successful and efficient at reproducing their preferred start technique. The order of testing trials was randomised, and participants were provided with detailed demonstrations and feedback during familiarisation time to reduce the technique bias. Time limitations and the availability of high performance swimmers meant that the study could not be carried out over a number of weeks to provide more familiarisation time. Even with this limitation, when participants were divided into groups based on their technique preference, start time using the FAW method was faster for all groups regardless of preference. This result provides strong evidence supporting the use of the FAW technique and its biomechanical advantages for a successful start performance. The athletes that preferred the FAW technique were considerably faster using this start, with the mean difference in start time being -0.17 s ^ 0.06 s. This may be due to greater familiarity of using the FAW method. Nonetheless, it was determined that start time was faster using FAW for the preference groups of feet partially underwater (difference ¼ -0.10 s ^ 0.10 s) and completely underwater (mean: -0.04 s ^ 0.06 s). Conclusion The key finding of this study was that start time to 15 m was significantly different between the FUW and FAW backstroke starts, with FAW being quicker ( p , 0.01). In addition, the FAW technique was considerably faster to 5, 7.5, and 10 m. This suggests that the impulse on the wall at the start had a greater influence on the early race phase than underwater technique and effort. Specifically, take-off horizontal velocity was the most significant predictor of start time regardless of feet position, with hip horizontal displacement at the start also being important when using the FAW start. A larger peak horizontal force was not relevant for the faster start times observed for the FAW start as it was in fact lower for this technique. In line with the hypothesis, the high take-off horizontal velocity produced when adopting the FAW start was a crucial variable in a fast start, with it being greater for the FAW technique. Most competing backstrokers have progressed to using a higher feet technique at the start deeming it as more efficient. Further research is still required to provide indisputable results; however, this study has shown that with correct execution, using the FAW technique is effective in improving times to at least the initial segments of the backstroke start. Acknowledgements The authors wish to acknowledge the Australian Institute of Sport for the use of their facilities for this study. In addition, kind acknowledgements are presented to Swimming Australia and all participants for their persistent cooperation and support. References Blanksby, B., Nicholson, L., & Elliott, B. (2002). Biomechanical analysis of the grab, track and handle swimming starts: An intervention study. Sports Biomechanics, 1, 11–24. Breed, R., & McElroy, G. (2000). A biomechanical comparison of the grab, swing and track start in swimming. Journal of Human Movement Studies, 39, 277–293. Cosser, J., & Mason, B. (2001). Swim start performance at the Sydney 2000 Olympic Games. In J. R. Blackwell & R. H. Sanders (Eds.), XIX International Symposium on Biomechanics in Sports (pp. 70–74). San Francisco, CA: International Society of Biomechanics in Sports.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
165
de Jesus, K., de Jesus, K., Figueiredo, P., Gonc alves, P., Pereira, P., Vilas-Boas, J. P., & Fernandes, R. J. (2010). Biomechanical characterization of the backstroke start in immerged and emerged feet conditions. In P.-L. Kjendlie, R. K. Stallman, & J. Cabri (Eds.), XIth International Symposium for Biomechanics and Medicine in Swimming (pp. 64–66). Norway: Nordbergtrykk. de Jesus, K., de Jesus, K., Figueirdeo, P., Gonc alves, P., Pereira, S., Villa-Boas, J. P., & Fernandes, R. J. (2011). Biomechanical analysis of backstroke swimming starts. International Journal of Sports Medicine, 32, 546–551. Federation Internationale de Natation. (2011). SW 6 backstroke swimming rules. Retrieved May 3, 2011, from http://www.fina.org/H2O/index.php?option¼com_content&view¼article&id¼283:sw-6-backstroke&catid¼82: swimming-rules&Itemid¼184 Getchell, N., & Roberton, M. A. (1989). Whole body stiffness as a function of developmental level in children’s hopping. Development Psychology, 25, 920 –928. Hohmann, A., Fehr, U., Kirsten, R., & Krueger, T. (2008), Biomechanical analysis of the backstroke start technique in swimming. E-Journal Bewegung und Training, 2, 28–33. Holthe, M., & McLean, S. (2001). Kinematic comparison of grab and track starts in swimming. In J. R. Blackwell & R. H. Sanders (Eds.), XIXth International Symposium on Biomechanics in Sports (pp. 31–34). San Francisco, CA: International Society of Biomechanics in Sports. Hooper, S. (1981). Identification of the biomechanical factors affecting the performance of the backstroke swimming start. Brisbane: University of Queensland. Jorgic´, B., Puletic´, M., Stankovic´, R., Okicˇic´, T., Bubanj, S., & Bubanj, R. (2010). The kinematic analysis of the grab and track start in swimming. Physical Education and Sport, 8, 31–36. Kruger, T., Hohmann, A., Kirsten, R., & Wick, D. (2006). Kinematics and dynamics of the backstroke start. In J. P. Vilas-Boas, F. Alves, & A. Marques (Eds.), Xth International Symposium for Biomechanics and Medicine in Swimming (pp. 58– 60). Portugal: Portuguese Journal of Sport Sciences. Lyttle, A., & Benjanuvatra, N. (2005). Start right? A biomechanical review of dive start performance. Retrieved August 9, 2011, from http://www.coachesinfo.com/index.php?option¼com_content&view¼article&id¼89: swimming-start-style&catid¼49:swimming-coaching&Itemid¼86 Mason, B., & Cosser, J. (2000). What can we learn from competition analysis at the 1999 Pan Pacific swimming championships? In Y. Hong, D. P. Johns, & R. Sanders (Eds.), XVIIIth International Symposium on Biomechanics in Sports (pp. 75–82). Hong Kong: International Society of Biomechanics in Sports. Mason, B., Mackintosh, C., & Pease, D. (2012). The development of an analysis system to assist in the correction of inefficiencies in starts and turns for elite competitive swimming. In E. J. Bradshaw, A. Burnett, & P. A. Hume (Eds.), XXXth International Symposium on Biomechanics in Sports (pp. 249–252). Melbourne: International Society of Biomechanics in Sports. Pereira, S., Ruschel, C., & Araujo, L. (2006). Biomechanical analysis of the underwater phase in swimming starts. In J. P. Vilas-Boas, F. Alves, & A. Marques (Eds.), Xth International Symposium for Biomechanics and Medicine in Swimming (pp. 79– 81). Portugal: Portuguese Journal of Sport Sciences. Ranavolo, A., Don, R., Cacchio, A., Serrao, M., Paoloni, M., Mangone, M., & Santilli, V. (2008). Comparison between kinematic and kinetic methods for computing the vertical displacement of the centre of mass during human hopping at different frequencies. Journal of Applied Biomechanics, 24, 271–279. Ruschel, C., Araujo, L. G., Pereira, S. M., & Roesler, H. (2007). Kinematical analysis of the swimming start: Block, flight and underwater phases. In H. J. Menzel & M. H. Chagas (Eds.), XXVth International Symposium on Biomechanics in Sports (pp. 385–388). Brazil: International Society of Biomechanics in Sports. Swimming Australia. (2012). 2012 Australian swimming championships entry qualifying times. Retrieved March 1, 2012, from http://www.210.247.205.118/assets/console/document/documents/Aus%202012%20LC%20QT’s. pdf Vantorre, J., Seifert, L., Fernandes, R., Vilas-Boas, J., & Chollet, D. (2010). Biomechanical influence of start technique preference for elite track starters in front crawl. Open Sports Sciences Journal, 3, 137–139. Zar, J. H. (1972). Significance testing of the Spearman rank correlation. Journal of the American Statistical Association, 67, 578–580.
Sports Biomechanics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rspb20
Is starting with the feet out of the water faster in backstroke swimming? ab
b
a
Cecilia Nguyen , Elizabeth J. Bradshaw , David Pease & Cameron Wilson
b
a
Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Canberra, Australia b
School of Exercise Science, Australian Catholic University, Melbourne, Australia Published online: 24 Feb 2014.
Click for updates To cite this article: Cecilia Nguyen, Elizabeth J. Bradshaw, David Pease & Cameron Wilson (2014) Is starting with the feet out of the water faster in backstroke swimming?, Sports Biomechanics, 13:2, 154-165, DOI: 10.1080/14763141.2014.885072 To link to this article: http://dx.doi.org/10.1080/14763141.2014.885072
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [McMaster University] at 02:59 23 December 2014
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
Sports Biomechanics, 2014 Vol. 13, No. 2, 154–165, http://dx.doi.org/10.1080/14763141.2014.885072
Is starting with the feet out of the water faster in backstroke swimming? CECILIA NGUYEN1,2, ELIZABETH J. BRADSHAW2, DAVID PEASE1, & CAMERON WILSON2 Downloaded by [McMaster University] at 02:59 23 December 2014
1
Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Canberra, Australia, and 2School of Exercise Science, Australian Catholic University, Melbourne, Australia
(Received 16 December 2012; accepted 14 January 2014)
Abstract This study aimed to determine if starting with the feet above the water (FAW) in male backstroke swimming resulted in faster start times (15-m time) than when the feet were underwater (FUW). It was hypothesised that setting higher on the wall would generate increased horizontal force and velocity, resulting in quicker starts. Twelve high-level male backstrokers performed three trials of the FAW and FUW techniques. A biomechanical swimming testing system comprising one force plate (1,000 Hz), four lateral-view (100 Hz), and five overhead (50 Hz) video cameras captured the swimmers’ performance. Data for each participant’s fastest trial for each technique were collated, grouped, and statistically analysed. Analysis included Wilcoxon, Spearman Rho correlation, and regression analysis. Wilcoxon results revealed a significantly faster start time for the FAW technique ( p , 0.01). Peak horizontal force was significantly smaller for FAW ( p ¼ 0.02), while take-off horizontal velocity was significantly greater ( p ¼ 0.01). Regression analysis indicated take-off horizontal velocity to be a good predictor of start time for both techniques, and the horizontal displacement of the centre of mass for the FAW start.
Keywords: Biomechanics, force, velocity, start
Introduction The start phase is defined by many as the time it takes the middle of the swimmers head to reach the 15-m marker and is considered a critical phase of a swimming race (e.g. Cossor & Mason, 2001; Jorgic´ et al., 2010; Mason & Cosser, 2000; Ruschel, Araujo, Pereira, & Roesler, 2007; Vantorre, Seifert, Fernandes, Vilas-Boas, & Chollet, 2010). The start represents between 0.8% and 26.1% of the final race time with the influence being greater for shorter events such as the 100 m (Mason & Cosser, 2000). In World Championship and Olympic competitions, races have been won or lost by a fraction of a second. For example, at the 2012 Summer Olympics in London, American Nathan Adrian edged out Australian James Magnussen in the 100-m freestyle event by 0.01 s to win gold. Also at the 2008 Correspondence: Cecilia Nguyen, Aquatic Testing and Research Unit, Sports Science and Sports Medicine, Australian Institute of Sport, Leverrier Street, Bruce, ACT 2617, Australia, E-mail: [email protected] q 2014 Taylor & Francis
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
155
Olympics, German Britta Steffen won the 50-m freestyle gold by the same margin over American Dana Torres. Swimmers are continually developing their start technique with the goal of gaining a faster start time and therefore an early edge over their competitors. For freestyle, breaststroke, and butterfly, the ventral dive start is used. This is where the swimmer stands on top of the starting block and dives into the water once the trigger sounds. The backstroke start requires different skills to the ventral start as it begins with the swimmer already in the water, holding onto the starting block and diving backwards. In July 2005, the Federation Internationale de Natation (FINA, 2011) modified the backstroke start rule to allow competitors to set their feet above the water (FAW) line prior to the take-off signal of a race. Since this rule change, many competitive backstrokers have altered their starting technique to place their feet completely out of the water. With advances in knowledge and technology, biomechanical parameters of the swimming start have been broadly investigated to determine if any significantly affect start time. Research is abundant comparing the grab and track starts, which are the two most popular ventral starts (e.g. Blanksby, Nicholson, & Elliott, 2002; Breed & McElroy, 2000; Hohmann, Fehr, Kirsten, & Krueger, 2008). However, those studying the differences in backstroke start techniques are scarce, particularly since the 2005 rule change. Kruger, Hohmann, Kirsten, and Wick (2006) noted that there were definite kinematic and dynamic influences of the overwater (take-off and flight) phases of the backstroke start. Comparable to ventral start studies, strong correlations were identified between resultant take-off force and start time (Kruger et al., 2006). The higher the impulse or resultant force on the wall, the greater the take-off velocity and quicker time to 7.5 m (r ¼ 0.83, p , 0.01). In a comparison study of backstroke start techniques, de Jesus et al. (2011) proposed that shorter start times would be attributed to the mechanical characteristics at the wall set position. It was suggested that setting the feet high on the wall and above the water at the start may produce a larger horizontal force off the wall (de Jesus et al., 2011). This was because the centre of mass (COM) was higher; hence, the swimmer was not required to project their body as much vertically (de Jesus et al., 2010; Hooper, 1981). The purpose of this study was to compare the biomechanics of the FAW and feet underwater (FUW) starting techniques for backstroke to determine if starting with the FAW led to a faster start time. It was hypothesised that placing the feet high on the wall would produce faster start times (15-m time) due to greater generation of horizontal force and velocity from a higher COM in the set up position.
Methods Participants Twelve Australian male backstroke swimmers aged 20.5 (SD: 1.6 years) (M ^ SD height ¼ 187.0 ^ 6.0 cm, body mass ¼ 80.7 ^ 6.0 kg) were recruited for this study. All participants were healthy (no serious injury or illness in the last 6 months), able-bodied, and were, at the time of this study, racing in state and national competitions with backstroke as their main event. Only males were recruited for this study as very few female backstrokers use the FAW starting technique. The FINA point score system was used to quantify skill level. The minimum scores required for participation in this study were 770 FINA points in the 50-m backstroke event and 602 FINA points in the 100 m. These scores corresponded to the qualifying times of the 2012 Energy Australia Swimming Championships (Swimming Australia, 2012). Both the Australian Institute of Sport (AIS) and Australian Catholic University ethics committees approved the study before testing commenced.
156
C. Nguyen et al.
Downloaded by [McMaster University] at 02:59 23 December 2014
Procedure The participants attended one testing session of approximately 1 h at the AIS technology pool in Canberra. The height (cm) and body mass (kg) of the participants were measured at the start of the session. Each swimmer was then asked what starting technique they would normally use in a race (feet completely underwater, partially underwater, or their whole FAW) and this was recorded. The participants then completed a warm up that involved swimming 400 m of freestyle at their own pace. The testing required participants to perform both variants of the backstroke start. These were starting backstroke with the feet fully above the water (FAW) and then completely underwater (FUW). For the FAW technique, the feet were set so that the heels were just above the water line (Figure 1a). For the FUW, the toes were placed just below the water’s surface (Figure 1b). The technique variants were randomised with the order determined before testing commenced. The two backstroke techniques were verbally described as well as visually depicted by video recordings to each participant. The swimmers were provided with familiarisation time of each technique before data collection commenced. This involved participants performing a number of practice trials of both techniques on the testing wall. Verbal instruction and feedback were given during familiarisation to ensure that the starts were performed correctly. Each participant performed three trials of the FAW and FUW variants, with 2 min rest in between each trial. Participants wore their standard training suit during testing.
Data collection The biomechanical systems embedded in the AIS technology pool have previously been described by Mason, Mackintosh, and Pease (2012). Four high-speed digital gigabit Ethernet cameras (TMC6740GE, 100 Hz, Pulnixw, San Jose, CA, USA) recorded the swimmer’s performance from a lateral view for the whole 15 m distance (Mason et al., 2012). Camera one viewed the above water left lateral side at initial set up position, take-off, and water entry; cameras 2–4 recorded a continuous view of the underwater left lateral side. The feeds from all cameras were integrated so that one complete video recording of the trial was produced. An overhead video system captured timing information. This overhead video system utilized five overhead cameras (SCC-C4301, 50 Hz, Samsung, Suwon, South Korea) positioned normal to the lane in which the athletes were performing at key distances (5, 7.5, 10, 15, and 20 m).
Figure 1. (a) FAW starting technique and (b) FUW starting technique. The heels of the feet are above the water’s surface for FAW and the toes of the feet are below the water’s surface for FUW.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
157
From these cameras, it was possible to obtain the time when the centre of the athlete’s head passed the given distance. The cameras were genlocked to provide a synchronised image that ran through in-house developed time-coding computer software. When watching the playback of the trials, a fixed line on the screen at each segment was used as a distance marker. Using this system, the time it took for the middle of the athlete’s head to reach each distance marker was determined. The middle of the head was used as the reference point because it is also used as the marker for ‘breakout distance’ during a race. Breakout distance is measured during the start phase and after a turn. It is the distance between the starting or turning wall and the point at which the swimmer’s head breaks the surface of the water. One force plate (Z20313, 600 mm £ 900 mm £ 40 mm, 100 Hz, Kistler, Winterthur, Switzerland) was inserted into the starting end of the pool, flush with the wall to measure the three-dimensional forces when the swimmer pushed off after the starting signal. All force measures were transformed from analogue to digital signals via amplifiers (Type 5070 Multichannel Charge Amplifier, Kistler, Winterthur, Switzerland) and a converter board and then relayed via cables to a personal computer that ran customised software (Mason et al., 2012). An auditory buzzer signal, similar to what is used in competition, was used to simultaneously start the swimmer and the data collection systems. Data analysis The customised computer program enabled semi-automatic analysis of force platform data and video images collected from the video cameras (Mason et al., 2012). Video footage from the four high-speed cameras (100 Hz) were manually digitised by finding the centre and top of the swimmer’s head and approximate COM during the start; specifically in the initial set up position, at the point when the feet left the wall and when the head entered the water. This was estimated using the grid markings from the computer software and the swimmer’s left greater trochanter (left hip). The hip is the closest point to the COM and therefore provides an easily digitisable, body landmark during analysis (Getchell & Robertson, 1989; Ranavolo et al., 2008). The left hip was therefore used to estimate the COM position in this study. From finding these points, and using a 100-Hz video camera that captured a frame every hundredth of a second, the computer program was able to calculate the biomechanical variables investigated. Feet position on the wall at the start was the independent variable (the feet position was either FAW or FUW). The main dependent performance variable was start time (time from the start signal to 15 m). Other variables examined were wall time (time from start signal until time the swimmer’s feet left the wall); 5-, 7.5-, 10-m time (all to when the middle of the swimmer’s head reached the distances); peak vertical, horizontal, and resultant forces (peak forces over the whole take-off phase); take-off vertical and horizontal velocities (at the instant of take-off); entry velocity (calculated from horizontal and vertical take-off velocities); entry distance (wall to head water contact); and horizontal and vertical displacement of the hip at the start and hip angle of entry (calculated using take-off horizontal and vertical velocities and the water line as the horizontal). Dependent measures for each trial and swimmer were collated in an Excel spreadsheet (Microsoft, Redmond, WA, USA). The trial with the fastest start time for each swimmer for the two start techniques was identified. The data were then grouped and descriptive statistical analysis was performed using Statistical Packages for the Social Sciences software (version 19.0.0, IBM Corporation, New York, NY, USA). An a level of 0.05 was set as the significance level for all statistical tests. Normality of the grouped data was tested using the Shapiro – Wilk test. The data was normally distributed with the exception of age, entry
Downloaded by [McMaster University] at 02:59 23 December 2014
158
C. Nguyen et al.
velocity for the feet above condition, time to 5 m for both conditions, and peak vertical force for both conditions. Due to the inconsistency in the normality of the data-set and the small sample size, non-parametric statistical methods were subsequently employed unless otherwise stated. A Wilcoxon test was used to identify any statistical differences between the biomechanics of the two starting techniques. Spearman Rho (r) correlation analysis was used to determine if there were any significant relationships between time variables and velocity, force, or start hip position components of the backstroke start for both FUW and FAW. Significant correlations were identified if jrsj . 0.59, p , 0.05 (Zar, 1972). Following the correlation tests, a forward stepwise regression analysis was performed on the data. The two starting conditions were analysed separately, with start time as the dependent variable and variables that had significant correlations set as the independent variables. Using this method, one variable at a time was added to the regression model until the entered variable’s p value was higher than the p value cutoff ( p . 0.05). From this forward stepwise regression, the best predictors of start time for the FUW and FAW techniques were determined. Results Backstroke start performance On average, the FAW technique resulted in a 0.08-s (1.1%) faster starting performance than the FUW technique ( p , 0.01). Out of the 12 swimmers, 4 participants preferred the FAW technique, 3 preferred the FUW technique, and 5 participants normally adopted the feet partially underwater technique. When the participants’ results were divided into groups based on their technique preference, descriptive analysis showed that FAW start time was faster than the FUW start time regardless of the swimmers technique preference (Table I). Furthermore, an examination of the individual data revealed that 10 out of the 12 participants were able to perform a quicker backstroke start to the 15-m lane marker when using the FAW technique. The largest improvement observed was 0.24 s for participant 3 which represented a 2.3% faster start time. The smallest improvement observed was 0.03 s for participant 8, which represented a 0.4% faster start time. Both swimmers (participants 6 and 10) who had performance decrements with the FAW technique had a 0.02-s (participant 6 –0.2%, participant 10 – 0.3%) slower start time. Kinematic and kinetic differences between start techniques The results for the kinematic and kinetic measures of start performance are summarised in Table II. Small technical differences were identified in the swimmers start displacement (start hip position), and the vertical and horizontal forces developed ( p , 0.05). The swimmers had a higher set position (start vertical hip displacement) for FAW which resulted in the generation of higher vertical but less horizontal force. This slight variation between the FUW and FAW techniques had no effect on resultant force generated but did result in a significantly faster take-off horizontal velocity ( p ¼ 0.010) and a steeper dive entry ( p , 0.01) for FAW. The timing results showed that the FAW start lead to faster times at the 5- and 10-m lane markers. Biomechanical attributes of a faster start The results of the Spearman r correlation analysis between biomechanical measures of the FUW and FAW techniques are summarised in Table III. The generation of a high peak
Foot position in backstroke start
159
Table I. Best start times
Downloaded by [McMaster University] at 02:59 23 December 2014
Start time (s)
Participant
Technique preference
1 2 3 4 5 6 7 8 9 10 11 12 Technique preference group Overall group
Group ranking
FUW
FAW
Difference FAW– FUW
Above Partial Above Above Above Under Under Partial Partial Partial Partial Under
8 3 2 4 1 11 10 12 5 7 6 9
7.72 7.2 7.02 7.22 6.88 8.42 7.88 8.61 7.28 7.6 7.56 7.74
7.54 7.06 6.78 7.06 6.78 8.44 7.78 8.58 7.14 7.62 7.34 7.7
-0.18 -0.14 -0.24 -0.16 -0.1 -0.02 -0.1 -0.03 -0.14 -0.02 -0.22 -0.04
Under Partial Above
Mean (SD) Mean (SD) Mean (SD)
8.01 (0.36) 7.65 (0.56) 7.21 (0.37)
7.97 (0.41) 7.55 (0.55) 7.04 (0.36)
-0.04 (0.06) -0.10 (0.10) -0.17 (0.06)
Mean (SD)
7.62 (0.40)
7.52 (0.47)
-0.10 (0.07)
Note: FUW, feet underwater; FAW, feet above water.
horizontal and resultant force off the starting block had moderate to strong correlations with 7.5 – 15 m times for both techniques. Similarly take-off horizontal velocity was moderately correlated with 7.5 – 15 m times for both techniques as well as block time for the FAW technique. On the contrary, peak vertical force and take-off vertical velocity were not related to any of the time measures. Dive entry velocity had a relationship with the later phase (10 – 15 m) of the start for the FUW technique (entry velocity and 15 m, rs ¼ -0.67, p ¼ 0.02) and the wall and middle phases (7.5 – 10 m) of the FAW technique (wall time and entry velocity, rs ¼ -0.81, p , 0.01). Biomechanical predictors of a faster start Forward stepwise regression modelling identified the strongest predictors of 15 m start time for the two backstroke start techniques. For the FUW technique, take-off horizontal velocity was the best predictor of start time: Start time ðsÞ ¼ 13:0721:56v ðr ¼ 0:84; r 2 ¼ 0:74; SEE ¼ 0:30; SEE% ¼ 3:94Þ; where v is the take-off horizontal velocity. For the FAW start, take-off horizontal velocity and the swimmer’s start horizontal displacement of the hip were the best predictors of start time as summarised below: Start time ðsÞ ¼ 19:5222:86v22:98d ðr ¼ 0:94; r 2 ¼ 0:88; SEE ¼ 0:22; SEE% ¼ 2:93Þ; where v is the take-off horizontal velocity and d is the start horizontal displacement of the hip. Both linear regression analyses reveal that faster start times using either techniques required larger take-off horizontal velocity from the starting wall. In addition, for the FAW
0.68 (0.06) 1.86 (0.10) 3.22 (0.18) 4.63 (0.29) 7.59 (0.53) 0.50 (0.11) 0.07 (0.09) 34.66 (3.14) -0.08 (0.62) 3.51 (0.25) 2.34 (0.16) 3.56 (0.28) 1.33 (0.11) 1.47 (0.16) 1.99 (0.14)
Wall time (s) 5 m time (s) 7.5 m time (s) 10 m time (s) 15 m time (s) Start hip horizontal displacement (m) Start hip vertical displacement (m) Hip angle of entry (8) Take-off vertical velocity (m/s) Take-off horizontal velocity (m/s) Dive entry distance (m) Dive entry velocity (m/s) Peak vertical force (BW) Peak horizontal force (BW) Peak resultant force (BW)
0.67 (0.08) 1.87 (0.09) 3.22 (0.19) 4.63 (0.30) 7.58 (0.64) 0.48 (0.20) 0.04 (0.17) 34.00 (5.35) -0.04 (0.95) 3.50 (0.25) 2.39 (0.27) 3.75 (0.19) 1.33 (0.07) 1.47 (0.07) 1.96 (0.19)
Median (IQR) 0.66 (0.03) 1.72 (0.25) 3.13 (0.23) 4.54 (0.30) 7.51 (0.89) 0.54 (0.10) 0.18 (0.12) 38.19 (3.92) -0.70 (0.54) 3.65 (0.19) 2.39 (0.27) 3.75 (0.19) 1.40 (0.11) 1.42 (0.16) 1.99 (0.16)
Mean (SD)
Note: SD, standard deviation; IQR, inter quartile range; BW, body weight. *Significant at the 0.05 level.
Mean (SD)
Kinematic measure
FUW
FAW
0.66 (0.04) 1.77 (0.13) 3.07 (0.26) 4.46 (0.35) 7.44 (0.70) 0.53 (0.20) 0.20 (0.27) 38.20 (6.75) -0.72 (0.98) 3.66 (0.31) 2.40 (0.41) 3.77 (0.20) 1.37 (0.11) 1.38 (0.25) 1.97 (0.41)
Median (IQR)
Wilcoxon 0.090 0.004* 0.080 0.020* 0.010* 0.020* 0.010* 0.005* 0.060 0.010* 0.170 0.100 0.003* 0.020* 0.720
Difference (%) -3.30 ^ 2.80 -3.30 ^ 1.60 -2.19 ^ 1.40 -2.00 ^ 1.10 -1.40 ^ 0.70 9.20 ^ 5.00 100 ^ 98.10 10.10 ^ 4.60 -0.60 ^ 0.30 4.10 ^ 3.00 1.70 ^ 3.10 5.80 ^ 3.20 5.80 ^ 2.90 -4.10 ^ 2.50 0.30 ^ 1.60
Table II. Descriptive statistics for the kinematic and kinetic measures of start performance for the FUW and FAW techniques
Downloaded by [McMaster University] at 02:59 23 December 2014
160 C. Nguyen et al.
0.06 (0.87) -0.03 (0.94) -0.25 (0.44) -0.28 (0.38) -0.45 (0.15) 0.10 (0.75) -0.37 (0.24)
-0.16 (0.62) -0.07 (0.82) 0.14 (0.68) -0.48 (0.11) -0.56 (0.06) 0.12 (0.71) -0.46 (0.13)
5 m time (s) 0.08 (0.81) 0.07 (0.82) 0.24 (0.45) -0.68 (0.02)* -0.60 (0.04)* 0.26 (0.41) -0.54 (0.07)
7.5 m time (s) -0.33 (0.29) -0.15 (0.65) 0.25 (0.43) -0.77 (,0.01)* -0.66 (0.02)* 0.31 (0.34) -0.61 (0.04)*
10 m time (s) -0.40 (0.20) -0.18 (0.58) 0.23 (0.48) -0.78 (,0.01)* -0.68 (0.02)* 0.37 (0.24) -0.67 (0.02)*
15 m time (s) 0.21 (0.52) 0.06 (0.85) -0.50 (0.10) -0.33 (0.28) -0.40 (0.19) 0.15 (0.65) -0.81 (,0.01)*
Block time (s) 0.12 (0.72) 0.54 (0.63) -0.24 (0.46) -0.20 (0.54) -0.28 (0.38) -0.08 (0.81) -0.17 (0.60)
5 m time (s) -0.08 (0.80) -0.20 (0.53) -0.38 (0.22) -0.74 (0.01)* -0.73 (0.01)* -0.17 (0.60) -0.65 (0.02)*
7.5 m time (s)
FAW
-0.15 (0.65) -0.19 (0.55) -0.36 (0.25) -0.75 (0.01)* -0.73 (0.01)* -0.16 (0.62) -0.62 (0.03)*
10 m time (s)
-0.28 (0.40) -0.33 (0.30) -0.36 (0.26) -0.73 (0.01)* -0.71 (0.01)* -0.18 (0.58) -0.53 (0.07)
15 m time (s)
Notes: Results are reported as rs ( p); rs, Spearman r correlation coefficient; COM, centre of mass; BW, body weights. *Correlation significant at the 0.05 level.
Resultant entry velocity (m/s)
Take-off vertical velocity (m/s)
Peak resultant force (BW)
Peak horizontal force (BW)
Peak vertical force (BW)
Start hip horizontal displacement (m) Start hip vertical displacement (m)
Block time (s)
FUW
Table III. Results of Spearman r correlation analysis between kinematic and kinetic measures of the backstroke start.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start 161
162
C. Nguyen et al.
start, the horizontal displacement of the swimmer’s hip in the set up position was critical in predicting start time. The greater the horizontal distance of the hip from the wall, the quicker the start time.
Downloaded by [McMaster University] at 02:59 23 December 2014
Discussion and implications Adopting the FAW technique resulted in faster 15-m start times over the FUW start, with take-off horizontal velocity being significantly higher for this technique ( p , 0.01). Regression analysis indicated that higher take-off horizontal velocity highly correlated with faster start times, agreeing with the initial hypothesis and previous research (Kruger et al., 2006). Conversely, peak horizontal force was less for the FAW technique compared to the FUW start. On investigation of take-off velocity, the change in the swimmer’s set up position was found to contribute to the significant difference in take-off horizontal velocity between the FAW and FUW starts. The swimmer sat considerably higher on the wall when using the FAW technique; the average start vertical displacement of the hip being 0.18 m compared to 0.07 m for the FUW start ( p ¼ 0.01). Less of the body was in contact with the water; hence, there was less resistance or drag from the water that the body needed to move through to push off the wall (Hooper, 1981). The higher horizontal velocity for FAW may also be attributed to better force-time distribution during the take-off phase of the start. The area under a force-time curve represents impulse of the movement. When comparing the forcetime curves of the two techniques (Figure 2), the FAW start has a much greater area under the curve and hence greater impulse and take-off horizontal velocity. Take-off vertical velocity was on average larger for the FAW start. Again, this was due to the higher positioning of the body on the wall. The vertical distance that the swimmer’s COM (estimated from the hip position) had to travel before water entry was greater, explaining the larger negative inclination of take-off vertical velocity. This was further validated by the greater hip angle of entry for FAW ( p ¼ 0.01). Altering starting feet positions resulted in a significant change in peak horizontal ( p ¼ 0.02) and peak vertical force ( p ¼ 0.003). However, no significant differences were identified for both peak resultant force and dive entry distance. This suggests that the total force that each swimmer produced from the start was more or less the same regardless of
Figure 2. Example of the difference in the FUW and FAW techniques force-time curves
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
163
where they set their feet on the wall; the difference was in the direction of force production. To compensate for the lower peak horizontal force production during FAW, peak vertical force was larger (Table III). The greater peak horizontal force produced using the FUW may be due to a number of reasons. According to Lyttle and Benjanuvatra (2005), a longer time on the block or more correctly the wall for backstroke, allows for increased generation of force due to the relationship between time and force production. The results of this current study are supportive of their concept. The FUW start had a longer wall time then the FAW start (0.68 s vs. 0.66 s, respectively) and peak horizontal force was significantly higher for FUW ( p ¼ 0.02). De Jesus et al. (2010) suggested that the horizontal displacement of the COM at the start may affect force production. Having a COM horizontal start displacement closer to the wall would lead to greater generation of horizontal force due to the tighter coiling of major leg and trunk muscles involved in driving the body backwards. This study identified a considerable difference in horizontal displacement of the hip ( p ¼ 0.02) during the start phase and peak horizontal force ( p ¼ 0.02) between FAW and FUW. The hip horizontal displacement for FUW at the start was closer to the wall (0.50 m vs. 0.54 m), and FUW also had a larger peak horizontal force consistent with the idea that greater pre-stretching of the muscles (like for vertical jumping) produces more force. A major finding of this study was that the initial time variables after water entry (5, 7.5, and 10 m) were faster for the FAW start. At these distances, the influence of the wall, flight, and entry phases of the start were greater than after the 10-m mark when swimmers were well into executing underwater work. Past studies have revealed the importance of the underwater phase of the swimming start. Ruschel et al. (2007) observed that as underwater average velocity increased, the swimmers’ start time became faster (Ruschel et al., 2007). Pereira Ruschel, and Araujo (2006) also found a similar significant negative correlation between underwater velocity and 15 m time ( p , 0.01). Although underwater distance and time were determinants of underwater velocity, the two variables by themselves did not present significant correlations to start time (r ¼ 0.11 and r ¼ 0.38, respectively). This demonstrates that although distance and time may be factors in the underwater phase, the most critical component was the swimmer’s ability to minimise drag and optimise propulsion underwater. Cosser and Mason (2001) found that for ventral or backstroke starts, underwater distance and underwater time together had the greatest affect on time to 15 m. Due to the time constraints of the current study, only variables until water entry and then discrete time points from water entry to 15 m were investigated. Underwater parameters (including total time underwater, glide velocity, and distance underwater) were not examined. In an attempt to reduce the impact of differing underwater skill level, swimmers were instructed to perform both start variations and intra-subject comparisons made. This was performed with the idea that differences seen between variations would be due to takeoff, flight, and entry characteristics as each swimmers’ underwater skill would be the same for each start variation (Holthe & McLean, 2001). Correlation analysis revealed significant relationships between start time and take-off horizontal velocity, but only weak correlations with take-off vertical velocity for both techniques. As take-off horizontal velocity increased, start time was faster. Forward stepwise regression analysis with start time as the outcome indicated that take-off horizontal velocity as a predictor was significant. This demonstrates how take-off horizontal velocity is a critical component for an effective backstroke start regardless of feet setting on the wall. All correlations between entry velocity and the time points were negative, signifying that when entry velocity was higher, swim times were lower. Each participant’s technique preference was a critical consideration for this study. The effect of how much training a participant had
Downloaded by [McMaster University] at 02:59 23 December 2014
164
C. Nguyen et al.
done on the techniques prior to testing would have major influences on results. In a study by Vantorre et al. (2010) on front crawl (freestyle) swimming, it was found that athletes were more successful and efficient at reproducing their preferred start technique. The order of testing trials was randomised, and participants were provided with detailed demonstrations and feedback during familiarisation time to reduce the technique bias. Time limitations and the availability of high performance swimmers meant that the study could not be carried out over a number of weeks to provide more familiarisation time. Even with this limitation, when participants were divided into groups based on their technique preference, start time using the FAW method was faster for all groups regardless of preference. This result provides strong evidence supporting the use of the FAW technique and its biomechanical advantages for a successful start performance. The athletes that preferred the FAW technique were considerably faster using this start, with the mean difference in start time being -0.17 s ^ 0.06 s. This may be due to greater familiarity of using the FAW method. Nonetheless, it was determined that start time was faster using FAW for the preference groups of feet partially underwater (difference ¼ -0.10 s ^ 0.10 s) and completely underwater (mean: -0.04 s ^ 0.06 s). Conclusion The key finding of this study was that start time to 15 m was significantly different between the FUW and FAW backstroke starts, with FAW being quicker ( p , 0.01). In addition, the FAW technique was considerably faster to 5, 7.5, and 10 m. This suggests that the impulse on the wall at the start had a greater influence on the early race phase than underwater technique and effort. Specifically, take-off horizontal velocity was the most significant predictor of start time regardless of feet position, with hip horizontal displacement at the start also being important when using the FAW start. A larger peak horizontal force was not relevant for the faster start times observed for the FAW start as it was in fact lower for this technique. In line with the hypothesis, the high take-off horizontal velocity produced when adopting the FAW start was a crucial variable in a fast start, with it being greater for the FAW technique. Most competing backstrokers have progressed to using a higher feet technique at the start deeming it as more efficient. Further research is still required to provide indisputable results; however, this study has shown that with correct execution, using the FAW technique is effective in improving times to at least the initial segments of the backstroke start. Acknowledgements The authors wish to acknowledge the Australian Institute of Sport for the use of their facilities for this study. In addition, kind acknowledgements are presented to Swimming Australia and all participants for their persistent cooperation and support. References Blanksby, B., Nicholson, L., & Elliott, B. (2002). Biomechanical analysis of the grab, track and handle swimming starts: An intervention study. Sports Biomechanics, 1, 11–24. Breed, R., & McElroy, G. (2000). A biomechanical comparison of the grab, swing and track start in swimming. Journal of Human Movement Studies, 39, 277–293. Cosser, J., & Mason, B. (2001). Swim start performance at the Sydney 2000 Olympic Games. In J. R. Blackwell & R. H. Sanders (Eds.), XIX International Symposium on Biomechanics in Sports (pp. 70–74). San Francisco, CA: International Society of Biomechanics in Sports.
Downloaded by [McMaster University] at 02:59 23 December 2014
Foot position in backstroke start
165
de Jesus, K., de Jesus, K., Figueiredo, P., Gonc alves, P., Pereira, P., Vilas-Boas, J. P., & Fernandes, R. J. (2010). Biomechanical characterization of the backstroke start in immerged and emerged feet conditions. In P.-L. Kjendlie, R. K. Stallman, & J. Cabri (Eds.), XIth International Symposium for Biomechanics and Medicine in Swimming (pp. 64–66). Norway: Nordbergtrykk. de Jesus, K., de Jesus, K., Figueirdeo, P., Gonc alves, P., Pereira, S., Villa-Boas, J. P., & Fernandes, R. J. (2011). Biomechanical analysis of backstroke swimming starts. International Journal of Sports Medicine, 32, 546–551. Federation Internationale de Natation. (2011). SW 6 backstroke swimming rules. Retrieved May 3, 2011, from http://www.fina.org/H2O/index.php?option¼com_content&view¼article&id¼283:sw-6-backstroke&catid¼82: swimming-rules&Itemid¼184 Getchell, N., & Roberton, M. A. (1989). Whole body stiffness as a function of developmental level in children’s hopping. Development Psychology, 25, 920 –928. Hohmann, A., Fehr, U., Kirsten, R., & Krueger, T. (2008), Biomechanical analysis of the backstroke start technique in swimming. E-Journal Bewegung und Training, 2, 28–33. Holthe, M., & McLean, S. (2001). Kinematic comparison of grab and track starts in swimming. In J. R. Blackwell & R. H. Sanders (Eds.), XIXth International Symposium on Biomechanics in Sports (pp. 31–34). San Francisco, CA: International Society of Biomechanics in Sports. Hooper, S. (1981). Identification of the biomechanical factors affecting the performance of the backstroke swimming start. Brisbane: University of Queensland. Jorgic´, B., Puletic´, M., Stankovic´, R., Okicˇic´, T., Bubanj, S., & Bubanj, R. (2010). The kinematic analysis of the grab and track start in swimming. Physical Education and Sport, 8, 31–36. Kruger, T., Hohmann, A., Kirsten, R., & Wick, D. (2006). Kinematics and dynamics of the backstroke start. In J. P. Vilas-Boas, F. Alves, & A. Marques (Eds.), Xth International Symposium for Biomechanics and Medicine in Swimming (pp. 58– 60). Portugal: Portuguese Journal of Sport Sciences. Lyttle, A., & Benjanuvatra, N. (2005). Start right? A biomechanical review of dive start performance. Retrieved August 9, 2011, from http://www.coachesinfo.com/index.php?option¼com_content&view¼article&id¼89: swimming-start-style&catid¼49:swimming-coaching&Itemid¼86 Mason, B., & Cosser, J. (2000). What can we learn from competition analysis at the 1999 Pan Pacific swimming championships? In Y. Hong, D. P. Johns, & R. Sanders (Eds.), XVIIIth International Symposium on Biomechanics in Sports (pp. 75–82). Hong Kong: International Society of Biomechanics in Sports. Mason, B., Mackintosh, C., & Pease, D. (2012). The development of an analysis system to assist in the correction of inefficiencies in starts and turns for elite competitive swimming. In E. J. Bradshaw, A. Burnett, & P. A. Hume (Eds.), XXXth International Symposium on Biomechanics in Sports (pp. 249–252). Melbourne: International Society of Biomechanics in Sports. Pereira, S., Ruschel, C., & Araujo, L. (2006). Biomechanical analysis of the underwater phase in swimming starts. In J. P. Vilas-Boas, F. Alves, & A. Marques (Eds.), Xth International Symposium for Biomechanics and Medicine in Swimming (pp. 79– 81). Portugal: Portuguese Journal of Sport Sciences. Ranavolo, A., Don, R., Cacchio, A., Serrao, M., Paoloni, M., Mangone, M., & Santilli, V. (2008). Comparison between kinematic and kinetic methods for computing the vertical displacement of the centre of mass during human hopping at different frequencies. Journal of Applied Biomechanics, 24, 271–279. Ruschel, C., Araujo, L. G., Pereira, S. M., & Roesler, H. (2007). Kinematical analysis of the swimming start: Block, flight and underwater phases. In H. J. Menzel & M. H. Chagas (Eds.), XXVth International Symposium on Biomechanics in Sports (pp. 385–388). Brazil: International Society of Biomechanics in Sports. Swimming Australia. (2012). 2012 Australian swimming championships entry qualifying times. Retrieved March 1, 2012, from http://www.210.247.205.118/assets/console/document/documents/Aus%202012%20LC%20QT’s. pdf Vantorre, J., Seifert, L., Fernandes, R., Vilas-Boas, J., & Chollet, D. (2010). Biomechanical influence of start technique preference for elite track starters in front crawl. Open Sports Sciences Journal, 3, 137–139. Zar, J. H. (1972). Significance testing of the Spearman rank correlation. Journal of the American Statistical Association, 67, 578–580.
