Content uploaded by David M Rouffet

Author content

All content in this area was uploaded by David M Rouffet on Dec 09, 2014

Content may be subject to copyright.

Introduction

The 200 m flying start is the qualifying race for the “Match

Sprint” competition. “Match Sprint” is considered to involve the

most explosive effort amongst high-performance track-cycling

events. It can be said to include three phases: an acceleration

phase (before the start of timed portion of the flying 200 m), a

maximal velocity phase, and a deceleration phase (the last both

included between the start and finish lines of the 200 m). Out-

standing national and international performances are commonly

completed within 10 – 11 s. This has not been significantly im-

proved over the last twenty years [40]. Also, in contrast with road

cycling, the absence of any gear system means that the constant

gear ratio chosen before the race directly influences the mean

Abstract

The aims of the present study were both to describe anthropo-

metrics and cycling power-velocity characteristics in top-level

track sprinters, and to test the hypothesis that these variables

would represent interesting predictors of the 200 m track sprint

cycling performance. Twelve elite cyclists volunteered to per-

form a torque-velocity test on a calibrated cycle ergometer, after

the measurement of their lean leg volume (LLV) and frontal sur-

face area (A

p

), in order to draw torque- and power-velocity rela-

tionships, and to evaluate the maximal power (P

max

), and both

the optimal pedalling rate (f

opt

) and torque (T

opt

) at which P

max

is

reached. The 200 m performances – i.e. velocity (V

200

) and pedal-

ling rate (f

200

) – were measured during international events

(REC) and in the 2002 French Track Cycling Championships

(NAT). P

max

, f

opt

, and T

opt

were respectively 1600 ± 116 W,

129.8 ± 4.7 rpm and 118.5 ± 9.8 N · m. P

max

was strongly correlated

with T

opt

(p < 0.001), which was correlated with LLV (p < 0.01).

V

200

was related to P

max

normalized by A

p

(p ≤ 0.05) and also to

f

opt

(p < 0.01) for REC and NAT. f

200

(155.2 ± 3, REC; 149 ± 4.3,

NAT) were significantly higher than f

opt

(p < 0.001). These find-

ings demonstrated that, in this population of world-class track

cyclists, the optimization of the ratio between P

max

and A

p

repre-

sents a key factor of 200 m performance. Concerning the major

role also played by f

opt

, it is assumed that, considering high val-

ues of f

200

, sprinters with a high value of optimal pedalling rate

(i.e. lower f

200

–f

opt

difference) could be theoretically in better

conditions to maximize their power output during the race and

hence performance.

Key words

Elite sprint cycling · 200 m flying start · maximal power (P

max

)·

optimal pedalling rate (f

opt

) · projected frontal area (A

p

)

Training & Testing

739

Affiliation

1

Laboratoire de Biomécanique et de Modélisation Humaine (Equipe Physiologie de l’Exercice) – Faculté de

Médecine Lyon-Sud, Oullins cedex, France

2

Laboratoire de Physiologie – Unité PPEH, CHU St-Jean-Bonnefonds, Saint-Etienne cedex 2, France

3

Laboratoire de Physiologie de l’Exercice (BAPS), Gabriel Montpied Hospital, Clermont-Ferrand cedex 1,

France

4

Centre de Recherche et d’Innovation sur le Sport (CRIS), Université Claude Bernard – Lyon I, Villeurbanne

cedex, France

Correspondence

S. Dorel · Laboratoire de Biomécanique et de Modélisation Humaine (Equipe Physiologie de l’Exercise) –

Faculté de Médecine Lyon-Sud · BP 12 · 69921 Oullins cedex · France · Phone: + 33478 863135 ·

Fax: + 33 478 86 3135 · E-mail: Dorel77@wanadoo.fr or Sylvain.Dorel@univ-lyon1.fr

Accepted after revision: September 30, 2004

Bibliography

Int J Sports Med 2005; 26: 739 – 746 © Georg Thieme Verlag KG · Stuttgart · New York ·

DOI 10.1055/s-2004-830493 · Published online January 24, 2005 ·

ISSN 0172-4622

S. Dorel

1,2

C. A. Hautier

4

O. Rambaud

1

D. Rouffet

4

E. Van Praagh

3

J.-R. Lacour

1

M. Bourdin

1

Torque and Power-Velocity Relationships in Cycling:

Relevance to Track Sprint Performance in

World-Class Cyclists

pedalling rate sustained in the sprint (f

200

). Retrospective anal-

ysis of the ten best peak performances obtained during the pre-

ceding 5 years’ world championships and world cup races ena-

bles us to calculate a strongly significant relationship between

performance in the 200 m time trial and the final competition

ranking (r = 0,95; p < 0.001). However, little data, to our knowl-

edge, is available concerning track sprint cycling. An anthropo-

metric analysis by McLean et al. [31] suggested a trend for corre-

lation of thigh girth with 200 m sprint performance. The authors

hypothesized that absolute strength may contribute to success in

this event. Thus, in this explosive event, muscle anaerobic me-

tabolism [20], neuromuscular and mechanical factors, including

maximal force and/or power of the lower limbs [43], may con-

tribute to final performance.

These muscle power characteristics demonstrated during cycling

can be accurately measured on a cycle ergometer, using the well

known torque-velocity test [1,19, 27, 28, 32,39,41]. The linear re-

lationship obtained between torque and pedalling rate enables

assessment of f

0

and T

0

, which have the dimensions of maximal

pedalling rate at the zero torque axis and the torque correspond-

ing to a zero pedalling rate, respectively (Fig.1). Maximal power

generation is described by a polynomial power-velocity relation-

ship [11,18, 28, 39]. Power output reaches a maximum value

(P

max

) at the optimal cycling rate (f

opt

). For example, sprint run-

ning and high jump performances have been demonstrated to

be correlated with P

max

[33,44] and sometimes also with f

opt

[16]. It is noteworthy that information concerning both power

and velocity data in elite track sprint cyclists is not well docu-

mented, neither is information concerning the possible relation-

ships between these factors and performance.

From the mechanical point of view, considering several endur-

ance cycling power equations [2, 5, 8,10,26,35,36] and the high

speeds achieved during sprint cycling (> 18 m · s

–1

), air resistance

(R

a

) represents the main resistive force acting on the rider + bi-

cycle system (> 90%). For a standardized environment, R

a

still de-

pends on speed (V) and two individual biomechanical-anthropo-

metric factors: the drag coefficient (C

d

) and the frontal area of the

rider + bicycle (A

p

) (i.e. R

a

=0.5ρA

p

C

d

V

2

, where ρ is the air den-

sity). Measurement of individual A

p

by photographic methods,

as proposed in some recent studies [5,17,37], might help a better

understanding of the mechanical linkage between power output

and performance (i.e. cycling speed).

The present study describes anthropometrics and the torque-

and power-velocity relationships in world-level sprint track cy-

clists, providing data for future comparative studies. The purpose

was to investigate whether these anthropometric and cycling

power characteristics were related to flying-start 200 m sprint

performance in this population. Finally, the present investigation

tested the hypothesis that f

200

attained during the 200 m is sim-

ilar or close to f

opt

and discussed how values of the latter would

influence performance.

Materials and Methods

Subjects and anthropometry

Twelve male elite cyclists volunteered to participate in the study.

Five were professional track cyclists, including three winners of

World or Olympic championships (F, H, K). All were competing

in national and international-level track races. Anthropometric

measurements are reported in Table 1. Four skinfold thicknesses

were measured to estimate the percentage of body fat (BF), using

the equations of Durnin and Rahaman [13]. Lean leg volume

(LLV) was calculated by anthropometry using Jones and Pear-

son’s technique [22]. The projected frontal area (A

p

) of ten of the

twelve cyclists (riding their own competition bicycles, clothes,

and helmet) was determined in the position they commonly

used during competition: i.e. from partially to fully bent-over

torso position (parallel with the ground), with the hands on the

drops portion of the handlebars and elbows more or less flexed.

Subjects were asked to maintain their head in the same position

as used during the race (Fig. 2). A

p

results were obtained accord-

ing to the method used by Heil et al. [17]. Digital photographs

were taken of each the participant sitting on his bicycle with a

square surface of known area at their side. The area enclosed by

Pedaling rate (rpm)

Power (W)

P

max

f

opt

R = 0,9865

2

R = 0,9911

2

R = 0,9687

2

R = 0,9538

2

0

50

100

150

200

250

300

0 50 100 150 200 250 300

2000

1750

1500

1250

1000

750

500

250

0

0

50 100 150 200 250 300

Pedaling rate (rpm)

Torque

(

Nm

)

.

f

0

T

0

ab

Fig.1a and b Torque (a) and power-velocity (b) relationships of two

athletes, F and K, fitted by average downstroke values obtained from

three sprints of 5-s duration. Determination of maximal power (P

max

),

optimal cycling rate (f

opt

), maximal cycling rate (f

0

), and maximal tor-

que (T

0

). F: filled circles; K: open circles.

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

740

the cyclist and his bicycle (including wheel spokes and the areas

between the lower limbs and the bicycle as well as that between

the upper limbs and the trunk) was delimited using a computer-

based image analysis software application (Scion Image Beta

4.02; Scion Corporation, Frederick, MD). The actual A

p

(in m

2

)

was determined by dividing this enclosed area by the area of

the calibration image and then multiplying by the known area

of the calibration frame (0.16m

2

). The study was performed dur-

ing racing season, one month before the National Track Cycling

Championships. The testing procedures were explained to the

participants before they gave their informed consent.

Torque-velocity test (T-V test)

After a 5-min warm-up, the cyclists performed three maximal

cycling sprints of 5-s duration, interspersed with 5-min rest peri-

ods according to the protocol proposed by Arsac et al. [1]. Each

sprint was performed against a specific force applied to the fric-

tion belt: 0.3, 0.6, or 0.9 N·kg

–1

body mass. The corresponding

friction torques were 0.287, 0.573, and 0.859 N · m·kg

–1

body

mass, respectively. The cycle ergometer (Monark 818E, Stock-

holm, Sweden) was equipped with the same handlebar and ped-

als as used during track sprint cycling. Saddle height, handlebar

height, and stem length were set to match the usual position of

the participants. Since it is well established that crank length in-

fluences the optimal pedalling rate [27], a standard crank length

of 170 mm (similar or close to the crank length used in the field,

i.e. 165 or 170 mm) was chosen to provide optimal test condi-

tions compared with field conditions. Subjects were told to re-

main seated on the saddle throughout the test, and were vigor-

ously encouraged to produce the highest acceleration possible.

Table 1 Individual anthropometric and best flying-start 200 m performance characteristics (n = 12)

Subject Age (yr) BM (kg) H (cm) BF (%) LLV (L) A

p

(m

2

)T

200

(s) % W

rec

(%)

A 22 80 182 8.6 10.03 0.533 10.23 96.5

B 22 83 172 12.3 9.99 0.512 10.35 95.3

C 21 78 178 10.4 9.62 0.525 10.6 93.1

D 23 92 185 12.6 10.69 0.558 10.81 91.3

E 22 84 185 13.2 11.05 0.536 10.24 96.3

F 29 78 176 11.4 8.19 0.510 10.04 98.3

G 26 88.5 181 11.9 10.7 0.537 10.44 94.5

H 27 78 183 10 10.07 – 10.33 95.5

I 19 86 178 15.6 9.49 0.535 10.38 95.1

J 21 76 184 10.2 8.53 0.525 10.91 90.4

K 29 87 182 8.1 11.09 0.538 10.07 98

L 31 85.5 183 11.7 10.85 – 10.42 94.7

Mean (SD) 24.3 (3.9) 83 (5) 180.8 (3.9) 11.3 (2) 10.03 (0.95) 0.531 (0.014) 10.40 (0.26) 94.9 (2.4)

BM, body mass (kg); H, height (cm); BF, body fat (%); LLV, lean leg volume (L); A

p

, projected frontal area of the cyclist and his bicycle (m

2

); T

200

, time reached during the best

performance in flying-start 200 m (s); %W

rec

, best performance, expressed as a percentage of the world record

Fig. 2 Sample A

p

photograph of a single

subject (A) on his own bicycle in the tradi-

tional body position and with the equipment

used during competition. The square surface

of known area (0.16m

2

) is presented at his

side.

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

741

The technical and mechanical characteristics of the friction-load-

ed cycle ergometer have been previously described [1, 32]. A

strain gauge (200 N, bandwidth 500 Hz) measured the frictional

force. An optical encoder (1969.2 points per meter of displace-

ment or 11815 points per pedal revolution) recorded flywheel

displacement (m = 22.5 kg). Force and displacement signals were

sampled (200 Hz) and stored on a PC via a 12-bit analog-to-digi-

tal interface card (DAS-8, 12 bits, Keithley Metrabyte, Taunton,

MA). First and second order derivatives of the flywheel dis-

placement were calculated to obtain flywheel velocity and ac-

celeration. The external torque produced was calculated as the

sum of the frictional torque (given by the strain gauge) plus the

torque necessary to accelerate the flywheel [14, 23, 32]. The mo-

ment of inertia of the flywheel (I) had been previously deter-

mined using the method of the free deceleration of the flywheel

described by Arsac et al. [1], Lakomy [23], and Martin et al. [28]:

I=0.927kg·m

2

. Power, torque and pedalling rate were averaged

for pedal down-stroke. After computation, the data obtained

from the three sprints were cumulated and used to draw up the

torque- and power-velocity relationships, using linear and quad-

ratic regressions, respectively [11,16,18, 28, 33,41].

Both maximal pedalling rate (f

0

, in rpm) and maximal torque (T

0

,

in N.m) were obtained by extrapolation. They correspond to the

intercept of the torque-velocity curve with the velocity and

torque axes, respectively. Maximal power (P

max

, in W) was iden-

tified as the apex of the power-velocity relationship. Optimal

pedalling rate (f

opt

) was the pedalling rate at which P

max

occurred.

Consecutive calculation considered the derivative of the equa-

tion of the power-velocity relationship to be equal to zero at this

point. The value of the torque at which P

max

occurred – i.e. opti-

mal torque (T

opt

) – was deduced from values of P

max

and f

opt

. The

cycling torque-velocity and power-velocity relationships for two

typical athletes (F and K) are presented in Fig.1. All maximal

power values (P

max

) were normalized for body mass (W · kg

–1

)

and for A

p

(W·m

–2

).

Flying-start 200 m cycling performances

Individual 200 m records (REC) were initially collected, all of

which had been achieved on indoor tracks over the last three

years. Each subject used the same competitive equipment and

bicycle (Look KG396, France). The testing protocol was carried

out during the competitive period. Different environmental as-

pects such as altitude and track composition are known to influ-

ence individual cycling performances [2, 40]. For this reason, per-

formances completed by the athletes during the French Track Cy-

cling Championships (NAT) – i.e. with the same training level

and the same environmental conditions (altitude: sea level;

250 m outdoor wooden track, with constant wind speed) – were

also analysed in the current study. These track test performances

were organized one month after the anthropometric and T-V test

measurements. 200 m sprint cycling performances were meas-

ured with an electronic timer (accuracy ± 0.001 s) by federation

officials, and corresponding mean velocities (V

200

,inm·s

–1

) were

calculated. The gear ratio determined the distance the bike trav-

elled in one pedal revolution (D

R

, in m). Considering the gear ra-

tio used for each performance and a constant wheel diameter,

the mean pedalling frequency (f

200

, in rpm) was calculated using

the following equation: f

200

=(V

200

·60)/D

R

.

Statistical analysis

All data were analyzed with StatView software (version 5.0).

They are expressed as means ± standard deviation (SD). Linear

torque-velocity and quadratic power-velocity regression models

were fitted by the least square method. Pearson product-mo-

ment correlation coefficients were calculated to determine

whether relationships between anthropometric data (LLV, A

p

),

T-V test data (f

0

, f

opt

,T

0

,T

opt

,P

max

) and 200 m track cycling per-

formances (V

200

, f

200

) were significant. Significant relationships

between variables were examined by linear and multiple regres-

sion analysis. Paired t-tests were used to study the differences in

cycling rates between the T-V test and track sprint cycling values

(f

opt

and f

200,

respectively). The level of significance was set at

p ≤ 0.05.

Results

Torque-velocity test

In the whole group, the torque- and power-velocity relationships

were described best by linear and quadratic models, respectively.

The determination coefficients averaged 0.981 (± 0.010) and

0.957 (± 0.015), respectively. Values of maximal and optimal cy-

cling rates (f

0

and f

opt

), maximal and optimal torques (T

0

and T

opt

),

and maximal absolute and relative power (P

max

) are shown in

Table 2.T

0

and T

opt

were significantly related to lean leg volume

(LLV) (r = 0.77, p < 0.01 and r = 0.69, p < 0.01). P

max

was signifi-

cantly correlated with T

0

and T

opt

(r = 0.92, p < 0.001 and r = 0.91,

p < 0.001; Fig. 3) but not with f

0

or f

opt

.

A

p

and 200 m track cycling performances

Values of A

p

are presented in Table 1.A

p

was significantly related

to the body mass (BM, r = 0.76, p < 0.01) and the height (H,

r = 0.75, p < 0.01). A

p

was also significantly correlated with LLV

(r = 0.66, p < 0.05).

Individual values of velocity and cycling rate obtained during

REC and NAT are presented in Table 2. NAT performances were

3.5% lower than REC, but both performances were strongly cor-

related (r = 0.90; p < 0.001). These performances represent re-

spectively 91.6% and 94.9 % of the current world record (set at

high altitude: 9.865 s – 20.27 m ·s

–1

; Bogotta, 1995). V

200

was

strongly correlated with f

200

for both REC and NAT (r = 0.87,

p < 0.001 and r = 0.93, p < 0.001, respectively).

Relationships with 200 m performances

No significant relationships were found between V

200

and P

max

(in W or W·kg

–1

) on linear regression analysis. Significant rela-

tionships were observed in V

200

and P

max

/A

p

for REC and NAT

(r = 0.75, p = 0.01; r = 0.63, p = 0.05, respectively; Fig. 4). A signifi-

cant relationship was observed between V

200

and f

opt

(r = 0.77,

p < 0.01 for both performances; Fig. 5). On the multiple regres-

sion analysis, the highest V

200

coefficient was obtained when

both independent variables f

opt

and P

max

/A

p

were taken into ac-

count (REC: r = 0.96, p < 0.0001; NAT: r = 0.87, p < 0.01). None of

the anthropometric variables were directly correlated with V

200

.

f

200

was higher than f

opt

by 19.2 % for REC and by 14.8% for NAT

(p < 0.001, Table 2).

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

742

Discussion

Torque-velocity relationships in world-class sprint cyclists

To the best of our knowledge, torque and power-velocity rela-

tionships of an elite track sprint cyclist group have not previously

been reported in the literature. Previous investigations have re-

ported maximum P

max

values of between 17.1 W · kg

–1

in elite

power athletes [7, 44] and 20.2 W· kg

–1

in one elite track cyclist

[28]. This last figure falls within our present range, which was

from 17.6 to 21.8 W · kg

–1

. Such high P

max

values measured in this

population support the hypothesis put forward by several au-

thors that sprint training background and genetics could be fac-

tors which might determine individual P

max

[1,18, 44]. Addition-

ally, our mean value for f

opt

(129.8 rpm) is higher than previously

reported values of 122 rpm [28]. In the same way, compared to

the few data reported for national-level sprint runners and elite

junior track cyclists (i.e. 247 and 249 rpm, respectively) [44], the

present mean f

0

value (260 rpm) was high. However, this value is

similar to that of the elite track cyclist reported by Martin et al.

(260 rpm) [28] and f

opt

value is slightly lower than values report-

ed by Hintzy et al. (134.3 rpm) [18] and found in our own labora-

tory (138 rpm, unpublished data), in athletes specialized in ex-

plosive-type events and in national-level sprint runners, respec-

Table 2 Individual values of maximal and optimal pedalling rates (f

0

, f

opt

), maximal and optimal torques (T

0

,T

opt

), maximal power (P

max

)

obtained during the T-V test, and 200 m sprint cycling data expressed in terms of individual record (REC) velocity and cycling rate

(V

200

, f

200

) or the mean values attained during national championships (NAT)

Partici-

pants

f

0

(rpm)

f

opt

(rpm)

T

0

(N · m)

T

opt

(N · m)

P

max

(W)

P

max

(W · kg

–1

)

V

200

(REC)

(m ·s

–1

)

f

200

(REC)

(rpm)

V

200

(NAT)

(m · s

–1

)

f

200

(NAT)

(rpm)

A 260 130 241 121 1640 20.5 19.55 157 19.18 154

B 267 133 224 108 1510 18.2 19.32 152 18.71 150.3

C 259 129 220 111 1500 19.2 18.87 154.7 18.42 151

D 247 123 254 127 1635 17.8 18.50 151.7 18.17 142.9

E 257 128 270 137 1830 21.8 19.53 156.9 18.53 148.8

F 282 141 215 107 1580 20.3 19.92 160 19.22 154.4

G 256 127 233 119 1587 17.9 19.16 153.9 18.45 145.1

H 262 131 225 112 1533 19.7 19.36 158.7 18.63 149.6

I 262 130 230 119 1615 18.8 19.27 154.8 18.75 150.6

J 251 123 217 113 1460 19.2 18.33 150.3 17.48 140.4

K 258 130 269 134 1800 20.7 19.86 159.5 18.92 152

L 259 132 230 114 1502 17.6 19.19 154.2 18.48 148.4

Mean (SD) 260.0

(8.6)

129.8

(4.7)

235.7

(19.1)

118.5

(9.8)

1600

(116)

19.3

(1.3)

19.24

(0.48)

155.3***

(3.1)

18.58

(0.46)

149***

(4.3)

*** p < 0.001, f

200

significantly different from f

opt

Fig. 3 The relationship between maximal power (P

max

) and T

opt

mea-

sured by the torque-velocity test (n = 12).

Fig. 4 Relationships between the record velocity (V

200

) of the cyclists

during their best 200 m performance (REC: open circles) and the mean

value attained during the French Track Cycling Championships (NAT:

filled circles) and the maximal power normalized by the frontal area

(P

max

/A

p

), (n = 10).

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

743

tively. The mean T

0

value (235.8 N · m) obtained in the present

study is in agreement with the previously elite track cyclist re-

ported value of 237 Nm [28]. Our range of values for T

0

(215 –

270 N · m) is similar to the highest values reported by McIntosh

et al. [25] for a track cyclist and a power lifter (i.e 235 –

287 N·m; calculated from the reported maximal resistance force

data).

According to the linear torque-velocity relationship, f

opt

and T

opt

were equal to 0.5 f

0

and 0.5 T

0

, respectively, for each individual.

Notwithstanding the fact that all these results presented in Table

2 provide data for future comparative studies, the following dis-

cussion will focus only on P

max

and its two components f

opt

and

T

opt

. In contrast to previous studies [1,18], no relationship was

observed between P

max

and f

opt

in this homogenous elite track-

cyclist group. P

max

was better explained by T

opt

(Fig. 3), which

was significantly related to LLV. In agreement with Driss et al.

[12], who suggested T

0

to be a good indicator of maximal

strength, this underlines the possible effect of the strength train-

ing that has been practiced in this group for the last five to ten

years. Comparing the high T

opt

and P

max

values with the compa-

rable f

opt

values to those obtained in other explosive events, it

would suggest that the training program applied in this popula-

tion (specific cycling training and strength conditioning) did not

represent a means of improving optimal pedalling rate. This is

coherent with the hypothesis that the velocity ability (f

opt

) is less

sensitive to the training. This may account as well for the fibre

conversion from type IIb (or IIx) to IIa which could result from

the enhanced strength training program run in this population

[9, 42].

P

max

,A

p

, and 200 m performances

Values of V

200

NAT (18.58 m · s

–1

, outdoor track) compared to V

200

REC (19.24 m·s

–1

, indoor track) confirm the high level of both

training and commitment of the athletes during the test period.

In contrast to our hypothesis and despite the trend, the linear re-

lationship between 200 m performance and P

max

(in W or

W·kg

–1

) was not statistically significant in this elite population

(r ranged from 0.39 to 0.57, and p ranged from 0.207 to 0.054 for

REC and NAT). However, minimizing aerodynamic parameters

(A

p

and C

d

) induces a decrease in the part of mechanical power

wasted in overcoming air resistance. Hence, for a given power

output, this results in an increase of the cycling speed [2,5,

8,26,35 – 37]. It could therefore be expected that the measure-

ment of these parameters would make sense to the relationship

between power output and cycling speed. The significant rela-

tionship observed between V

200

and P

max

related to A

p

(for REC

and NAT, Fig. 4) supports this expectation. C

d

(the second compo-

nent of the drag area, which is equal to the product of A

p

and C

d

)

was not measured, although it probably also plays a major role in

performance. This represents a limitation in the interpretation of

the results. Further investigations using wind tunnel [37] or free

deceleration techniques [4] may provide more information

about the link between these two aerodynamic parameters,

maximal power, and performance.

The mean value of A

p

(0.531 m

2

) was comparable with the total

A

p

values reported by Heil et al. [17] in road cyclists in compara-

ble positions and is in agreement with the Body A

p

data (0.42 m

2

)

reported by Capelli et al. [5] considering a Bicycle A

p

value of

0.117 – 0.130 m

2

[17]. The finding that both body mass and height

significantly influenced variation in A

p

which is consistent with

previous studies, confirms the influence of anthropometrics (i.e.

the rider’s size) on aerodynamics and hence on performance [5].

However, in the present population, the significant relationships

also observed between A

p

and LLV and between LLV and T

opt

emphasize the paradox regarding the rider’s size: on the one

hand decreasing to minimize A

p

and on the other hand increas-

ing to improve muscle volume, torque, and hence power output.

Regarding all these parameters, further improvements of the

200 m performance still seem to be possible for some subjects

by increasing P

max

and optimizing, in the same time, the body

fat (which remains slightly higher than values reported for pro-

fessional road cyclists, from 5 to 8– 9%, [6,24, 45]) to decrease or

at least maintain A

p

. Other ways to decrease A

p

and/or C

d

would

be to optimize the position on the bicycle [21] and to systemati-

cally use an aerodynamic helmet.

Pedalling rates and 200 m performances

The significant relationship obtained between V

200

and f

200

is

mainly explained by the fact that only three very similar gear ra-

tios were used in this group (i.e. G

R

= 7.32, 7.47, or 7.63 m). The

major finding of the present study was that V

200

was significantly

related to f

opt

in these top-level sprint cyclists (Fig. 5). Although

f

opt

is assumed to be related to explosive training status, to the

best of our knowledge no previous study has assessed the rela-

tionship between this factor and top performance in any sprint

cycling event. According to Vandewalle et al. [44], Sargeant et al.

[39] and Hautier et al. [16] who suggest that f

opt

is related to the

percentage of fast twitch muscle fibres, a higher f

opt

must be ben-

eficial for “all-out” explosive events. Even if f

opt

represents one of

the two components of P

max

, no relationship was found between

these two variables: the multiple regression analysis showed

that these parameters taken together explained 91% and 80% of

the V

200

variation for REC and NAT, respectively. It seems to indi-

cate that having a high f

opt

value is not an advantage regarding

Fig. 5 Relationships between the record velocity (V

200

) of the cyclists

during their best 200 m performance (REC: open circles) and the mean

value attained during the French Track Cycling Championships (NAT:

filled circles) and the optimal cycling rate (f

opt

) measured by the T-V

test (n = 12).

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

744

the absolute value of P

max

and that the influence of f

opt

must be

elucidated independently.

Regarding the specific pedalling conditions on the track (f

200

), the

hypothesis that f

200

was similar to f

opt

was not confirmed, i.e. f

200

being higher by 19.8 % and 14.8% for REC and NAT, respectively.

Moreover, the main finding is that this discrepancy is not equiv-

alent among subjects (range: 13.5 % to 23.3 % for REC). Such an

observation is particularly important as it refers to the power-

velocity relationship. Thus, a previous study by our group dem-

onstrated that during a short (< 5-s) “all-out” exercise the closer

the velocity was to f

opt

, the greater the total work performed [11].

As illustrated in Fig. 6, the maximal value of power produced at

this pedalling rate (Pf

200

) and calculated via the power-velocity

equation, does not represent the same percentage of P

max

among

the subjects. Associated with the fact that only three very similar

gear ratios were used in this group, the f

opt

of the sprinters being

higher and hence made the pedalling rate during the 200 m (f

200

)

closer to that in racing conditions, so their ability was greater to

produce high Pf

200

. This is confirmed by the strong relationship

obtained between the relative Pf

200

(expressed as a percentage

of P

max

) and f

opt

(r = 0.81, p < 0.001 for REC; r = 0.67, p < 0.05 for

NAT).

Hypothesis on the choice of pedalling rate during

the track event

As discussed above, it can be reasonably assumed that using a

higher gear ratio could result in lower pedalling rate and hence

lead to higher power during the race and performance, especially

for individuals presenting low f

opt

values. Additionally, cycling

rates closer to f

opt

may decrease the energy cost of the internal

power used to move the legs [11,15,30], the negative muscle

work [34] and may increase the mechanical efficiency [11, 38].

Furthermore, the capacity of the athletes to resist fatigue

throughout the race (acceleration: > 6-s and timed portion; 10 –

11-s) should not be neglected in the understanding of the per-

formance. Once more, previous studies clearly demonstrated

that fatigue index during maximal cycling exercise was also de-

creased at lower pedalling velocities (60 and 100 rpm) compared

to higher (120 and 140 rpm) [3, 29].

One question remains: why are f

200

values higher than f

opt

val-

ues? The results of the present study allow asking the question

rather than answering to this one. Firstly, an alternative interpre-

tation is that gears the riders choose allow maximal power to oc-

cur during the acceleration phase. Further studies, currently

underway, will estimate the effect of using higher gear ratios on

global performance and on power production during each specif-

ic phase. Secondly, it should be kept in mind that, historically, the

200 m is usually performed with the gear ratio used in the

“Match Sprint”. During this event, the choice of the gear ratio is

more complex since it depends on other mechanical (great accel-

eration) and tactical factors (opponent reaction).

In conclusion, the main findings of our study that, both P

max

related to A

p

and f

opt

were significant predictors of 200 m per-

formance, indicate that the cycling torque-velocity test as well

as measurement of the frontal area have several applications in

the understanding of the top-level track sprint cycling perform-

ance. Variations in these parameters could be used to character-

ize the general ability of sprinters and/or to test the efficiency of

a specific training program. Additionally, the performance on the

track was characterized by specific high-rate pedalling condi-

tions (f

200

, 150.3 to 160 rpm) compared to f

opt

(123 to 141 rpm).

This difference being far from uniform among subjects, this

could probably account for the important role played by f

opt

on

performance and primarily questioned the adjustment of the

gear ratios to the individual’s power-velocity characteristics. Fur-

ther studies are needed to investigate whether an individualized

gear ratio (i.e. a lower f

200

value, closer to f

opt

) may improve

200 m performance and what impact it could have on the fatigue

process.

Fig. 6a and b Determination of P

f200

(REC) from the power-velocity

relationship of two typical athletes. F (a) and K (b) and from the value

of the mean cycling rate maintained during the 200 m performance

(f

200

). Observe the smaller discrepancy between P

max

and P

f200

for ath-

lete F (f

opt

: 141 rpm, f

200

: 160 rpm) compared with K (f

opt

: 130 rpm, f

200

:

159.5 rpm).

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

745

Acknowledgements

A part of the submission has been presented at the 8th Annual

Congress of the European College of Sport Science, Salzburg,

2003.

This study was possible only because of the cooperation of sev-

eral individuals. The authors wish to extend their sincere appre-

ciation to the participants, to Daniel Morelon and Gerard Quin-

tyn, the coaches of the French Federation of Cycling for allowing

us to conduct the tests, to Dr. Ronan Martin, Dr. Carine Bret, Cy-

rille Routier for assistance during the protocol and to Dr. Frédéric

Grappe for material assistance.

References

1

Arsac LM, Belli A, Lacour JR. Muscle function during brief maximal ex-

ercise: accurate measurements on a friction-loaded cycle ergometer.

Eur J Appl Physiol 1996; 74: 100 – 106

2

Bassett DR Jr, Kyle CR, Passfield L, Broker JP, Burke ER. Comparing cy-

cling world hour records, 1967 – 1996: modeling with empirical data.

Med Sci Sports Exerc 1999; 31: 1665– 1676

3

Beelen A, Sargeant AJ. Effect of fatigue on maximal power output at

different contraction velocities in humans. J Appl Physiol 1991; 71:

2332 –2337

4

Candau RB, Grappe F, Menard M, Barbier B, Millet GY, Hoffman MD,

Belli AR, Rouillon JD. Simplified deceleration method for assessment

of resistive forces in cycling. Med Sci Sports Exerc 1999; 31: 1441 –

1447

5

Capelli C, Schena F, Zamparo P, Monte AD, Faina M, di Prampero PE.

Energetics of best performances in track cycling. Med Sci Sports Exerc

1998; 30: 614 – 624

6

Coyle EF, Feltner ME, Kautz SA, Hamilton MT, Montain SJ, Baylor AM,

Abraham LD, Petrek GW. Physiological and biomechanical factors

associated with elite endurance cycling performance. Med Sci Sports

Exerc 1991; 23: 93– 107

7

Davies CT, Sandstrom ER. Maximal mechanical power output and ca-

pacity of cyclists and young adults. Eur J Appl Physiol 1989; 58: 838–

844

8

de Groot G, Sargeant A, Geysel J. Air friction and rolling resistance dur-

ing cycling. Med Sci Sports Exerc 1995; 27: 1090– 1095

9

Delecluse C. Influence of strength training on sprint running perform-

ance. Current findings and implications for training. Sports Med

1997; 24: 147 – 156

10

di Prampero PE, Cortili G, Mognoni P, Saibene F. Equation of motion of

a cyclist. J Appl Physiol 1979; 47: 201– 206

11

Dorel S, Bourdin M, Van Praagh E, Lacour JR, Hautier CA. Influence of

two pedalling rate conditions on mechanical output and physiologi-

cal responses during all-out intermittent exercise. Eur J Appl Physiol

2003; 89: 157– 165

12

Driss T, Vandewalle H, Le Chevalier JM, Monod H. Force-velocity rela-

tionship on a cycle ergometer and knee-extensor strength indices.

Can J Appl Physiol 2002; 27: 250 – 262

13

Durnin JV, Rahaman MM. The assessment of the amount of fat in the

human body from measurements of skinfold thickness. Br J Nutr

1967; 21: 681 – 689

14

Falgairette G, Billaut F, Giacomoni M, Ramdani S, Boyadjian A. Effect of

inertia on performance and fatigue pattern during repeated cycle

sprints in males and females. Int J Sports Med 2004; 25: 235–240

15

Francescato MP, Girardis M, di Prampero PE. Oxygen cost of internal

work during cycling. Eur J Appl Physiol 1995; 72: 51 – 57

16

Hautier CA, Linossier MT, Belli A, Lacour JR, Arsac LM. Optimal velocity

for maximal power production in non-isokinetic cycling is related to

muscle fibre type composition. Eur J Appl Physiol 1996; 74: 114– 118

17

Heil DP. Body mass scaling of frontal area in competitive cyclists not

using aero-handlebars. Eur J Appl Physiol 2002; 87: 520– 528

18

Hintzy F, Belli A, Grappe F, Rouillon JD. Optimal pedalling velocity

characteristics during maximal and submaximal cycling in humans.

Eur J Appl Physiol 1999; 79: 426 –432

19

Jaskolska A, Goossens P, Veenstra B, Jaskolski A, Skinner JS. Compari-

son of treadmill and cycle ergometer measurements of force-velocity

relationships and power output. Int J Sports Med 1999; 20: 192 – 197

20

Jeukendrup AE, Craig NP, Hawley JA. The bioenergetics of World Class

Cycling. J Sci Med Sport 2000; 3: 414 – 433

21

Jeukendrup AE, Martin J. Improving cycling performance: how should

we spend our time and money. Sports Med 2001; 31: 559 –569

22

Jones PR, Pearson J. Anthropometric determination of leg fat and

muscle plus bone volumes in young male and female adults. J Physiol

1969; 204: 63 – 66

23

Lakomy HK. Measurement of work and power output using friction-

loaded cycle ergometers. Ergonomics 1986; 29: 509 – 517

24

Lucia A, Hoyos J, Chicharro JL. Physiology of professional road cycling.

Sports Med 2001; 31: 325– 337

25

MacIntosh BR, Neptune RR, Horton JF. Cadence, power, and muscle ac-

tivation in cycle ergometry. Med Sci Sports Exerc 2000; 32: 1281 –

1287

26

Martin JC, Milliken DL, Cobb JE, MacFadden KL, Coggan AR. Validation

of a mathematical model for road cycling power. J Appl Biomech

1998; 14: 276– 291

27

Martin JC, Spirduso WW. Determinants of maximal cycling power:

crank length, pedalling rate and pedal speed. Eur J Appl Physiol

2001;84:413–418

28

Martin JC, Wagner BM, Coyle EF. Inertial-load method determines

maximal cycling power in a single exercise bout. Med Sci Sports Exerc

1997; 29: 1505 –1512

29

McCartney N, Heigenhauser GJ, Jones NL. Power output and fatigue of

human muscle in maximal cycling exercise. J Appl Physiol 1983; 55:

218 –224

30

McDaniel J, Durstine JL, Hand GA, Martin JC. Determinants of meta-

bolic cost during submaximal cycling. J Appl Physiol 2002; 93: 823 –

828

31

McLean BD, Parker AW. An anthropometric analysis of elite Australian

track cyclists. J Sports Sci 1989; 7: 247– 255

32

Morin JB, Belli A. A simple method for measurement of maximal

downstroke power on friction-loaded cycle ergometer. J Biomech

2004; 37: 141 – 145

33

Morin J-B, Hintzy F, Belli A, Grappe F. Force-velocity relationships and

sprint running performances in trained athletes. Sci Sports 2002; 17:

78 –85

34

Neptune RR, Herzog W. The association between negative muscle

work and pedalling rate. J Biomech 1999; 32: 1021 –1026

35

Olds TS, Norton KI, Craig NP. Mathematical model of cycling perform-

ance. J Appl Physiol 1993; 75: 730– 737

36

Olds TS, Norton KI, Lowe EL, Olive S, Reay F, Ly S. Modeling road-cy-

cling performance. J Appl Physiol 1995; 78: 1596– 1611

37

Padilla S, Mujika I, Angulo F, Goiriena JJ. Scientific approach to the 1-h

cycling world record: a case study. J Appl Physiol 2000; 89: 1522–

1527

38

Sargeant AJ. Neuromuscular determinants of human performance. In:

Whipp BJ, Sargeant AJ (eds). Physiological Determinants of Exercise

Tolerance in Humans. London: Portland Press Ltd, 1999: 13 – 28

39

Sargeant AJ, Hoinville E, Young A. Maximum leg force and power out-

put during short-term dynamic exercise. J Appl Physiol 1981; 51:

1175 – 1182

40

Schumacher YO, Mueller P, Keul J. Development of peak performance

in track cycling. J Sports Med Phys Fitness 2001; 41: 139 – 146

41

Seck D, Vandewalle H, Decrops N, Monod H. Maximal power and

torque-velocity relationship on a cycle ergometer during the acceler-

ation phase of a single all-out exercise. Eur J Appl Physiol 1995; 70:

161 – 168

42

Staron RS, Karapondo DL, Kraemer WJ, Fry AC, Gordon SE, Falkel JE,

Hagerman FC, Hikida RS. Skeletal muscle adaptations during early

phase of heavy-resistance training in men and women. J Appl Physiol

1994; 76: 1247– 1255

43

van Soest O, Casius LJ. Which factors determine the optimal pedalling

rate in sprint cycling? Med Sci Sports Exerc 2000; 32: 1927– 1934

44

Vandewalle H, Peres G, Heller J, Panel J, Monod H. Force-velocity rela-

tionship and maximal power on a cycle ergometer. Correlation with

the height of a vertical jump. Eur J Appl Physiol 1987; 56: 650– 656

45

Wilber RL, Zawadzki KM, Kearney JT, Shannon MP, Disalvo D. Physio-

logical profiles of elite off-road and road cyclists. Med Sci Sports Exerc

1997; 29: 1090 – 1094

Dorel S et al. Sprint Performance of Elite Track Cyclists … Int J Sports Med 2005; 26: 739 – 746

Training & Testing

746