To swim as fast as possible, expert swimmers cover a high distance per cycle (i.e., the stroke length [SL]) (in m) while maintaining high stroke rate values (SR) (Hz) (7). It is now well established that the improvement in performance throughout a career is mainly explained by an increase in SL (7,8,11,20). Race analysis has underlined that SL is one of the most important factors to discriminate level of expertise whatever the type of swimming race (7,11,20). Consequently, coaches pay attention to the development of SL. The higher the SL, the higher the swimming economy because it reflects the ability of the swimmer to decrease drag (27) and create a great force production (13) and a high propulsive efficiency (8). Arellano (4) discriminate between two kinds of force: the “muscular force” which has to be transferred into “applied force” on water. The development of modern training in swimming must be oriented to improve applied force, which can only be developed by interacting directly with the water through the enhancement of a feeling for the water in a specific kinesthetic and tactile sense (4). To enhance SL, trainers usually design training sets in which speed (S, m·s−1) and the number of stroke cycles permitted per distance unit are fixed. More specifically, coaches can require the swimmers to change their usual SL-SR combination at a given S. A well−known example of such a task constraint could be one in which a given S has to be maintained with a lower SR than the one spontaneously adopted at this pace.
From a biomechanical point of view, this kind of task constrains the swimmer to alter his or her motor organization to propel him/herself at the same S. Indeed, aquatic displacement induces resistive impulse (i.e., the integral of drag with respect to time) for which drag magnitude depends notably on swimming S squared. The achievement and maintenance of a targeted S depends on the characteristics of the propulsive impulses yielded and the rate of their application (the SR value) (2,27). A decrease in the spontaneous SR at a given S requires the swimmer to both increase the force that can be applied efficiently on the water during each impulse and also to decrease the forward resistance magnitude so as to reduce the induced resistive impulse. To date, these acute changes in technical characteristics remain unknown. Knowledge on how a swimmer adapts his or her stroke technique to such constraints would help coaches chose with greater relevance the kind of task constraints they impose on their swimmers.
The dynamical theories of motor organization can help to formulate the changes that could be expected. Recent studies have shown the effect of SR and swimming S in the temporal structure of the stroke cycle (24,25,29). The higher the SR, the higher the relative time spent in the propulsive phases, irrespective of the change in S. On the contrary, the lower the S and the SR, the higher the time spent to glide and control the arm before propulsive action. According to these results, at constant S, a decrease in SR would lead to greater relative durations of the nonpropulsive phases and thus to a greater time between 2 consecutive propulsive impulses of both arms. To better understand the effect of a decrease in SR while maintaining an identical swimming S, a study of the absolute duration of stroke phases is needed because this would permit the duration of the generated impulses to be estimated and thus enable an assessment of the alterations in the swimmer's motor organization.
Thus, the aim of the study is to analyze the acute technical adaptations subsequent to a reduction in spontaneous SR at different swimming S. It is hypothesized that swimmers spend more time in the nonpropulsive phases when the stroke cycle duration is lengthened, whatever the swimming S. This could increase or induce a nonpropulsive time in between each propulsive arm action.
Experimental Approach to the Problem
To assess whether swimmers spend more time in the nonpropulsive phases to fulfil a reduction in spontaneous SR at different swimming S, the absolute duration of the stroke phases were measured. The obtained values were compared between paced exercises performed at a spontaneous SR value and at a reduced SR value.
Ten well-trained high school swimmers (20.3 ± 1.7 yr; 8 males and 2 females, national level 2 and 3) volunteered for this study. Mean values (±SD) of height, body mass, and arm span were 180.4 ± 6.3 cm and 169.8 ± 6.9 cm, 72.8 ± 5.3 kg and 61.5 ± 5.3 kg, and 187.2 ± 5.5 cm and 172.0 ± 7.3 cm for males and females, respectively. They had previously trained for 12 ± 2 years and underwent a frequency of 6 ± 2 training sessions per week during the time the study took place. Their performances in the 400-m front crawl stroke were 280.23 ± 13.73 seconds and 322.73 ± 12.47 seconds, which corresponds to a mean S that represents 76.8 ± 3.7% and 73.3 ± 2.7% of mean S of the short-course pool world record for men and women, respectively. Subjects were informed of the procedure and the potential risks and gave written consent to participate and then underwent a complete medical examination. The protocol was reviewed and approved by the consultative committee for the protection of human subjects in biomedical research of the Nord-Pas-de-Calais (France) before the start of the study. The experiment took place during a high-volume endurance training cycle, and therefore the subjects consequently adapted to the intensity and total volume of training.
The tests were performed in a 25-m indoor swimming pool using the front crawl. All swimmers started the trials in the water, with a push off from the wall. For each test, environmental conditions were constant (water temperature was fixed at 29°). All tests, with at least 24 hours of rest in between, occurred at the same time of the day (±2 hr) for each swimmer to minimize the possible effect of circadian rhythm on performance (5). No dietary restrictions were imposed during the experiment, with the exception of alcohol and caffeine, which were not authorized.
All swimmers started the study with a 400-m maximal swim of which the mean S (S400) was calculated. The mean S over this distance is known to be highly correlated to the maximal aerobic S in swimming (14,26). Subjects were then required to do 3 sets of 3 swims for a time to exhaustion (TTE) (in s) test at targeted speed that correspond to 95%, 100%, and 110% of the previously recorded S400 (TTE95, TTE100, and TTE110, respectively). These S were selected because they represent paces that are currently used to design training sets. These paces allow additional constraints such as an unusual number of stroke cycles per distance unit. Thirteen tests were performed, and the experiment lasted 3 weeks.
Speed was imposed by 2 operators walking on each edge of the pool at the prescribed pace. Marks 5 m apart were laid out, and the corresponding split times were provided to the operator. The swimmer was asked to keep his feet level with the pacer. When the operator's feet caught up with the swimmer's head, the test was stopped.
The swimmers were filmed in the sagittal plane during all tests with 2 cameras, 1 above the water surface (JVC Mini DV GR-DVX 407EG operating at 25 Hz and contained in a JVC WR-DV96 box) and 1 below the surface of the water (Sony Mini DV DCR-HC40E operating at 25 Hz and contained in a waterproof Sony HandyCam MPK-DVF6); the 2 were carried by a trolley and pushed by an experimenter.
First Set: Free Condition
For this set, (named “Free”), the swimming speed was set at 95%, 100%, and 110% of the 400 m mean speed and the swimmers were free to adopt their preferred stroke rate value and to change it. For each TTE test, times of stroke cycles(s) were measured cycle by cycle using PC software developed in our laboratory. At each entry of the swimmer's same hand in the water, an experimenter recorded a split time. All the split times recorded during 1 length were averaged to obtain 1 mean value of stroke cycle duration (TCycle) per 25 m. The values obtained per lap were averaged to obtain a mean SR value (in Hz) per test. The SR were derived from the split times as follows: SR = 1/TCycle. The values obtained for each TTE test were used for the second part of the testing procedure.
Second Set: Fixed Condition
For this set (named Fixed), the swimming S was set at 95%, 100%, and 110% of the 400-m mean S, and SR was set. Swimmers were required to repeat the set of tests previously described (TTE95, TTE100, and TTE110) with an imposed SR equal to their mean individual SR measured during the tests of the Free set. They were asked to maintain this SR as long as possible. The mean differences between the target SR and the actual SR were 0.01 ± 0.01 Hz, 0.02 ± 0.01 Hz, and 0.03 ± 0.01 Hz for TTE95, TTE100, and TTE110, respectively.
Third Set: Lowered Stroke Rate Condition
For this set (named Lowered), the swimming S was set at 95%, 100%, and 110% of the 400-m mean S, and a lower SR was imposed. The reduced SR was 95% of the SR in the Fixed set. Swimmers were asked to maintain this SR as long as possible. The mean differences between the target SR and the actual SR were 0.01 ± 0.01 Hz, 0.02 ± 0.01 Hz, and 0.03 ± 0.01 Hz for TTE95, TTE100, and TTE110, respectively. For Fixed and Lowered sets, the SR was imposed with a metronome (Tempo Trainer, Finis) placed in the swimmer's cap.
All the tests were conducted in a random order. The swimmers were asked to carry out their own warm-up, but this lasted only 20 minutes. They were also allowed to get accustomed to swimming at the prescribed S for a few lengths with the metronome. During each trial, the subjects were encouraged to perform as well as possible.
Time to Exhaustion Analysis
The duration of each TTE test (s) and the corresponding distance (in m) were measured. The time during which the operator's feet were catching up with the swimmer's head was subtracted from the total time to obtain the real duration of the TTE.
Stroking Parameter Analysis
A previous study has shown that the imposition of a double-constraint task (S and SR) does not induce any significant changes in the temporal structure of the arm stroke cycle (stroke phase duration and arm coordination) (1). Moreover, no significant difference has been detected for the studied variables between the beginning and the end of the TTE test for both sets. Therefore, the presented values were calculated according to the following process: for each swimmer, values measured at the beginning and the end of each TTE test were averaged, and a mean value from the experimental sample was calculated for each TTE.
The stroking parameters were measured continuously for each test with PC software, allowing the time per pool length to be measured. This time corresponded to split times between 2 consecutive contacts of the feet on the wall during each turn. The swimming S (in m·s−1) was then calculated. Finally, the SL (in m) was calculated as the ratio between S and SR.
Video analysis was performed to measured the duration (in s) of the 4 arm stroke phases that compose the arm stroke cycle:
Phase A: entry and catch. This phase corresponds to the time from the hand's entry into the water to the beginning of its backward movement and is considered a nonpropulsive phase.
Phase B: pull. This phase ends at the hand's arrival in a plane vertical to the shoulder. It is the first propulsive phase of the stroke cycle.
Phase C: push. This phase ends with the release of the hand from the water. It is the second propulsive phase.
Phase D: recovery. This phase ends with a new entry of the hand into the water and is considered a nonpropulsive phase.
Therefore, the mean duration of a complete arm movement, defined as the sum of the 4 distinct phases (A + B + C + D), was calculated.
Views of both cameras were postsynchronized with a visual signal that was visible in each footage. Videos analysis, performed by the same experimenters for the whole test, was performed using DartTrainer software (Version 4.0.5 DartFish TeamPro), allowing the measurement of each stroke phase duration (in s) with an accuracy of 0.02 seconds as each video was descreening.
Means and SDs were used to represent the average and the typical spread of values of the variables studied. The normal Gaussian distribution of the data was verified by the Shapiro-Wilks test. After verification of homoscedasticity by the Levene test, a paired t-test was used to compare the mean values of studied parameters between Fixed and Lowered sets for each TTE test.
Variations (in %) of the variables between sets Fixed and Lowered were calculated for each swimmer according to the following formula:
where Δ is the variation in the variable, and MFixed and MLowered are the mean values of a variable in the Fixed and Lowered sets, respectively. The Δ value for the SR was correlated to the Δ value of the other variables to detect any associations. When normality of the distribution was met, a Bravais-Pearson coefficient of correlation was used; otherwise, a Spearman coefficient of correlation was used. Only significant correlations observed for the 3 TTE tests will be discussed here. The threshold for significance was set at the 0.05 level of confidence. This statistical analysis was carried out by means of the STATISTICA 6.0 PC software.
Duration of Time to Exhaustion Test
The durations of each TTE test for both sets are presented in Table 1. Durations of TTE95 and TTE100 declined significantly from the Fixed set to the Lowered set (p < 0.05). Durations of TTE110 did not differ significantly between sets.
Arm Stroke Phases and Arm Coordination
All the variables studied concerning the temporal organization of the arm stroke cycle are presented in Table 2. The duration of the glide + catch phase (A) always increased significantly from the Fixed set to the Lowered set (p < 0.05). For TTE95 and TTE110, the lengthening in the stroke cycle duration was also made by a significant increase in the duration of the recovery phase (D) (p < 0.05). The durations of the pull (B) and push (C) phases did not vary significantly between sets (p < 0.05).
A significant negative correlation was detected between the reduction in SR value and the lengthening in the absolute duration of the glide + catch phase (A) for all TTE tests (r = −0.89, −0.96, and −0.94; p < 0.01 for TTE95, TTE100, and TTE110, respectively).
The aim of this study was to analyze the acute changes in the motor organization of the arm stroke cycle subsequent to a decrease in spontaneous SR during paced exercises. Because the magnitude of the drag the swimmer encounters depends on the swimming S, the achievement and maintenance of a targeted pace depends on the characteristics of the propulsive impulses yielded and the rate of their application (i.e., the SR). Such underlying motor processes are reflected in the individual stroke/length-stroke rate combination spontaneously used by the swimmer (23). Changes in this usual combination require modulation of both propulsive and resistive impulses and of the way they are coordinated. To fulfill to an acute increase in the stroke length, the strategy to modify the propulsive impulses could correspond to changes in the time over which the forces act and/or in the intensity of these forces (10,27).
The main results of the study were that the reduction in SR induced similar motor adaptations for all the tested S. The swimmers spent more time in the nonpropulsive phases, verifying our hypothesis. Moreover, the change in SR value was made primarily by a change in the glide + catch phase duration, as highlighted by the significant associations only observed between these 2 variables. Consequently, the consecutive propulsive actions of both arms became more distantly spaced in time for the TTE tests in the Lowered set.
The body decelerates between 2 consecutive propulsive impulses (9) because resistive impulses predominate (27). In the front crawl stroke, such body deceleration occurs when no propulsive impulse is directly generated by the arms (i.e., when the swimmer performs the glide + catch phase while the other arm recovers). Our results revealed an increase in the absolute duration of the glide + catch phase from the Fixed set to the Lowered set. Such a technical adaptation could lead to an increase in the resistive impulse, at least in its temporal dimension. Because no drag measurement was made, whether the resistive impulse actually increased remains a hypothesis. However, if no attention is paid to body alignment during the glide + catch phase, drag magnitude could increase, and the pace could not be reached and maintained. Coaches should control the way the swimmers perform this streamlined position during the lengthened glide + catch phases of each arm (with a longer time during which the hand is maintained forward or with greater body roll magnitude).
Our results revealed a stability in the durations of both pull and push phases for all the TTE tests from Fixed sets to Lowered sets. Given a decrease in the number of stroke cycles allowed per distance unit (i.e., a pool length for the Lowered set) for the same absolute S, the propulsive impulses generated per stroke cycle should be increased by a greater force application, as previously suggested by Craig and Pendergast (11). The swimmers could solve the problem of such a constraint by increasing the magnitude in force or the efficiency with which this force is applied. The question of how this propelling factor could be enhanced remains. In considering the relationship between the hand path and body roll (16,21,27), a first answer could be in a lengthening of the aquatic hand path thanks to a greater body roll, thus allowing for a better forward movement of the hand. If this is true, because the duration of the pull and push phases did not vary significantly, the hand S could be greater, which is believed to strongly influence the magnitude of force (30). A second answer is through an increase or a better orientation of the surface of body segments useful for propulsion. If this is so, the greater duration of the glide + catch phase could reflect the time spent in placing a body segment (notably the hand and forearm) in a more propulsive position (27), which would give an amplified “high elbow” position.
Our results showed decreases in the exercise durations from the Fixed set to the Lowered set, which is in line with the results of Nickleberry and Brooks (19), who found a significant decline in a TTE performed at 75% of maximal oxygen consumption (V̇O2max) at a lower cadence rate value than the spontaneous rate in cycling. The generation of propulsive impulses to counteract resistive ones induce a physiologic cost (27), which can be altered as swimmers, performing at a particular pace, use an unusual stroke/length-stroke rate combination. Changes in the physiologic cost of swimming could explain the decreases observed in the exercise durations from the Fixed set to the Lowered set. However, attempts to interpret such observations by analyzing the metabolic responses appear inconclusive. Whereas some authors showed that, as soon as the cyclical task is performed at a lower cadence than the spontaneous cadence, energy cost was significantly increased (3,17), others have not (6,15,31). Moreover, in swimming, a decrease in the spontaneous SR value at an imposed paced (not maintained until exhaustion) did not affect the blood lactate concentration or heart rate (22). The premature ending of the exercise for the tests of the Lowered set could be caused by an unusual muscular stimulation of the muscles directly involved in the propelling action (during the pull and push phases) and also indirectly (with muscles more activated during the streamline action during the glide + catch phase, with greater hypothetical body roll and stretching of the body) (10), but how this could affect the physiologic cost of swimming remains unanswered. Yet, because this double-task constraint is commonly used by trainers and because it reduces the swimmer's ability to swim at a determined S, the present results can be taken into account to better monitor the imposed training load. The use of Borg's rate of perceiving exertion scale could be relevant for this purpose, as is already done in cycling (12). Finally, the nonsignificant decline in TTE110 in the Lowered set could be related to the intensity at which the test was swum. Indeed, such an intensity implies a very high energy release that could not be significantly affected by such an added constraint.
The present study observed that a reduction in spontaneous SR value, for the same relative swimming S, induced a lengthening of the absolute duration of the nonpropulsive phases, whereas the pull and push phases were unaffected, whatever the swimming S tested. Such a double-task constraint appears to amplify the double mechanical constraint the swimmer has to solve to maintain the same S: the increase in magnitude and efficiency of the propulsive force and the improvement in body streamlining to limit an increase in resistive impulse between 2 consecutive arm propelling actions. This aspect has to be monitored by a coach's observation. The benefit of such a kind of exercise is offset by a significant decline in exercise duration, probably because of unusual muscular solicitation from a mechanical point of view, and this should be taken into account by coaches when they define the training load. The consecutive changes in spatial characteristics of the arm stroke cycle, as well as in the workload, appear as interesting to assess in the near future to better rationalize and include the impact of such task constraints in a training plan.
Our results showed that maintaining a certain S while using an SR lower than the spontaneous one will lead to several changes in motor organization that coaches should consider in a training design. Indeed, according to the technical adaptation sought, coaches could put emphasis on one particular aspect in the maximization of the propulsive impulse or in the minimization of the resistive impulse. Although providing the same general task constraint, the way the swimmer adapts his or her motor organization could correspond to different training aims.
If one wished to focus on the stroke cycle time devoted to the propulsive impulse, attention should primarily focus on the propulsive parameters such as form, surface orientation, and path of the arm. Coaches could insist on one of the above-mentioned factors to improve the efficiency with which force is applied to the water, according to the main technical aim of the training period. A decrease in the propulsive impulse allowed per distance unit obliges the swimmers to enhance the efficiency with which force is applied. Such task constraints could thus be useful to enhance feelings for the water in a specific kinesthetic and tactile sense (4), which is crucial to transfer the general muscular force into the so-called applied force on the water. Coaches should put emphasis on one of these aspects while applying the same general task constraint.
If one wishes to focus on minimizing drag, coaches could pay attention to the way the swimmer can decrease drag magnitude while increasing the glide + catch phase duration. A greater forward movement of the hand should be sought because this may decrease drag magnitude. This technical adaptation should be made while the swimmer keeps an adequate streamlining of the body to avoid an increase in the form drag.
It has previously been supposed that the glide + catch phase duration could be increased with a longer stand forward of the hand. Even if this kind of precise technical adaptation has been observed to put a main emphasis on body alignment, we do not support the idea that this kind of technical adaptation has to be stabilized because it could disturb the whole rhythm of the stroke. We advise coupling the studied task constraint with other learning tasks in which the stroke would also be changed.
Finally, because the capacity of the swimmer is reduced when he or she performs such a double-task constraint, the training load is also changed, and these changes have to be taken into account. The rate of perceiving exertion scale could be useful for this aim.
This study is the result of work supported with resources and the use of facilities at the Faculté des Sciences du Sport et de l'Education Physique, Laboratoire d'Etudes de la Motricité Humaine, Université de Lille, Ronchin, France. The authors thank all the participants for their enthusiastic cooperation and Odile Deray, who helped us conduct the experiment. The results of this study do not constitute endorsement by the NSCA.
1. Alberty, M, Potdevin, F, Dekerle, J, Pelayo, P, Gorce, P, and Sidney, M. Swimming techniques changes during time to exhaustion at freely chosen and controlled stroke rate. J Sports Sci
26: 1191-1200, 2008.
2. Alberty M, Sidney M, Pelayo P, and Toussaint H. Effect of fatigue on stroking characteristics in time to exhaustion test. Medicine & Science in Sports & Exercise.
41: 637-644, 2009.
3. Anderson, T. Biomechanics and running economy. Sports Med
22: 76-89, 1996.
4. Arellano R. Vortices and propulsion. 1999. Applied proceeding of the XVII International Symposium on Biomechanics in Sports, 53-66.
5. Atkinson, G and Reilly, T. Circadian variation in sports performance. Sport Med
21: 292-312, 1996.
6. Bessot, N, Nicolas, A, Moussay, S, Gauthier, A, Sesboüé, B, and Davenne, D. The effect of pedal rate and time of day on the time to exhaustion from high-intensity exercise. Chronobiol Int
23: 1009-1024, 2006.
7. Chollet, D, Pelayo, P, Delaplace, C, Tourny, C, and Sidney, M. Stroking characteristic variations in the 100-M freestyle for male swimmers of differing skill. Percept Mot Skills
85: 167-177, 1997.
8. Costill, DL, Kovaleski, J, Porter, D, Kirwan, J, Fielding, R, and King, DL. Energy expenditure during front crawl swimming: Predicting success in middle-distance events. Int J Sports Med
6: 266-270, 1985.
9. Counsilman, JE. Hand Speed and acceleration. Swim Tech
18: 22-26, 1981.
10. Craig, AB and Pendergast, DR. Relationships of stroke rate, distance per stroke and velocity in competitive swimming. Med Sci Sports Exerc
11: 178-283, 1979.
11. Craig, AB, Skehan, PL, Pawelczyk, JA, and Boomer, WL. Velocity, stroke rate, and distance per stroke during elite swimming competition. Med Sci Sports Exerc
17: 625-634, 1985.
12. Jameson, C and Ring, C. Contributions of local and central sensations to the perception of exertion during cycling: effects of work rate and cadence. J Sports Sci
18: 291-298, 2000.
13. Keskinen, KL, Tilli, LJ, and Komi, PV. Maximum velocity swimming: Interrelationship of stroking characteristics, force production and anthrometrics variables. Scand J Sports Sci
11: 87-92, 1989.
14. Lavoie, JM and Montpetit, RR. Applied physiology of swimming. Sports Med
3: 165-189, 1986.
15. Lepers, R, Millet, GY, Maffiuletti, NA, Hausswirth, C, and Brisswalter, J. Effect of pedalling rates on physiological response during endurance cycling. Eur J Appl Physiol
85: 392-395, 2001.
16. Liu, Q, Hay, JG, and Andrews, JG. Body roll and handpath in freestyle swimming: an experimental study. J Appl Biomech
9: 238-253, 1993.
17. Marais, G, Weissland, T, Robin, H, Vanvelcenaher, JM, Lavoie, JM, and Pelayo, P. Physiological effects of variations in spontaneously chosen crank rate during sub-maximal and supra-maximal upper body exercise. Int J Sports Med
20: 239-245, 1999.
18. Martin, PE and Marsh, AP. Step length and frequency effects on ground reaction forces during walking. J Biomech
25: 1237-1239, 1992.
19. Nickleberry, BR Jr and Brooks, GA. No effect of cycling experience on leg cycling ergometer efficiency. Med Sci Sports Exerc
28: 1396-1401, 1996.
20. Pai, Y-C, Hay, JG, and Wilson, BD. Stroking techniques of elite swimmers. J Sports Sci
2: 225-239, 1984.
21. Payton, CJ, Bartlett, RM, Baltzopoulos, V, and Coombs, R. Upper extremity kinematics and body roll during preferred-side breathing and breath-holding front crawl swimming. J Sport Sci
17: 65-70, 1999.
22. Pelayo, P, Alberty, M, Sidney, M, Potdevin, F, and Dekerle, J. Aerobic potential, stroke parameters, and coordination in swimming front-crawl performance. Int J Sports Physiol Perf
2: 345-357, 2007.
23. Pelayo, P, Sidney, M, and Weissland, T. Effects of variations in spontaneously chosen rate during crank upper-body and swimming exercise. J Human Mov Studies
33: 171-180, 1997.
24. Potdevin, F, Bril, B, Sidney, M, and Pelayo, P. Stroke frequency and arm coordination in front crawl swimming. Int J Sports Med
27: 193-198, 2006.
25. Potdevin F, Pelayo P, Delignières D, Dekerle J, Alberty M, and Sidney M. Does stroke rate determine swimming velocity values and coordination? In Chatard, JC, ed. Swimming Science IX
. University of Saint Etienne, France (press). 2003. pp. 163-168.
26 Rodriguez, FA. Maximal oxygen uptake and cardiorespiratory response to maximal 400-m free swimming, running and cycling tests in competitive swimmers. J Sports Med Phys Fitness
40, 87-95, 2000.
27. Sanders, R. New analysis procedures for giving feedback to swimming coaching and swimmers. In: Proceedings of the XXth International Symposium on Biomechanics in Sports Swimming
, 2002. pp. 1-14.
28. Seifert, L, Chollet, D, and Bardy, BG. Effect of swimming velocity on arm coordination in the front crawl: a dynamic analysis. J Sports Sci
22: 651-660, 2004.
29. Seifert, L, Chollet, D, and Rouard, A. Swimming constraints
and arm coordination. Hum Mov Sci
26: 68-86, 2007.
30. Toussaint, HM, de Hollander, AP, van den Berg, C, and Vorontsov, RA. Biomechanics of swimming. In: Exercise and Sport Science
. Garret, WE Jr and Kirkendall, DT, eds. Philadelphia: Lippincott Williams & Wilkins, 2000. pp. 639-659.
31. Watson, G and Swensen, T. Effects of altering pedal cadence on cycling time-trial performance. Int J Sports Med
27, 4: 296-300, 2006.