Which Starting Style is Faster in Sprint Running -Standing or Crouch Start?
AKI SALO and IAN BEZODIS
The purpose of this study was to further understand the biomeehanical differences between the standing and crouch starting methods, unci to investigate
whether one of the Starting styles provides better acceleration and proves to he faster. Six university track team sprinters performed 2x3x50m trials. Digitised video, photocell timing, and velocity data revealed that during the first steps of the performance the standing start produced higher body centre of mass horizontal velocity than the crouch start. I’his may be due to the longer distance between the feet in the standing start, which caused longer push-ojl phases, and the work against gravity in the crouch Start. However, this advantage in horizontal velocity disappeared by the 10m mark, where similar velocities were recorded with both start styles. Further, there was no statistically significant difference between the two starting styles in horizontal velocity at the 25m mark, nor in the time to reach the 25m or 50m mink. Regarding relay running, where athletes need to decide to adopt either a crouch start without starting blocks or a standing start, there seems to be no specific reason for outgoing athletes to use a crouch start, although this area warrants further investigation.
Keywords: sprint start. 2-dimensional. digitising, photocell, relay running.
The start and a quick acceleration are important bases tor excellent performance in sprint running. A crouch start is compulsory for all events up to 400 m under International Association of Athletics Federations (IAAF) rule no. 161 (IAAF. 2002). However, athletes use the standing start in their practice sessions and there are other situations in which the standing start may well be beneficial to the athlete, such as the 4 x 100m relay. Only the first leg runner is permitted to use starting blocks while others need to decide whether to adopt a crouch start without blocks or a standing start.
The crouch start has been more widely studied from a biomeehanical point of view over the years (e.g. Henry. 1952; Baumann. 1976; Mero et at., 1983; Guissard and Duchateau, 1990: Mero and Komi. 1990: Guissard et al., 1992; Schot and Knutzen. 1992) than the standing start (e.g. Ostarello, 2001: Kraan et al… 2001). Only very tew studies have compared both starting techniques (Desipres. 1973: Gagnon. 1978). However, the problem still exists as to what biomeehanical variables contribute to a fast start technique (Harland and Steele. 1997).
Gagnon (1978) proposed that as the body centre of mass (CM) is closer to the start line in a standing start than in a crouch start, a natural running position could be attained almost immediately. The study found that skilled sprinters were 0.030s faster to 50m with a crouch start than an elongated standing start. However, unskilled sprinters were 0.043s faster when using the elongated standing technique (feet were 0.46m apart). Gagnon (1978) acknowledged that these data may be biased by the fact that the skilled sprinters were more familiar with the crouching technique. The study also found that the crouch start led to quicker block clearance with a reduced velocity, but at a greater acceleration during the block contact. Desipres (1973) studied 17 experienced sprinters over the first second of a sprint from crouch and standing starts, and concluded that the crouch start held an advantage over the standing start. Increased CM velocil} after one second and reduced time to cover a set distance were found in the crouch start compared to the standing start.
Regarding the crouch start, some authors have been interested in changing and controlling the block positions and/or the block angles. Henry (1952) found that different block distances revealed different block velocities (velocity at the end of block push-off) w ith the elongated start being the fastest. Schot and Knutzen 11992) revealed significant differences between the starting positions in the length of the first step and in the horizontal velocity at the end of the first step (elongated start yielding larger values than the bunch start). Guissard et al. (1992) reported that the block clearance velocity increased when the front block angulation was reduced, as this changed ankle and knee angles and consequently the muscle lengths at the calf. Jacobs and van Ingen Schenau ( 19921 showed further interest in joint angular data in sprint start and emphasised the correct liming and the amount of hip. knee and ankle joint flexions and extensions.
Interestingly, the crouch start was only introduced after athletes started to dig small holes to gain support for their feet, and later starling blocks were established. It seems that the athletes found the crouch start to be faster than the standing start, and thus adopted the crouching technique. However, Ostarello (2001 ) claimed that the standing start is faster than the crouch start, and that perhaps officials should reconsider the rule stipulating a crouch start. Ostarello’s (2001) Study, though, was a single subject study, although the author was adamant that the athlete’s background should have favoured a crouch start. Also, some other parts of the research design m the study warrant some caution. Nevertheless, the question still remains, which starting style is faster?
Anecdotal evidence about relay running shows that there is a considerable split between the athletes concerning which starting style to adopt in consecutive legs of the 4 x 100m relay at the elite level. Often, although not always, this split seems to be based on the nationality of the athletes. That is. it seems that the coach of the relay team and/or the national traditions in relay running dictate the preferred starting style. Outgoing athletes in the relay can run only
a maximum of 3()m before they must receive the baton. Thus, prompt and consistent acceleration to achieve as high a velocity as possible before receiving the baton is one of the key skills for the athletes. Consequently, any scientific indication of which starting style provides better acceleration is useful knowledge in deciding the starting style of relay runners.
Specificity of training is a generally accepted coaching principle tor improving performance, although scientific evidence of this in relation to sprinting is more physiology based (e.g. Dawson et al… 1998) than biomeehanical. As mentioned above, athletes are required to use a crouch start in competition, yet in training short sprints are often initialed with a standing start. Thus, further awareness of potential differences between the two starting styles will enhance coaching knowledge in sprint running. Therefore, the aim of this study was to further understand the biomeehanical differences between these two starting styles, and to investigate whether one of the starting styles provides better acceleration ami proves to be faster.
Six university track team sprinters (4 males. 2 females, aged 19.5 ±1.1 years) volunteered as subjects. The subjects personal best times for 100m were 10.98 ± 0.40s for men and 12.55 ± 0.35 s for women. After completing their own preferred warm-up. all subjects performed si\ maximal 50m sprints in two sets of different starting styles (three Standing starts and three crouch starts, or vice versa based on randnmh assigned groups). Subjects were allowed to use their own preferred block spacing for the crouch start, and to adopt a standing position with only feet in contact with the track (no blocks or hand contact > for the standing start. Normal starting commands with an electrical start sound signal were utilised. Recovery between the trials was approximately 4 minutes and between changing the starting style approximately 10 minutes.
Video data over a 5 m range covering the athlete in the set position and the first two steps of the sprint were recorded using a digital camera (Sony DCR-TRV 9001:. Sony Corporation. Japan) operating at 50 fields per second. The camera was set up perpendicular to the direction of motion, 1.0m in front of the start line. 30.0 m to the left side of the body. Photocells i New test Powertimer. Finland) were set at the 25m and 50m marks on the straight in order to measure subjects’ time taken to reach these points. The photocells were set at the approximated hip height in line with the recommendations by Ycadon et al. (1999) for a single beam system. The electrical starting signal initiated the photocell system. Additionally, subjects’ displacement during trials was measured using a sport laser gun (Laveg LDM 300 C. Jenoptik. Germany), positioned 20m behind the start line, at 1.5m above ground level. The target area lor the beam was the athletes’ lower back. Athletes" horizontal velocity at l()m and 25m was calculated from the positional data as an average velocity over outstep at these respective marks. The average velocity value instead of the instantaneous velocity was selected due to the fact that during the step the velocity of an athlete varies. Thus, it cannot be guaranteed that the instantaneous velocity at a specific mark is from the same step phase. For calculation, the local velocity minimum on either side of (he selected distance was found (Figure 1). It was assumed that these minimum values occurred very close lo the same phase of consecutive step cycles. The respective displacement was calculated and divided h\ the time taken lor this step to gain the average velocity value at this point of the run. Averaging in this way reduced the effect of potential measurement errors.
All videotape sequences lor each athlete were manually digitised using the Peak Motus system (Peak Performance Technologies. Inc., USA). Two-dimensional calibration was carried out using affine scaling. A calibration frame of 2.03m x 1.03m was positioned 0.52m above the track surface with the long sides parallel to the track, and w as \ ideotaped before the athletes" performances. An 18-point model of the human body was utilised, and inertia data were based on Zatsiorsky and Seluyanov (1983). The digitised points were the top of the head, the neck, the shoulders, elbows, wrists, tip of the middle fingers, hips, knees, ankles and the tip of the footwear (spikes). The operator visually evaluated and dijjitised the joint centres (and the top of the head, tip of the fingers and footwear) with a mouse from the computer screen. No joint markers were used, as the markers would not Stay perfectly in the viewed plane.
After reconstruction and Peak Motus embedded generalised cross-validatory quintie spline smoothing (Woltring. 1986). data of selected variables were collected. The step was defined to start at the first field of the video when the foot had made contact with the ground. The foot lea\ ing the ground (push-off) was defined as the first field of the video when the foot had lost the contact with the ground. Statistical tests were carried out using SPSS for Windows version 10.0. Repeated measures ANOVA tests with two within-subject factors (start style: crouch or standing, and trials: I, 2, 3) were performed. For reporting the variable values (in the results section), the mean and standard deviation in each style were calculated from the average of each athlete’s performance.
Although only six subjects were used in this study, several variables revealed statistically significant results. Regarding the set position, the CM location was higher (F = 44.5, p <.(X)I) in the standing start (0.71 ±0.04m) than in the crouch start (0.57 ±0.03m). Also, the distance between the feet was longer in the standing start than in the crouch start (0.56 ± 0.06m and 0.29 ± 0.03 m. respectively, F = 61.5. p <.00l). The time difference between the rear and front feet leaving the ground or blocks was longer in the standing start than in the crouch start (0.23 ± 0.00s and 0.21 ±0.02s, F = 15.3. p <.()5). The first stride length of the rear foot from its original position was longer in the standing stall compared with the crouch start (Table 1). This was also the case for the second step (contralateral feet).
Centre of mass horizontal velocity was statistically higher at the instant of rear foot push-off in the standing start than in the crouch start, as well as at the instant of front foot push-off (Figures 2 and 3). Regarding the vertical direction. CM velocity values were reversed, i.e. standing start revealed lower values at both feet push-off instants than the crouch start (0.10±0.21 m.s vs. 0.81 ±0.15m.s’, F = 38.5, p <.01 and 0.24±0.14m.s 1 vs. 0.42±0.25m.s F= 13.6. p <.05. respectively for the rear foot and front fool push-off instants). Horizontal velocity values remained higher in the standing start than in the crouch start alter the first and second contacts (Figures 4 and 5). although the difference was statistically significant only at the end of the first push-off
phase. I lori/ontal velocity change from the liisi step to the second step, though, was lower in the standing start than in the crouch start (0.7 ± 0.3 m.s-1 and 1.0 ± 0.2m.s■’. respectively). This difference was not statistically significant (Figure 6).
Athletes" horizontal velociti&S at M)m and 25 m were not statistically different between the two starling styles (Table I ). neither were the times taken to reach 25m and 50m. Regarding athletes’ technique, the maximum posterior knee angle of the rear foot during the push-off was larger in the standing start. However, this was not the case during front foot push-off. Maximum anterior
hip angle was larger during both rear foot and front loot push-off in the standing stall (Table I). Further, maximum knee and hip angular velocities yielded nonsignificant differences during each push-off phase (including the first 2 steps), as well as thigh segmental maximum angular velocities during each recovery phase of the running cycle.
Initially, it appears that the standing stail is the faster of the two starting methods studied in this paper. Centre of mass horizontal velocity was significantly larger at each push-off phase from the ground in the standing start than from the blocks in the crouch start (Figures 2 and 3). This trend was also found by Gagnon (1978) at the end of the front block contact. The statistically significant difference of the horizontal velocity continued at the end of the first step in favour of the standing start (Figure 4). However, this advantage was already disappearing between the first and second step. The crouch start style gained more velocity (although not statistically significantly - Figure 6) in this pail of the start than when using the standing start technique. Also, the horizontal velocity at the end of the second step yielded non-significant differences (Figure 5), although at the group level the velocity was still 0.5m.s-’ higher in the standing start than in the crouch start.
The advantage of the standing start was unquestionably lost by the 10m mark, when the crouch start trials had reached similar horizontal velocity to the trials of standing start. This finding is in line with Henry’s (1952) findings that elongated start with 26*’ block spacing provided the fastest block velocity, but the advantage was lost within the first ten yards. However, as the shortest outdoor sprint competition for adults is 100m, it is not relevant to look at the start in isolation. Thus, a further horizontal velocity value was measured at the 25m mark, as well as the time that Was taken from the start to 25 m and 50m. Timing was taken using a single beam system of photocells which can amplify errors, especially if the photocells are close to each other (Ycadon et <//… 1999). However, as the separations between the timing equipment were 25m. it was considered that potential errors had minimal effect on the results in this study. As neither the velocities at the 10 and 25m mark nor timings at 25 and 50m revealed statistical!) significant results, it can be concluded that neither starting style was shown to be better beyond the first step for this subject group. This contradicts Ostarello’s 120(11 i finding that the standing Start was faster than the crouch start. Based on the results in this study, there is no need to reconsider the start rule as suggested by Ostarello, especially when the false shirr detection systems are facilitated by the starting blocks.
However, it is interesting to consider why the standing start produced higher horizontal velocities during the initial phase of the start. The feet were further apart in the set position for the standing start than in the same situation for the crouch start (Table 1). This meant that the athletes pushed for longer with then front foot after the rear loot had left the ground. This supports observations by Henry (1952) who. although studying onlv block starts, found that the duration of front fOOt impulse as well as the velocity of leaving the blocks increased progressive!) based on the increased block spacing. The maximum knee angle at the rear foot push-off. partly due to initial position, was larger in the standing start, possibly providing a more favourable muscle length to produce forces for push-off. These differences (feet location in the set position, longer push-off phase from the front foot, and the larger maximum knee angle of the rear fool i resulted in a higher CM horizontal velocity at the end of each push-off phase for the standing start. Mero et al. (1983) concluded that the level of horizontal force during the block clearance was a more important factor than the time to produce it. Thus, if longer push produces higher force, this may be beneficial even considering the initial delay. Although force measurements were not taken in this study, it can be speculated that this finding by Mero et al. (1983) may also he \alid for the results between the two different starting styles in this study.
Another explanation of the higher horizontal values at the beginning of the start may be in relation to the finding by Cavagna el al. (1971). The study found that during the first second after the start from the blocks, the work against gravity was about 2ti 309! of the work aecessar) to move the body horizontally. In the current study, the centre of mass was lower in the set position of the crouch start (average 0.57m) than in the standing start (average 0.71 m). Thus, during the block clearance there were significantly higher CM vertical velocities at the end of each push-off phase than in pushing from the ground in the standing start. This shows that the work carried out against gravity in the crouch start can potentially explain the higher horizontal velocities in the standing start. However, it is still unclear why the favourable horizontal velocity in the standing start alter the first step does not continue further. This mechanism of transferring the initial velocities to further stages in running warrants further investigation.
It was not questioned at the time of \ ideotaping why the athletes hail longer distances between the feet in the standing start. This may be due to seeking better balance in the standing start, which is not an issue in the crouch start as both feet and hands create a steady base of support. It would be logical to think that the advantages gained in the standing start due to longer foot placement should be transferred to the block positions in the crouch start. As this was not the case in this studs, this area warrants further investigation. The difference between the athletes’ feet in the crouch start in this study were in full agreement with findings of Mero et al. (1983). showing that the distances were typical for sprinters, although longer spacings have also been reported (Baumann. 1976). Regarding the standing start, the mean feet difference of 0.56m was larger than in Gagnon’s (1978) study, in which the maximum spacing (0.46m) at the standing start produced better biomeehanical characteristics, e.g. higher velocity at takeoff. Differences in angular displacement values between the two different starting Styles were mainly due to different initial set positions, which meant that the knee and hip angles were much smaller in the set position of the crouch starts.
It is clear that both the starting styles were similar after the first two steps. The differences in joint angles and timings of movements at the initial push-off phases and at the first step merit the question, whether it is beneficial to repeatedly use standing starts in training for individual events. Any real effect of specificity, though, should be studied further in a controlled long-term training siuil} before making final judgement over this issue.
The time to run to 25 m and the velocity of athletes at the 25 m mark were selected to emulate relay running. In good 4 x 100m relay exchanges the outgoing athlete has run about 25 m when receiving the baton. The results in this study showed that the different starting techniques were very similar at that point of the trials. Further, it is worth considering that the crouch starts in this study were implemented using starting blocks, equipment which the outgoing athletes cannot use in the relax competition. There seems to be no specific reason, therefore, for outgoing athletes to use a crouch start in the relay. However, to receive further assurance on the matter, crouch starts without blocks should be studied, properly emulating the relay situation. It would also be worth replicating the study using elite athletes to ensure whether similar trends to this study are found for the specific subject group of high calibre athletes.
The authors are grateful to Mr Andreas Wallbaum and Mr Neil Be/oilis for their assistance during the data collection.