See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/6493059

The effects of resisted sprint training on acceleration performance and

kinematics in soccer, rugby union, and Australian football players

Article in The Journal of Strength and Conditioning Research · March 2007

DOI: 10.1519/R-18145.1 · Source: PubMed

CITATIONS READS

224 6,635

4 authors, including:

Aron Murphy Robert George Lockie

University of New England (Australia) California State University, Fullerton
110 PUBLICATIONS 9,128 CITATIONS 301 PUBLICATIONS 7,428 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Robert George Lockie on 07 October 2017.

The user has requested enhancement of the downloaded file.

Journal of Strength and Conditioning Research, 2007, 21(1), 77–85
䉷 2007 National Strength & Conditioning Association

THE EFFECTS OF RESISTED SPRINT TRAINING ON

ACCELERATION PERFORMANCE AND KINEMATICS IN
SOCCER, RUGBY UNION, AND AUSTRALIAN FOOTBALL
PLAYERS
CHRISTOPHER D. SPINKS,1 ARON J. MURPHY,2 WARWICK L. SPINKS,3 AND ROBERT G. LOCKIE2
1
Strength and Conditioning Department, New South Wales Institute of Sport, Sydney, New South Wales,
Australia; 2Human Performance Laboratory, University of Technology, Sydney, New South Wales, Australia;
3
Human Performance Laboratory, Institute of Sport and Exercise Science, James Cook University, Townsville,
Queensland, Australia.

ABSTRACT. Spinks, C.D., A.J. Murphy, W.L. Spinks, and R.G. formance is largely dependent upon the propulsive force
Lockie. The effects of resisted sprint training on acceleration provided by the extensors of the hip, knee, and foot (19).
performance and kinematics in soccer, rugby union, and Austra- Specific strength training activities are necessary to en-
lian football players. J. Strength Cond. Res. 21(1):77–85. 2007.— sure that the lower-leg musculature of football players
Acceleration is a significant feature of game-deciding situations
has the capacity to contribute to acceleration perfor-
in the various codes of football. However little is known about
the acceleration characteristics of football players, the effects of mance.
acceleration training, or the effectiveness of different training Improved strength levels allow for the production of
modalities. This study examined the effects of resisted sprint greater force and decreased ground contact time, leading
(RS) training (weighted sled towing) on acceleration perfor- to a possible increase in stride frequency. Increased stride
mance (0–15 m), leg power (countermovement jump [CMJ], 5- length may be achieved by improved utilization of elastic
bound test [5BT], and 50-cm drop jump [50DJ]), gait (foot con- energy during the support stage of the sprint cycle (28,
tact time, stride length, stride frequency, step length, and flight 32). Various training modalities, including sprint loading
time), and joint (shoulder, elbow, hip, and knee) kinematics in (e.g., weighted sled towing), have been used in an effort
men (N ⫽ 30) currently playing soccer, rugby union, or Austra- to increase muscular force output, especially at the hip,
lian football. Gait and kinematic measurements were derived knee, and ankle, leading to a potential increase in stride
from the first and second strides of an acceleration effort. Par-
ticipants were randomly assigned to 1 of 3 treatment conditions:
length.
(a) 8-week sprint training of two 1-h sessions·wk⫺1 plus RS train- Resisted sprint (RS) training in the form of weighted
ing (RS group, n ⫽ 10), (b) 8-week nonresisted sprint training sled towing is a training protocol often prescribed for foot-
program of two 1-h sessions·wk⫺1 (NRS group, n ⫽ 10), or (c) ball players in an effort to improve acceleration and
control (n ⫽ 10). The results indicated that an 8-week RS train- sprinting performance (11, 18). Although this mode of
ing program (a) significantly improves acceleration and leg pow- training is believed to increase lower-limb strength, there
er (CMJ and 5BT) performance but is no more effective than an are concerns that weighted sled towing may not transfer
8-week NRS training program, (b) significantly improves reac- to acceleration performance because of negative influenc-
tive strength (50DJ), and (c) has minimal impact on gait and es on acceleration kinematics (18, 22). There have been
upper- and lower-body kinematics during acceleration perfor- relatively few longitudinal studies that have examined
mance compared to an 8-week NRS training program. These
the effects of sprint training on sprint performance (6, 9),
findings suggest that RS training will not adversely affect ac-
celeration kinematics and gait. Although apparently no more ef- and fewer studies that have considered the effects of RS
fective than NRS training, this training modality provides an training on acceleration performance. Previous studies
overload stimulus to acceleration mechanics and recruitment of (18, 21) using RS training methods have examined the
the hip and knee extensors, resulting in greater application of effects of this training modality on maximum running
horizontal power. speed rather than on acceleration performance. The pur-
pose of this study was to determine the long-term effects
KEY WORDS. stride length, stride frequency, strength, towing of RS sprint training in the form of weighted sled towing
on acceleration performance, leg power, and acceleration
kinematics. It was hypothesized that (a) RS training
INTRODUCTION would improve acceleration performance in trained foot-
cceleration is integral to successful perfor- ballers to a greater extent than nonresisted sprint (NRS)

A mance in the various football codes, including

Australian rules, rugby union, and soccer (4,
10, 12), and is potentially decisive in deter-
mining the outcome of a game (7, 37). Al-
though speed is very important in most game situations
training, and (b) RS training would result in changes in
the biomechanical characteristics of acceleration perfor-
mance in trained football players. This study was de-
signed to provide coaches, athletes, and trainers with new
knowledge that could be used in the application of sled
in football, acceleration is of relatively greater impor- towing for RS training purposes.
tance, as players cover only short distances (approxi-
mately 20 m) at maximal effort (10, 35). Thus, training METHODS
for acceleration is an essential component of football con- Experimental Approach to the Problem
ditioning programs (8, 35), involving alteration of the ath- A longitudinal study was used to examine the effects of
letes’ stride length and stride frequency. Acceleration per- weighted sled towing on the acceleration performance

77
78 SPINKS, MURPHY, SPINKS ET AL.

TABLE 1. Sprint training program for the resisted sprint (RS, n ⫽ 10) and nonresisted sprint (NRS, n ⫽ 10) groups.*
Week Distance (m) Repetitions Sets Total distance (m) Recovery
1 0–5 4 2 40 45 s
0–10 4 1 40 W/B
0–15 5 1 75 W/B
0–20 3 1 60 (215 m) W/B (1 min B/W)
2 0–5 6 2 60 45 s
0–10 3 2 60 60 s
0–15 3 1 45 W/B
0–20 3 1 60 (225 m) W/B (1.5 min B/W)
3 0–5 5 3 75 60 s
0–10 4 2 80 45 s
0–15 3 1 45 W/B
0–20 3 1 60 (260 m) W/B (2 min B/W)
4 0–5 5 3 75 45 s
0–10 4 2 80 60 s
0–15 3 1 45 W/B
0–20 3 1 60 (260 m) W/B (1.5 min B/W)
5 0–5 6 2 60 45 s
0–10 3 2 60 60 s
0–15 3 1 45 W/B
0–20 3 1 60 (225 m) W/B (1.5 min B/W)
6 0–5 5 3 75 60 s
0–10 4 2 80 45 s
0–15 3 1 45 W/B
0–20 3 1 60 (260 m) W/B (2 min B/W)
7 0–5 5 3 75 60 s
0–10 4 2 80 45 s
0–15 3 1 45 W/B
0–20 3 1 60 (260 m) W/B (2 min B/W)
8 0–5 5 2 50 45 s
0–10 4 2 80 60 s
0–15 3 2 90 90 s
0–20 3 2 120 (340 m) 120 s
* Number in parentheses ⫽ weekly total distance in meters. W/B ⫽ walk back recovery period; B/W ⫽ between sets.

and kinematics of trained football players. The research (SSC) capabilities; (d) reactive strength index; (e) gait
design was a multiple group pretest-posttest control characteristics; and (f) kinematic characteristics of accel-
group design with the participants matched for 15-m eration performance.
sprint performance and then randomly assigned to 1 of 3
treatment conditions: (a) sprint training plus RS (weight- Subjects
ed sled towing) training (RS group), (b) NRS training Thirty healthy men (age, 21.8 ⫾ 4.2 years; height, 181.9
(NRS group), and (c) control. The RS and NRS groups ⫾ 6.2 cm; mass, 83.3 ⫾ 8.7 kg) volunteered and gave writ-
undertook an 8-week sprint training program of two 1-h ten informed consent to participate in this study. Partic-
sessions·wk⫺1 (Table 1). As is typical of an intervention ipants were recruited if they (a) had a strength training
study involving athletes in training, logistical factors, history (ⱖ2 times·wk⫺1) extending over the previous 12
namely periodization of sprint training, influenced the months, (b) were currently strength training, and (c) were
length of the intervention. There is some limitation on the currently participating in a code of football, namely soccer
potential for morphological changes in 16 training ses- (n ⫽ 8), rugby union (n ⫽ 12), or Australian football (n ⫽
sions conducted over an 8-week period. However, there is 10), at first-grade level. The participants were required
evidence that neurological changes can occur in periods to have a strength training history and to be currently
of 4–6 weeks of resistance training and that these have strength training in order to reduce the potential for in-
the potential to affect muscle morphology (40). jury and to prevent delayed onset muscle soreness be-
The control group did not engage in any training pro- cause of the dynamic nature of the intervention and test-
gram, completing only the pretests and posttests. The lit- ing protocols. Prior to involvement in the study, all par-
erature (22) has indicated that a desirable load for RS ticipants had completed or were near completion of pre-
training is equal to a 10% reduction in maximum running season training and competition. During involvement in
velocity. Because the participants were required to run the study, all participants were instructed to continue
over a short distance (15 m), maximum running velocity with their normal training programs, which consisted of
was not attainable. The load required to reduce running at least 2 football training and 2 strength and fitness
velocity by approximately 10% without measuring maxi- training sessions·wk⫺1 and at least 1 football game·wk⫺1.
mum running velocity was determined from the velocity The study was approved by the University of Technology,
that resulted from towing a certain load established as a Sydney, Human Research Ethics Committee.
function of body weight using the regression equation de-
veloped by Lockie et al. (22). The effects of the interven- Procedures
tions were determined by analysis of (a) acceleration per- Following completion of a preparticipation questionnaire
formance over 0–5 m, 5–10 m, 10–15 m, and 0–15 m; (b) and determination of age, height, and mass, each partic-
lower-body power; (c) horizontal stretch-shortening cycle ipant undertook the same warm-up routine consisting of
 RESISTED SPRINT TRAINING AND ACCELERATION 79

10 minutes of continuous cycling and 10 minutes of static

and dynamic stretches of the muscles of the lower limbs
(hamstrings, gluteals, and quadriceps). The dynamic
warm-up consisted of athletics drills including fast foot
strikes, ankle hops, high knee skips, and heel pickup
runs. A series of warm-up runs at increasing speeds (60,
70, 80, 90, and 100% of perceived maximum effort) over
15 m was then performed. A similar cool-down routine
was conducted after the sprint tests. The participants
then completed 2 trials of 15-m maximal sprint efforts
without resistance with a walk-back recovery (1.5 min-
utes) between each trial. All testing procedures were con-
ducted on an indoor basketball court (timber surface).
The running velocity and acceleration were recorded
using a velocimeter (Onspot, Wollongong, Australia),
which consisted of a nylon wire connected to the partici-
pant’s shorts that unwound over a precisely machined
wheel as the subject moved forward. An optical sensor
mounted over the wheel sent impulses to the processor
for every 0.1 m of path length, and 5-m interval split FIGURE 1. Joint angle conventions. TL ⫽ trunk lean; S ⫽
times (0–5 m, 5–10 m, 10–15 m, and 0–15 m) were re- shoulder; E ⫽ elbow; H ⫽ hip; K ⫽ knee.
corded by an attached stopwatch (Seiko, Tokyo, Japan).
The split times were used to calculate the velocity for that
interval. The velocities recorded for both trials were av- determine stride length, stride frequency, first- and sec-
eraged and recorded as the test score. The accuracy of the ond-step flight times, and first- and second-step contact
velocity measures was determined by matching the velo- times. For the purposes of this study, step length was
cimeter split times with split times recorded by a series taken as the distance between the initial ground contact
of timing lights (Swift Technologies, Lismore, Australia). of 1 foot to the initial ground contact of the other foot.
A force platform (Onspot) was then used to measure Stride length was calculated as the distance from succes-
lower-body power via countermovement jumps (CMJs) sive initial ground contacts of the same foot.
and reactive strength index via 50-cm drop jumps Following pretesting, the data obtained from the 15-
(50DJs). Jump height was calculated from the force plat- m sprint assessment was used to rank participants from
form data using standard projectile motion equations. 1 to 30. The fastest participant was matched with the
The force data was sampled at 1,000 Hz. A 5-bound test next 2 fastest participants, and so on. The participants in
(5BT) was used to measure horizontal SSC capabilities in each triplet were then randomly assigned (using a table
distance traveled. Participants undertook 3 trials for the of random numbers) to 1 of the 3 treatment conditions:
CMJ, 50DJ, and 5BT, with the best trial used for analy- (a) RS group, (b) NRS group, or (c) control. The RS and
sis. A rest period of 1 minute was allowed between trials. NRS groups undertook an 8-week sprint training pro-
A 10-minute recovery period was allowed between the gram of two 1-h sessions·wk⫺1 (Table 1). The control
various test protocols. group did not engage in any training program, completing
Prior to all sprint testing, major joint segments were only the pretests and posttests. During the 8-week train-
identified using reflective markers placed on 8 landmarks ing period, the RS and NRS group participants undertook
on the right-hand side of the body. The landmarks were the training protocol on 2 nonconsecutive days·wk⫺1. The
the acromion (shoulder), lateral epicondyle of ulna (el- sprint training program was designed to mimic a series
bow), midpoint between the styloid processes of the ra- of acceleration efforts over varying short distances. Elite
dius and ulna (wrist), greater trochanter of femur (hip1), sprinters reach maximum running velocity at 40–50 m,
anterior superior iliac spine (hip2), lateral epicondyle of whereas intermediate sprinters reach maximum running
femur (knee), lateral malleolus of fibula (ankle) and toe velocity at 20–30 m (13). The aim of the current study
(5th metatarsal). Hip2 was used to measure horizontal was to investigate the acceleration phase of sprint run-
hip velocity, and this velocity was used to determine hor- ning in trained football players; therefore, the training
izontal running speed, as a high horizontal hip velocity consisted of a series of maximum sprint efforts over typ-
indicates a good sprint performance (30). Hip1 was used ical acceleration distances of 5, 10, 15, and 20 m. The
to determine joint angles. sprint training protocol was changed periodically to facil-
Sprint efforts were recorded on 2 motion analysis sys- itate a progressive overload in order to increase the train-
tems, each set at a frame rate of 100 Hz and placed per- ing stimulus. The progressive overload was determined
pendicular to the participant’s line of motion. Digital im- via total distance traveled for each session, with the total
agery was obtained via 2 infrared cameras (Qualysis, distance increasing each week (every 2 sessions) for the
Sävedalen, Sweden). From these data, horizontal hip ve- 8-week duration (Table 1).
locity at takeoff into the second stride, trunk lean at The warm-up and cool-down routines used for the
touchdown of the first stride, maximum extension and sprint tests were also utilized before and after each train-
flexion, range of motion (ROM), and average angular ve- ing session, with the addition of lower-limb power exer-
locity at the shoulder, elbow, hip, and knee joints for the cises in the form of low-level plyometric activities (squat
first and second strides were measured. The first 2 strides jump, 30- and 50-cm box jumps, bounding, traveling
were chosen for evaluation because the initial accelera- squat jumps, CMJ) to the dynamic warm-up, in order to
tion features the greatest changes in speed. The joint an- reduce the risk of injury and to facilitate a specific warm-
gle conventions are shown in Figure 1. A high-speed video up for the ensuing sprint training session. All training
camera (JVC-DV 9800; JVC, Tokyo, Japan) was used to sessions were supervised to ensure that the participants
80 SPINKS, MURPHY, SPINKS ET AL.

TABLE 2. Mean ⫾ SD horizontal hip velocities for all distances.

Resisted Nonresisted Control
Test (n ⫽ 10) (n ⫽ 10) (n ⫽ 10)
0–5 m (m·s⫺1) Pretest 3.62 ⫾ 0.25 3.51 ⫾ 0.32 3.55 ⫾ 0.20
Posttest 3.95 ⫾ 0.30* 3.79 ⫾ 0.29* 3.69 ⫾ 0.19
5–10 m (m·s⫺1) Pretest 6.41 ⫾ 0.23 6.39 ⫾ 0.32 6.43 ⫾ 0.26
Posttest 6.80 ⫾ 0.29*† 6.67 ⫾ 0.32*† 6.47 ⫾ 0.29
10–15 m (m·s⫺1) Pretest 7.19 ⫾ 0.38 6.99 ⫾ 0.33 7.20 ⫾ 0.31
Posttest 7.72 ⫾ 0.34*‡ 7.38 ⫾ 0.27*‡ 7.28 ⫾ 0.30
0–15 m (m·s⫺1) Pretest 5.25 ⫾ 0.26 5.13 ⫾ 0.33 5.21 ⫾ 0.18
Posttest 5.66 ⫾ 0.32*† 5.45 ⫾ 0.29*† 5.32 ⫾ 0.21
* p ⬍ 0.001 significant within-subjects main effect for time of test (pretraining-posttraining)
† p ⬍ 0.05 significant between-subjects interaction effect (posttraining).
‡ p ⬍ 0.01 siginificant within-subjects interaction effect for time of test (pretraining-posttraining).

TABLE 3. Mean ⫾ SD lower-body power (CMJ), horizontal stretch-shortening cycle capabilities (5BT) and reactive strength index
(50DJ).*
Resisted Nonresisted Control
Test (n ⫽ 10) (n ⫽ 10) (n ⫽ 10)
CMJ (cm) Pretest 37.4 ⫾ 4.4 35.3 ⫾ 3.7 36.2 ⫾ 3.9
Posttest 39.6 ⫾ 4.2† 38.5 ⫾ 3.7† 38.6 ⫾ 3.6
5BT (m) Pretest 11.4 ⫾ 0.7 10.7 ⫾ 0.8 11.2 ⫾ 0.8
Posttest 12.2 ⫾ 0.9† 11.6 ⫾ 0.8† 11.5 ⫾ 0.8
50DJ (cm) Pretest 117.2 ⫾ 27.1 112.2 ⫾ 18.9 127.2 ⫾ 36.2
Posttest 137.3 ⫾ 27.7‡ 121.8 ⫾ 21.3 128.7 ⫾ 25.1
* CMJ ⫽ countermovement jump; 50DJ ⫽ 50 cm drop jump; 5BT ⫽ 5-bound test.
† p ⬍ 0.001 significant within-subjects main effect for time of test (pretraining-posttraining).
‡ p ⬍ 0.05 significant within-subjects main effects for time of test (pretraining-posttraining).

completed the 8-week training programs. No participants assumptions were considered, with the Levene statistic
were injured during the 8-week intervention program. used to determine homogeneity of variance. Post hoc tests
The external resistance applied to the RS group was were performed using the Tukey honestly significant dif-
determined in 2 parts. First, the load required as a per- ference procedure. A power value of 0.8 and an alpha lev-
centage of body mass was determined using the equation el of 0.05 were adopted for this study. Intraclass corre-
%body mass ⫽ (⫺1.96 ⫺ %velocity) ⫹ 188.99 lation coefficients were performed on the mean scores of
each group for each dependent variable to determine test-
where %velocity ⫽ the required training velocity as a per- retest reliability.
centage of the maximum velocity, e.g., 90% of maximum.
Second, the sled load required to produce a certain RESULTS
velocity was determined using the following formula:
Horizontal Hip Velocity
load ⫽ ([body mass ⫻ %body mass] ⫺ sled weight)
where %body mass ⫽ the solution to the first equation Horizontal hip velocity was significantly higher at all dis-
expressed as a decimal (e.g., 5% body mass ⫽ 0.05) and tance intervals for the RS and NRS groups posttraining
sled weight ⫽ 4 kg. than pretraining (0–5 m, p ⬍ 0.001; 5–10 m, p ⬍ 0.001;
Once the load had been determined, it remained con- 10–15 m, p ⬍ 0.001). However, there was no significant
stant for all RS group participants for the duration of the between-subjects effect for horizontal hip velocity for
8-week training period. these groups (Table 2). There were significant interaction
effects (time of test ⫻ group) for the velocity intervals 5–
Statistical Analyses 10 m (p ⬍ 0.01), 10–15 m (p ⬍ 0.01), and 0–15 m (p ⬍
Statistical analyses were performed using SPSS for Win- 0.01). Post hoc interaction effect analysis indicated that
dows (version 11.0; SPSS, Inc., Chicago, IL). Descriptive the RS group participants were significantly faster than
statistics were expressed as mean (⫾ SD). Data were an- the control group participants for the posttest velocity
alyzed via a 2-factor mixed factorial analysis of variance values 5–10 m (p ⬍ 0.05), 10–15 m (p ⬍ 0.01), and 0–15
(ANOVA). Three levels corresponding to the training ac- m (p ⬍ 0.05).
tivities of the 3 groups (i.e., RS training, NRS training,
and no sprint training) were specified as the between- Lower-Body Power, Reactive Strength Index, and
subjects factor (group). The within-subjects factor (time of Horizontal Stretch-Shortening Cycle Capability
test) represented the pretraining and posttraining mea-
sures. Because only 2 repeated measures were employed, Lower-body power (CMJ, p ⬍ 0.001) and horizontal SSC
the assumption of sphericity, determined by Mauchly’s performance (5BT, p ⬍ 0.001) were significantly higher
test of sphericity for homogeneity of covariance, was not posttraining for the RS and NRS groups. However, there
applicable, as there was only 1 correlation coefficient (the were no significant between-subjects effects for these
correlation between pretest and posttest measures) that groups. Reactive strength (50DJ) was significantly higher
could be calculated. All other repeated measures ANOVA (p ⬍ 0.05) posttraining for the RS group only (Table 3).
 RESISTED SPRINT TRAINING AND ACCELERATION 81

TABLE 4. Mean ⫾ SD pretraining-posttraining gait values.*

Resisted Nonresisted Control
Test (n ⫽ 10) (n ⫽ 10) (n ⫽ 10)
CTR (s) Pretest 0.17 ⫾ 0.02 0.16 ⫾ 0.01 0.16 ⫾ 0.02
Posttest 0.15 ⫾ 0.02†‡ 0.15 ⫾ 0.01† 0.16 ⫾ 0.01
CTL (s) Pretest 0.18 ⫾ 0.02 0.20 ⫾ 0.02 0.19 ⫾ 0.03
Posttest 0.18 ⫾ 0.02 0.19 ⫾ 0.02 0.19 ⫾ 0.02
SL (m) Pretest 2.15 ⫾ 0.4 2.03 ⫾ 0.3 2.01 ⫾ 0.2
Posttest 1.96 ⫾ 0.4 2.01 ⫾ 0.1 2.09 ⫾ 0.1
SF (Hz) Pretest 1.90 ⫾ 0.3 1.88 ⫾ 0.2 1.94 ⫾ 0.1
Posttest 2.10 ⫾ 0.5 1.94 ⫾ 0.1 1.89 ⫾ 0.1
StL1 (m) Pretest 0.99 ⫾ 0.1 0.98 ⫾ 0.1 1.00 ⫾ 0.2
Posttest 0.98 ⫾ 0.1 0.95 ⫾ 0.1 0.97 ⫾ 0.1
StL2 (m) Pretest 1.01 ⫾ 0.1 1.04 ⫾ 0.1 1.07 ⫾ 0.3
Posttest 1.07 ⫾ 0.2 1.03 ⫾ 0.1 1.04 ⫾ 0.2
FT1 (s) Pretest 0.08 ⫾ 0.02 0.05 ⫾ 0.02 0.06 ⫾ 0.02
Posttest 0.06 ⫾ 0.02 0.06 ⫾ 0.02 0.07 ⫾ 0.02
FT2 (s) Pretest 0.09 ⫾ 0.02 0.09 ⫾ 0.02 0.08 ⫾ 0.01
Posttest 0.09 ⫾ 0.02 0.08 ⫾ 0.02 0.08 ⫾ 0.01
* CTR ⫽ contact time, right foot; CTL ⫽ contact time, left foot; SL ⫽ stride length; SF ⫽ step frequency; StL1 ⫽ first step length;
StL2 ⫽ second-step length, FT1 ⫽ first-step flight time; FT2 ⫽ second-step flight time.
† p ⬍ 0.01 significant main effect for time (pretraining-posttraining).
‡ p ⬍ 0.01 significant interaction effect for time (pretraining-posttraining).

Gait Values step 1 and step 2 knee angular velocity and a significant
interaction effect for step 2 knee angular velocity. Post
The RS training program had no significant effect on
hoc analysis of the interaction effect indicated that post-
stride length, left-foot contact time, steps 1 and 2 flight
test step 2 knee angular velocity for the control group was
time, stride frequency, or steps 1 and 2 step length. There
significantly slower than for the RS and NRS groups. Post
was a significant (p ⬍ 0.01) pretest to posttest decrease
hoc between-subjects analysis indicated that steps 1 and
in right-foot (second-step) contact time for the RS and
2 knee angular velocities for the control group partici-
NRS groups (Table 4). However, there was no significant
pants were significantly slower than those of the RS
between-subjects effect for right-foot contact time. There
group participants (Table 6). Because of the presence of
was a significant (p ⬍ 0.01) interaction effect for right-
a significant Levene statistic (indicating lack of homoge-
foot contact time. Post hoc interaction effect analysis in-
neity of variance), the posttest step 1 and step 2 knee
dicated that right-foot contact time for the RS group de-
angular velocity measures should be interpreted with
creased significantly from pretest to posttest.
caution.
Upper-Body Kinematics Test-retest intraclass correlation coefficients were
above 0.90 for all dependent variables except for trunk
Control group subjects demonstrated significantly re- lean (ICC R ⫽ 0.81).
duced step 1 shoulder flexion (p ⬍ 0.05), step 1 shoulder
extension (p ⬍ 0.001), step 2 shoulder extension (p ⬍ DISCUSSION
0.001), and step 1 (p ⬍ 0.05), and step 2 (p ⬍ 0.01) shoul-
der ROM pretest to posttest. Post hoc interaction effect Resisted sprint training in the form of weighted sled tow-
analysis indicated significantly (p ⬍ 0.01) greater posttest ing is a training protocol often prescribed for football
step 1 shoulder extension values for the RS group com- players in an effort to improve acceleration and sprinting
pared to the NRS group, significant (p ⬍ 0.01) pretest performance (11, 18). Although this mode of training is
differences in step 2 shoulder extension for the NRS and believed to increase lower-limb strength, there are con-
control groups, and a significant (p ⬍ 0.05) pretest to cerns that weighted sled towing may not transfer to ac-
posttest increase in step 1 shoulder angular velocity for celeration performance because of negative influences on
the RS group. Post hoc between-subjects analysis indi- acceleration kinematics (18). Therefore, the purpose of
cated that the NRS group subjects had a significantly (p this study was to determine whether RS training in the
⬍ 0.05) larger step 1 elbow flexion than the RS and con- form of weighted sled towing is more beneficial in the
trol group subjects. Trunk lean was significantly (p ⬍ development of acceleration performance in trained foot-
0.001) increased for the RS and NRS groups. Post hoc ballers than NRS training. Also of interest were the long-
analysis indicated that the RS group had a significantly term effects of RS training on the kinematics of acceler-
(p ⬍ 0.01) greater posttest trunk lean than the control ation performance in trained football players.
group (Table 5). The sprint velocities recorded for all intervals (0–5, 5–
10, 10–15, and 0–15 m) for both experimental groups
Lower-Body Kinematics showed a significant improvement from pretraining to
There were no significant within- or between-subjects ef- posttraining (Table 2). However, there were no significant
fects for steps 1 and 2 hip extension, steps 1 and 2 hip differences between the 2 training groups, indicating that
ROM, steps 1 and 2 hip angular velocity, or step 1 knee RS training is no more effective than NRS training in
ROM. However, the control group demonstrated a signif- developing acceleration performance. The RS and NRS
icant pretest to posttest decrease in step 2 knee ROM groups improved their 0–5-m running velocity by 9% and
compared to the RS group. 8% respectively. For the 5–10-m interval, the RS group
There were significant between-subjects effects for improved running velocity by 6%, whereas the NRS group
82 SPINKS, MURPHY, SPINKS ET AL.

TABLE 5. Mean ⫾ SD upper-body kinematic variables pretraining-posttraining for steps 1 (S1) and 2 (S2).*
Resisted Nonresisted Control
Test (n ⫽ 10) (n ⫽ 10) (n ⫽ 10)
Shoulder: S1 flexion (⬚) Pretest 22.9 ⫾ 7.2 23.7 ⫾ 4.7 29.1 ⫾ 5.9
Posttest 24.2 ⫾ 5.6 20.7 ⫾ 3.0 24.3 ⫾ 3.9†
Shoulder: S2 flexion (⬚) Pretest 21.8 ⫾ 7.2 22.1 ⫾ 3.3 26.4 ⫾ 4.9
Posttest 24.2 ⫾ 3.4 22.1 ⫾ 2.3 21.4 ⫾ 3.8‡
Shoulder: S1 extension (⬚) Pretest 78.9 ⫾ 12.4 69.2 ⫾ 8.9 81.6 ⫾ 10.8
Posttest 76.5 ⫾ 8.0‡ 65.1 ⫾ 13.3 68.7 ⫾ 8.3§
Shoulder: S2 extension (⬚) Pretest 76.9 ⫾ 12.7 68.5 ⫾ 7.8‡ 81.4 ⫾ 9.6‡
Posttest 74.0 ⫾ 7.4 67.1 ⫾ 10.4 68.9 ⫾ 8.6§
Shoulder: S1 ROM (⬚) Pretest 56.1 ⫾ 13.9 45.6 ⫾ 9.2 52.5 ⫾ 13.4
Posttest 52.4 ⫾ 11.0 44.4 ⫾ 12.0 44.3 ⫾ 8.6†
Shoulder: S2 ROM (⬚) Pretest 55.0 ⫾ 13.3 46.4 ⫾ 8.3 55.0 ⫾ 11.5
Posttest 49.8 ⫾ 7.3 44.9 ⫾ 10.8 47.5 ⫾ 10.1㛳
Shoulder: S1 ang vel (⬚·s⫺1) Pretest 402.2 ⫾ 92.0 325.1 ⫾ 53.7 364.7 ⫾ 81.4
Posttest 396.9 ⫾ 83.6† 398.4 ⫾ 70.2 362.1 ⫾ 56.8
Shoulder: S2 ang vel (⬚·s⫺1) Pretest 420.5 ⫾ 114.9 375.2 ⫾ 54.1 427.2 ⫾ 39.8
Posttest 393.5 ⫾ 109.5 396.1 ⫾ 83.5 378.7 ⫾ 65.0
Elbow: S1 flexion (⬚) Pretest 68.4 ⫾ 15.5 82.5 ⫾ 17.4 72.2 ⫾ 13.5
Posttest 69.5 ⫾ 15.8 89.8 ⫾ 24.9# 69.2 ⫾ 11.9
Elbow: S2 flexion (⬚) Pretest 70.4 ⫾ 16.1 79.9 ⫾ 14.3 67.8 ⫾ 12.7
Posttest 68.8 ⫾ 17.6 85.3 ⫾ 23.3 68.6 ⫾ 12.3
Elbow: S1 extension (⬚) Pretest 139.2 ⫾ 10.0 136.4 ⫾ 11.7 138.4 ⫾ 15.7
Posttest 142.2 ⫾ 8.2 141.1 ⫾ 10.5 133.9 ⫾ 11.3
Elbow: S2 extension (⬚) Pretest 142.8 ⫾ 5.9 136.9 ⫾ 9.8 135.1 ⫾ 16.6
Posttest 143.9 ⫾ 8.9 141.8 ⫾ 7.2 136.6 ⫾ 14.5
Elbow: S1 ROM (⬚) Pretest 70.8 ⫾ 18.6 53.8 ⫾ 18.6 66.3 ⫾ 14.4
Posttest 72.7 ⫾ 21.2 51.3 ⫾ 19.3 64.7 ⫾ 15.2
Elbow: S2 ROM (⬚) Pretest 72.4 ⫾ 15.6 56.9 ⫾ 14.5 67.3 ⫾ 16.8
Posttest 75.2 ⫾ 20.6 56.5 ⫾ 21.4 68.0 ⫾ 17.2
Elbow: S1 ang vel (⬚·s⫺1) Pretest 396.8 ⫾ 145.3 309.1 ⫾ 128.7 453.9 ⫾ 188.6
Posttest 455.9 ⫾ 191.5 300.5 ⫾ 174.7 425.8 ⫾ 163.1
Elbow: S2 ang vel (⬚·s⫺1) Pretest 461.0 ⫾ 138.7 368.2 ⫾ 150.3 440.6 ⫾ 218.3
Posttest 476.3 ⫾ 195.7 407.3 ⫾ 231.8 430.7 ⫾ 156.9
Trunk lean (⬚) Pretest 31.1 ⫾ 11.4 32.8 ⫾ 7.7 25.5 ⫾ 4.7
Posttest 49.3 ⫾ 8.8§# 44.9 ⫾ 7.3§ 38.4 ⫾ 6.9
* ROM ⫽ range of motion; ang vel ⫽ angular velocity.
† p ⬍ 0.05 significant within-subjects main effect for time of test (pretraining-posttraining).
‡ p ⬍ 0.01 significant between-subjects interaction effect for time of test (pretraining-posttraining).
§ p ⬍ 0.001 significant within-subjects main effect for time of test (pretraining-posttraining).
㛳 p ⬍ 0.01 significant within-subjects main effect for time of test (pretraining-posttraining).
# p ⬍ 0.05 significant between-subjects effect for time of test (pretraining-posttraining).

improved by 4%. For the total acceleration distance (15 the inertia of the stationary start position. Therefore, the
m), the experimental groups improved by 6–8%, which greatest development in strength and power appears to
equated to 0.41 m·s⫺1 and 0.32 m·s⫺1 improvement for the have been produced over the first and second steps.
RS and NRS groups, respectively (Figure 2). The first 30 m of a 100-m sprint effort has been de-
Sprint training programs comprising a combination of scribed as a period of mass acceleration in which 60–70%
training modalities (plyometrics, flexibility, weight train- of maximum speed is achieved (13, 14), with close to 50%
ing, and sprint training) have been shown to increase of maximum speed being achieved in the first 10 m and
running velocity by 7–8% (9) compared to 4–5.5% for further 10% increases recorded for the 10–20- and 20–30-
sprint training alone (6). The training modalities used in m intervals. The speed increases plateaued between 30
the more effective of these studies were similar to those and 60 m, with only small increments (approximately 5%)
used in the current study, and may explain the compa- in speed over subsequent 10-m intervals. Rimmer and
rable improvement in running velocity. It would appear Sleivert (38) found that an 8-week sprint and plyometric
that a combination of sprint and plyometric exercises re- training program improved velocity over the first 10 m,
sults in a specific replication of the dynamic activities in- which in turn accounted for 50% of the improvement over
volved in sprinting, leading to improved running velocity. 40 m. The velocity patterns outlined above are similar to
The greatest improvement in 0–15 m acceleration per- that recorded over the 15-m acceleration distance exam-
formance was in the 0–5 m interval, in which approxi- ined in the current study indicating that the most impor-
mately 50% of the overall velocity increase was achieved tant phase of any sprint running is the period of mass
for both training groups (Figure 2). In contrast, the 5–10 acceleration.
m interval contributed 37 and 39% of the overall increase Stride length, stride frequency, and the subqualities
in velocity for the RS and NRS groups, respectively, of ground contact time and flight time are of particular
whereas the 10–15 m interval accounted for approxi- importance to running speed (23, 24). Sprinting speed is
mately 10% of the overall improvement in velocity. These the product of stride length and stride frequency (25, 36).
results reinforce the importance of the first 5 m of an Research (15, 16) indicates significant variability in stride
acceleration effort, wherein the athlete has to overcome lengths recorded at the start (2.04 m), at 20 m (1.53–1.66
 RESISTED SPRINT TRAINING AND ACCELERATION 83

TABLE 6. Mean ⫾ SD lower-body kinematic variables pretraining-posttraining for steps 1 (S1) and 2 (S2).*
Resisted Nonresisted Control
Test (n ⫽ 10) (n ⫽ 10) (n ⫽ 10)
Hip: S1 flexion (⬚) Pretest 97.2 ⫾ 13.6 95.6 ⫾ 11.2 95.2 ⫾ 11.4
Posttest 91.3 ⫾ 8.6 85.2 ⫾ 20.8 94.6 ⫾ 8.3
Hip: S2 flexion (⬚) Pretest 99.9 ⫾ 14.2 101.5 ⫾ 11.9 103.3 ⫾ 8.4
Posttest 93.3 ⫾ 7.8 92.4 ⫾ 15.6 101.3 ⫾ 7.1
Hip: S1 extension (⬚) Pretest 153.5 ⫾ 8.2 151.5 ⫾ 5.8 153.3 ⫾ 5.1
Posttest 154.4 ⫾ 11.0 141.0 ⫾ 27.8 157.0 ⫾ 6.2
Hip: S2 extension (⬚) Pretest 155.0 ⫾ 5.6 151.3 ⫾ 5.8 156.3 ⫾ 3.7
Posttest 160.2 ⫾ 6.7 148.2 ⫾ 27.9 159.6 ⫾ 6.2
Hip: S1 ROM (⬚) Pretest 56.2 ⫾ 11.7 55.8 ⫾ 9.7 58.1 ⫾ 12.4
Posttest 63.2 ⫾ 7.1 55.8 ⫾ 11.6 62.5 ⫾ 8.2
Hip: S2 ROM (⬚) Pretest 55.1 ⫾ 14.5 49.8 ⫾ 9.1 53.0 ⫾ 7.4
Posttest 66.9 ⫾ 6.8 55.8 ⫾ 15.0 58.3 ⫾ 6.2
Hip: S1 ang vel (⬚·s⫺1) Pretest 259.9 ⫾ 34.3 237.9 ⫾ 44.8 266.9 ⫾ 72.3
Posttest 224.2 ⫾ 118.8 275.7 ⫾ 41.5 276.9 ⫾ 40.8
Hip: S2 ang vel (⬚·s⫺1) Pretest 351.2 ⫾ 292.8 235.3 ⫾ 44.5 249.6 ⫾ 36.3
Posttest 667.6 ⫾ 339.8 270.7 ⫾ 32.2 267.7 ⫾ 31.8
Knee: S1 ROM (⬚) Pretest 88.0 ⫾ 23.9 83.8 ⫾ 11.1 81.0 ⫾ 11.4
Posttest 89.0 ⫾ 10.9 82.4 ⫾ 10.3 66.7 ⫾ 26.7
Knee: S2 ROM (⬚) Pretest 99.4 ⫾ 22.1 84.9 ⫾ 7.8 93.1 ⫾ 10.7
Posttest 96.8 ⫾ 9.1b 86.9 ⫾ 15.7ab 61.9 ⫾ 33.6a†§
Knee: S1 ang vel (⬚·s⫺1) Pretest 459.2 ⫾ 110.7 447.2 ⫾ 58.7 432.8 ⫾ 84.1
Posttest 497.7 ⫾ 50.2b 465.7 ⫾ 62.3ab 344.5 ⫾ 173.1a
Knee: S2 ang vel (⬚·s⫺1) Pretest 552.7 ⫾ 110.5 483.1 ⫾ 51.9 523.7 ⫾ 76.5
Posttest 567.8 ⫾ 108.4b 494.5 ⫾ 70.6ab 345.2 ⫾ 182.5a‡
* ROM ⫽ range of motion; ang vel ⫽ angular velocity. Group means sharing the same subscript (a or b) in the same row are not
significantly different at the p ⬍ 0.01 level.
† p ⬍ 0.05 significant within-subjects main effect for time (pretraining-posttraining).
‡ p ⬍ 0.05 significant between-subjects main effect for time of test (pretraining-posttraining).
§ p ⬍ 0.01 significant within-subjects interaction effect for time of test (pretraining-posttraining).

step) contact time for the RS (11.8%) and NRS (6.3%)

groups. The mean contact times recorded for the first and
second steps ranged from 0.15 to 0.19 seconds and are
similar to published block start data, which range from
0.15 to 0.22 seconds (31, 34).
It could be argued that the observed changes in stride
length, stride frequency, and contact time were sufficient
to bring about the significant increases in running veloc-
ity for both experimental groups. Rimmer and Sleivert
(38) found no significant changes in stride length or stride
frequency for both plyometric and sprint training groups,
but as in the current study, significant decreases in con-
tact time coincided with improved running velocity.
FIGURE 2. Contribution of velocity intervals to overall Lockie et al. (22) determined that acceleration contact
acceleration velocity. RS ⫽ resisted sprint; NRS ⫽ nonresisted time increased with larger loads, whereas Letzelter et al.
sprint.
(21) found greater contact times with increased resistance
during maximum-velocity running. A reduction in contact
time coupled with an increase in running velocity may
m), and at 30 m (1.71–1.85 m) of a sprint effort. The cur-
rent study indicated nonsignificant changes in stride also be explained by greater power output of the lower
length and stride frequency. Stride length decreased by limbs, specifically the hip and knee extensors. The in-
9% (0.19 m) and 1% (0.02 m) for the RS and NRS groups crease in ground contact time may be because of the ath-
respectively, whereas stride frequency increased by 9.5% lete’s requiring more time to produce greater muscular
(0.20 Hz) and 4% (0.06 Hz) for the RS and NRS groups power to overcome the increased resistance. An increase
respectively (Table 4). in power output may explain the decrease in first-step
Interestingly, Rimmer and Sleivert (38) found that contact time and increase in stride frequency. The power
stride frequency decreased by 7% after an 8-week sprint output of these muscles was measured by the lower-body
training program; however, running velocity increased power and reactive strength index tests (CMJ and 50DJ).
significantly. The step frequency of elite sprinters over Lower-body power, as measured by the CMJ, significantly
10–30 m ranges from 3.90 to 4.61 Hz (1, 28). These step increased by 5.9 and 9.1% for the RS and NRS groups
frequencies are higher than those recorded in the current respectively (Table 3).
study, and may be because of the nature of the partici- Sprint and plyometric training programs have been
pants (football players vs. elite sprinters) and the data shown to increase leg power, coinciding with a significant
collection distance (0–5 m vs. 0–30 m). increase in 30-m running velocity (19). Horizontal SSC,
There was a significant decrease in right-foot (first- as measured by the 5BT, significantly increased by 7 and
84 SPINKS, MURPHY, SPINKS ET AL.

8.4% for the RS and NRS groups respectively. The SSC lower than that recorded by Atwater (3) (step 2 ⫽ 60⬚)
action involved in the 5BT test is similar to that of the who examined the trunk lean of sprinters leaving starting
sprint stride, with high braking forces rapidly followed by blocks. However, the mean posttraining trunk lean values
propulsive forces applied over a short period of time (33, were closer to those recorded by Atwater (3). The signif-
38). If the power output of the hip and knee extensors is icant posttraining difference in trunk lean between the
improved, then changes in running speed might be ex- RS and control groups indicated that towing loads results
pected. Research (1, 31) has shown a high correlation (r in the athlete adopting a trunk lean position similar to
⫽ approximately 0.7) between increased running velocity that of a block start. Letzelter et al. (21) and Lockie et al.
and force production during the ground contact (propul- (22) showed that trunk lean increased with increased re-
sive) phase of acceleration efforts. sistance. However, this does not explain why the NRS
The 50DJ test indicated significant changes in reac- group significantly increased trunk lean. McInnis (28)
tive strength for the RS group (Table 3). Reactive stated that peak trunk lean should occur at the point of
strength is analogous with musculotendinous stiffness, maximum acceleration. In the current study, the greatest
and the changes in this variable for the RS group may increase in running velocity occurred in the 0–5-m inter-
explain the mechanism underlying the improved accel- val. The changes in trunk lean could have been because
eration performance demonstrated by this group for the of a practice effect and/or the participants adopting a
0–5-m interval. The change in reactive strength could be more efficient movement pattern allowing greater appli-
attributed to the RS subjects’ having to overcome the in- cation of horizontal force to the ground.
ertia of the weighted sled, and may also explain the sig- Few significant changes were observed in hip and
nificantly greater decrease in right-foot (second-step) con- knee joint kinematics (Table 6). It has been suggested
tact time for this group (11.7%) compared to the NRS that the thigh should make an angle of approximately 90⬚
(6.3%) group. The decrease in contact time for the NRS with the trunk during acceleration/sprinting (29). The
group was similar to that determined by Rimmer and first-stride hip flexion values of the RS and NRS groups
Sleivert (38) (⬃5%) over 7 and 37 m. were similar to this figure (91.3 and 85.2⬚ respectively),
Powerful and coordinated arm movement is deemed whereas the second-stride hip flexion was slightly higher
to be essential in promoting forward drive during the en- (93.3 and 92.4⬚ respectively). Experienced sprinters have
tire sprint cycle (28) but critically important during the demonstrated mean hip extension values at takeoff dur-
initial acceleration phase (20). The horizontal accelera- ing acceleration of 144⬚ (30). Hip extension during maxi-
tion of the arm swing is believed to increase stride length, mal speed running for experienced and elite sprinters
whereas the vertical component is believed to enhance leg ranges from 173 to 215⬚ (17, 27). The results obtained in
drive during ground contact (5) However, Ropret et al. the current study are similar to those recorded by Merni
(39) showed that loading the arms caused no significant et al. (30), with first-step hip extension values of 154.4
reduction in initial acceleration or maximum velocity over and 141.0⬚ for the RS and NRS groups respectively, and
30 m. second-step hip extension values of 160.2 and 148.2⬚ for
Across a range of sprint ability levels, maximum flex- the RS and NRS groups respectively. It would appear
ion and extension of the shoulder during maximal speed that RS training with loads approximating 10% of body
vary from 75 to 90⬚ and 37 to 60⬚ respectively (11, 27). mass may cause a change in hip flexion while not limiting
These findings are similar to those observed in the cur- hip extension. Again, there are evident differences be-
rent study for shoulder extension (65.1–76.5⬚) but are tween elite sprinters and football players.
greater than those determined for shoulder flexion (20.7– Mean knee ROM ranges from 114 to 121⬚ for poor and
24.2⬚) (Table 5). Shoulder ROM has been found to range elite sprinters respectively (2, 26). These results are high-
from 100 to 150⬚ (2, 27); however, the current study in- er than those found in the current study, in which knee
dicated a much lower range (44.4–52.4⬚), which could be ROM ranged from 89 to 82.4⬚ for the first step and 96.8
attributed to the smaller mean flexion results and per- to 86.9⬚ for the second step for the RS and NRS groups
haps to the use of less well-trained sprint athletes. respectively. The difference in value can be attributed to
The elbow ROM values were lower (56.5–75.2⬚) than the level of sprinter (elite sprinter vs. football player) and
those in previously reported research (2, 27), which in- the running velocity at which the values were recorded
dicated that maximum flexion, maximum extension, and (maximum running velocity vs. initial acceleration).
ROM of the elbow joint during maximum speed vary from
55 to 70⬚, 100 to 150⬚, and 70 to 90⬚ respectively. The PRACTICAL APPLICATIONS
results of the current study showed mean elbow flexion
values of 68.4 and 89.8⬚ and mean elbow extension values Resisted sprint training in the form of weighted sled tow-
of 136.4 and 142.2⬚ for the RS and NRS groups respec- ing is believed to increase lower-limb strength; however,
tively. The differences between the published research there are concerns that this training modality may not
and the current study may be explained by the distance transfer to acceleration performance because of negative
interval analyzed. Previous research examined elbow ki- influences on acceleration kinematics (18). Although the
nematics at maximum speed, whereas the current study current study indicated that long-term (8-week) use of RS
analyzed the acceleration phase. training has only a relatively minor influence on accel-
It appears that upper-body kinematics were not sig- eration kinematics and gait, acceleration performance, re-
nificantly altered by either of the training programs. Sim- active strength, lower-body power, and horizontal SSC
ilarly, increased running velocity was not associated with performance were enhanced, but no more effectively than
significant changes in upper-body kinematics. Therefore, by a NRS training program.
in terms of the current study, upper-body kinematics had The utilization of a sled load approximating 10% of
little impact on acceleration performance. body mass would appear to ensure that acceleration ki-
The RS and NRS groups displayed a significant in- nematics are not adversely affected, while still overload-
crease in forward trunk lean from pretraining to post- ing an athlete’s acceleration mechanics and developing
training (Table 5). The mean pretraining trunk lean was the specific recruitment of the hip and knee extensors,
 RESISTED SPRINT TRAINING AND ACCELERATION 85

resulting in greater application of power in the horizontal 22. LOCKIE, R.G., A.J. MURPHY, AND C.D. SPINKS. Effects of resisted sled
towing on sprint kinematics in field-sport athletes. J. Strength Cond.
direction. Res.. 17:760–767. 2003.
23. LOIZEAU, J., D. PARGANIN, AND M.P. LUONG. Dynamics of the stance
REFERENCES phase in sprinting using high-speed digital imaging. In: The Engineering
of Sport: Design and Development. S.J. Haake, ed. Oxford: Blackwell Sci-
1. ALEXANDER, M.J.L. The relationship between muscle strength and sprint ence, 1998. pp. 449–454.
kinematics in elite sprinters. Can. J. Sport Sci. 14:148–157. 1989. 24. LUHTANEN, P., AND P.V. KOMI. Mechanical factors influencing running
2. ARMSTRONG, L.E., D.L. COSTILL, AND G. GEHLSEN. Biomechanical com- speed. In: Biomechanics V1-B. E. Asmussen and K. Jorgensen, eds. Bal-
parison of university sprinters and marathon runners. Track Tech. 87: timore: University Park Press, 1978. pp. 23–29.
2781–2782. 1984. 25. MAJDELL, R., AND M.J.L. ALEXANDER. The effect of overspeed training
3. ATWATER, A.E. Kinematic analyses of sprinting. Track Field Q. Rev. 82: on kinematic variables in sprinting. J. Hum. Mov. Stud. 21:19–39. 1991.
12–16. 1982. 26. MANN, R. Biomechanical analysis of the elite sprinter and hurdler. In:
4. BANGSBO, J., L. NøRREGAARD, AND F. THORSø. Activity profile of com- The Elite Athlete. N.K. Butts and T.T. Gushiken, eds. New York: Spec-
petition soccer. Can. J. Sport Sci. 16:110–116. 1991. trum, 1985. pp. 43–80.
5. BHOWMICK, S., AND A.K. BHATTACHARYYA. Kinematic analysis of arm 27. MANN, R., AND J. HERMAN, J. Kinematic analysis of Olympic sprint per-
movements in sprint start. J. Sports Med. Phys. Fitness 28:315–323. formance: Men’s 200 meters. Int. J. Sport Biomech. 1:151–162. 1985.
1988. 28. MCINNIS, A. In quest of speed—Improving forward body lean in the
6. CALLISTER, R., M.J. SHEALY, S.J. FLECK, AND G.A. DUDLEY. Performance sprint start and early acceleration phase. Track Field Q. Rev. 80:25–31.
adaptations to sprint, endurance and both modes of training. J. Appl. 1980.
Sport Sci. Res. 2:46–51. 1988. 29. MECKEL, Y., H. ATTERBOM, A. GRODJINOVSKY, D. BEN-SIRA, AND A. ROT-
7. COMETTI, G., N.A. MAFFIULETTI, M. POUSSON, J.C. CHATARD, AND N. STEIN. Physiological characteristics of female 100 metre sprinters of dif-

MAFFULI. Isokinetic strength and anaerobic power of elite, subelite and ferent performance levels. J. Sports Med. Phys. Fitness 35:169–175. 1995.
amateur French soccer players. Int. J. Sports Med. 22:45–51. 2001. 30. MERNI, F., A. CICCHELLA, F. BOMBARDI, S. CIACCI, L. MAGENTI, S. OL-
MUCCI, AND L. COPPINI. Kinematic and dynamic analysis of sprint start.
8. DAVIS, J.A., J. BREWER, AND D. ATKIN. Pre-season physiological charac-
teristics of English first and second division soccer players. J. Sports Sci. In: ISBS ’92 Proceedings of the 10th Symposium of the International So-
10:541–547. 1992. ciety of Biomechanics in Sports. R. Rodano, ed. Milan, Italy, Edi-Ermes,
9. DELECLUSE, C., H. VAN COPPENOLLE, E. WILLEMS, M. VAN LEEMPUTTE, 1992. pp. 120–123.
31. MERO, A. Force-time characteristics and running velocity of male sprint-
R. DIELS, AND M. GORIS. Influence of high-resistance and high-velocity
ers during the acceleration phase of sprinting. Res. Q. Exerc. Sport 59:
training on sprint performance. Med. Sci. Sports Exerc. 27:1203–1209.
94–98. 1988.
1995.
32. MERO, A., AND P.V. KOMI. Effects of supramaximal velocity on biome-
10. DEUTSCH, M.U., G.A. KEARNEY, AND N.J. REHRER. A comparison of com-
chanical variables in sprinting. Int. J. Sport Biomech. 1:240–252. 1985.
petition work rates in elite club and Super 12 rugby. In: Science of Foot-
33. MERO, A., AND P.V. KOMI. Force, EMG and elasticity-velocity relation-
ball IV. W.L. Spinks, T. Reilly, and A.J. Murphy, eds. London: Routledge, ships at submaximal, maximal and supramaximal running speeds in
2002. pp. 160–166. sprinters. Eur. J. Appl. Physiol. 55:553–561. 1986.
11. DINTIMAN, G.B. Acceleration and speed. In: High Performance Sports 34. MERO, A., AND P.V. KOMI. Reaction time and electromyographic activity
Conditioning. B. Foran, ed. Champaign, IL: Human Kinetics, 2001. pp. during a sprint start. Eur. J. Appl. Physiol. Occup. Physiol. 61:73–80.
167–192. 1990.
12. DOUGE, B. Football: The common threads between the games. In: Science 35. NICHOLAS, C.W. Anthropometric and physiological characteristics of rug-
and Football. T. Reilly, A. Lees, K. Davids, and W.J. Murphy, eds. Lon- by union football players. Sports Med. 23:375–396. 1997.
don: E & FN Spon, 1988. pp. 3–19. 36. RADFORD, P.F. Sprinting. In: Physiology of Sport. T. Reilly, N. Secher, P.
13. GAFFNEY, S. Acceleration phase of the 100m sprint. Mod. Athlete Coach Snell, and C. Williams, eds. London: E & FN Spon, 1990. pp. 71–99.
28:35–38. 1990. 37. RIENZI, E., B. DRUST, T. REILLY, J.E.L. CARTER, AND A. MARTIN. Inves-
14. GAJER, B., C. THEPAUT-MATHIEU, AND D. LEHENAFF. Evolution of stride tigation of anthropometric and work-rate profiles of elite South American
and amplitude during course of the 100 m event in athletics. New Stud. international soccer players. J. Sports Med. Phys. Fitness 40:162–169.
Athletics 14:43–50. 1999. 2000.
15. GLIZE, D., AND M. LAURENT. Controlling locomotion during the acceler- 38. RIMMER, E., AND G. SLEIVERT. Effects of a plyometrics intervention pro-
gram on sprint performance. J. Strength Cond. Res. 14:295–301. 2000.
ation phase in sprinting and long jumping. J. Sports Sci. 15:181–189.
39. ROPRET, R., M. KUKOLJ, D. UGARKOVIC, D. MATAVULJ, AND S. JARIC. Ef-
1997.
fect of arm and leg loading on sprint performance. Eur. J. Appl. Physiol.
16. HARLAND, M.J., AND J.R. STEELE. Biomechanics of the sprint start.
Occup. Physiol. 77:547–550. 1988.
Sports Med. 23:11–20. 1997. 40. KAMEN, G. Aging, resistance training, and motor unit discharge behav-
17. JACOBS, R., AND G.J. VAN INGEN SCHENAU. Intermuscular coordination iour. Can. J. Appl. Physiol. 30:341–351. 2005.
in a sprint push-off. J. Biomech. 25:953–965. 1992.
18. JAKALSKI, K. The pros and cons of using resisted and assisted training Acknowledgments
methods with high school sprinters: Parachutes, tubing and towing.
This research was funded by a Football Science Scholarship
Track Coach 144:4585–4589, 4612. 1998.
granted to the first author by the Australian Rugby Union, the
19. KRAEMER, W.J., N.A. RATAMESS, J.S. VOLEK, S.A. MAZZETTI, AND A.L.
GOMEZ. The effect of the Meridian Shoe on vertical jump and sprint per-
Australian Football League, Australian Touch and the School of
formances following short-term combined plyometric/sprint and resis- Leisure, Sport and Tourism, University of Technology, Sydney.
tance training. J. Strength Cond. Res. 14:228–238. 2000. None of the authors have any conflict of interest. The assistance
20. LENTZ, D., AND A. HARDYK. Speed training. In: Training for Speed, Agil- provided by Dr Anthony Leicht, Dr Melissa Crowe, and Rob
ity and Quickness. L.E. Brown, V.A. Ferrigno, and J.C. Santana, eds. Bower is gratefully acknowledged.
Champaign, IL: Human Kinetics, 2000. pp. 17–77.
21. LETZELTER, M., G. SAUERWEIN, AND R. BURGER. Resistance runs in speed Address correspondence to Christopher Spinks, Cspinks@
development. Mod. Athlete Coach 33:7–12. 1995. pafc.com.au.

View publication stats