Power Output during Stage Racing in

Professional Road Cycling

STEFAN VOGT, LOTHAR HEINRICH, YORCK OLAF SCHUMACHER, ANDREAS BLUM, KAI ROECKER,
HANS-HERMANN DICKHUTH, and ANDREAS SCHMID
University of Freiburg, Center for Internal Medicine, Department of Prevention, Rehabilitation and Sports Medicine,
Freiburg, GERMANY

ABSTRACT
VOGT, S., L. HEINRICH, Y. O. SCHUMACHER, A. BLUM, K. ROECKER, H-H. DICKHUTH, and A. SCHMID. Power Output
during Stage Racing in Professional Road Cycling. Med. Sci. Sports Exerc., Vol. 38, No. 1, pp. 147–151, 2006. Purpose: The aim of
the study was to evaluate the power output during a multistage professional road race using direct power measurements and to compare
these results with the performance measurements using competition heart rate recordings. Methods: Six professional road cyclists
performed an incremental cycling test during which peak power output, power output, and heart rate at the lactate threshold (LT) and
at a lactate increase of 1 mM above the LT (LT ⫹ 1) were assessed. During a six-stage road race competition, power output was
measured directly (SRM crankset). To analyze the time spent at different intensities during competition, the amount of competition time
spent below LT (zone 1), between the LT and LT ⫹ 1 (zone 2), and above LT ⫹ 1 (zone 3) determined during laboratory testing were
calculated for power output and heart rate. Results: During the five mass start stages, a mean power output of 220 ⫾ 22 W (3.1 ⫾
0.2 W䡠kg⫺1) with a mean heart rate of 142 ⫾ 5 bpm was measured. Average power output during an uphill time trial was 392 ⫾ 60
W (5.5 ⫾ 0.4 W䡠kg⫺1) with a mean heart rate of 169 ⫾ 3 bpm. For the mass start stages, the average distribution of exercise time spent
in different intensities calculated for power output and heart rate was 58 versus 38% for zone 1, 14 versus 38% for zone 2, and 28 versus
24% for zone 3. Conclusion: Most of the competition time during the mass start stages was spent at intensities near the LT. Compared
with power output, heart rate measurement underestimated the time spent at intensity zones 1 and 3, and overestimated the time spent
in zone 2. Key Words: PERFORMANCE, SRM, HEART RATE, PROFESSIONAL BICYCLING

T
he demands in professional cycling are miscella-
neous, from cold springtime classics on cobblestone
roads in Flanders to hot summer mountain stages in
the Pyrenees during the Tour de France. During competi-
tion, the cyclists cover daily distances from 1 to 300 km over
a period of 1 to 21 d and have an annual training volume up
to 35,000 km. Therefore, training has to be as specific as
possible to meet the physiological requirements of these
races.
Until present, the workload demands during professional
races were mainly estimated indirectly through heart rate
(2,4,14,15,18,19,22,23). This method disregards influences
on the heart rate like the “cardiovascular drift” or the posi-
tion on the bike (9,10,20).
The central variable to determine the demands during
cycling is the mechanical power output that is produced by
the cyclist to propel the bike (3). This variable can be
measured directly and more precisely on the bicycle using a
mobile crank dynamometer (6,12).
Address for correspondence: Stefan Vogt, MD, University of Freiburg,
Center of Internal Medicine Department of Preventive and Rehabilitative
Sports Medicine, Hugstetterstr. 55, D-79106 Freiburg, Germany; E-mail:
stefan.vogt@uniklinik-freiburg.de.
Submitted for publication March 2005.
Accepted for publication July 2005.
0195-9131/06/3801-0147/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE®
Copyright © 2006 by the American College of Sports Medicine
DOI: 10.1249/01.mss.0000183196.63081.6a
147
The purpose of this study was to determine power output
during professional road race competition using direct
power measurements and to compare these results with the
measurements using the traditional heart rate method. These
data might help to improve the development of adapted
training programs.
MATERIALS AND METHODS
Subjects. The data of six professional road cyclists
were studied. Their anthropometrical and physiological
characteristics are presented in Table l.
The total training workload performed during a year by
these cyclists ranges between 30,000 and 35,000 km. All
cyclists were members of the same professional cycling
team and had competed in at least one of the three major
tours (Tour de France, Giro d’Italia, or Vuelta a Espana). All
athletes gave their informed written consent to participate in
the study, which was approved by the scientific committee
of our department.
Laboratory testing. One day before the race each
cyclist performed an incremental cycling test on an elec-
tronically braked cycle ergometer (SRM Ergometer, SRM
Trainingsystems, Schoberer Rad Messtechnik, Jülich, Ger-
many). Positions of the handlebars, seat height, and crank
length were adjusted to the measures used by the athletes on
their racing bike.
The test started with an initial resistance of 100 W, with
further increments of 20 W every 3 min. Heart rate was
recorded telemetrically during the entire test (Polar S725,
Polar Electro Oy, Kempele, Finland). Capillary blood (20
␮L) from a hyperaemized ear lobe for lactate analysis
(EBIO plus, Eppendorf, Hamburg, Germany) was collected
during the last 30 s of every step.
As markers of effort, heart rate and power output at the
lactate threshold (LT) and at 1 mM above the lactate thresh-
old (LT ⫹ 1) were determined. The LT was taken from the
original definition of Wasserman et al. (30) at the workload
that corresponds to the start of the increase in blood lactate
concentration.
As a marker of a higher level of effort, various concen-
trations of lactate above the LT have been used, and some-
times referred to as the “individual anaerobic threshold”
(IAT) (24,25). Here, an increase of 1 mM above the LT is
used for this purpose, and referred to as LT⫹1. The athletes
did not perform until complete exhaustion during the test.
The test was carried out 24 h before the first stage of
competition; to allow maximal recovery for the race, an
effort until exhaustion was therefore not practicable. How-
ever, the effort was sufficient for the determination of sub-
maximal variables such as LT and LT ⫹ 1.
The peak workload (Ppeak) was determined as the highest
workload a cyclist maintained for a complete 3-min period.
When the last workload was not maintained for three full
minutes, Ppeak was computed as follows (13):
Ppeak ⫽ Wf ⫹ 关共t/180兲 ⫻ 20兴
where Wf is the value of the last complete workload (W), t
is the time the last incomplete workload was maintained (s),
and 20 is the power output difference (W) between the last
two workloads.
Field testing. During the six stages of the road race
competition, workload was continuously monitored through
the SRM-Trainingsystem (SRM Trainingsystems, Scho-
berer Rad Messtechnik, Jülich, Germany). The SRM system
continuously and reliably measures power output, heart rate,
cadence, velocity, and distance (6,12).
The SRM system is a crank-based device that measures
the mechanical power output to the bike through the mul-
tiplication of the torque applied to the cranks, and the speed
at which they turn. The crankset experiences tiny deforma-
tion when torque is applied. The SRM system measures this
deformation using strain gauges attached to components
inside the crank. The system weighs 300 g, comparable to a
conventional bicycle crank, and thereby does not influence
performance.
The measured torque and cadence values are digitalized
inside the crank, and converted to a high-frequency, pulse-
width modulated electrical signal. The data are transmitted
to a microcomputer on the handlebar, where the torque is
averaged over each complete pedal revolution and multi-
plied by the cadence to calculate the power reading using the
following equation:
Power output ⫽ 共关measured frequency ⫺ zero offset frequency兴
⫻ cadence ⫻ 2⌸兲/共slope ⫻ 60兲.
The slope for each SRM crank dynamometer is calculated
dynamically at the SRM factory. The zero offset frequency
148 Official Journal of the American College of Sports Medicine
was adjusted by the investigator before each trial. The total
mechanical energy expenditure is calculated by integrating
the power measurements over the entire training session.
The data can be transmitted from the microcomputer on the
handlebars to any commercial computer for processing.
The energy expenditure displayed by the SRM represents
the mechanical work output (J) that the rider applied to the
bicycle. To calculate the total energy expenditure from the
SRM data, these data have to be corrected by the efficiency
of cycling. Although individual differences occur, physio-
logical studies (1,21) have estimated the average mechanical
efficiency of a human body on a bicycle to range around
25%. Therefore, the energy expenditure obtained from the
SRM System was multiplied by four to estimate the total
energy expenditure in joules.
To analyze the time spent at different power outputs
during competition, the amounts of competition time (in
percentage of race time) spent below LT (zone 1), between
LT and LT ⫹1 (zone 2), and above LT ⫹ 1 (zone 3)
determined during laboratory testing were calculated for
power outputs and heart rate.
Race characteristics. The study took place during
the Regio-Tour International, an international stage race for
elite cyclists (UCI Classification 2.3). This 5-d stage race
covered 758 km in five mass start stages and one uphill time
[1]
trial. The winner covered this distance in 18 h 23 min at an
average speed of 41.1 km·h⫺1.
Statistical analysis. Data registered during the com-
petition by the SRM System were transmitted to a conven-
tional PC and processed with the software provided with the
SRM System. Descriptive statistics were used, with all data
expressed as mean ⫾ standard deviation (SD).
RESULTS
Laboratory testing. The anthropometrical and physi-
ological characteristics of the studied athletes and the results
of the incremental ergometer test are displayed in Table 1.
Race characteristics. The average power output,
power output per kilogram, heart rate, energy expenditure,
stage length, and altitude difference (accumulated altitude
gain obtained from race officials) per stage are displayed in
TABLE 1. Subject characteristics (N ⫽ 6) and ergometer test results.
Age (yr) 27.3 ⫾ 2.1
Height (cm) 183.7 ⫾ 7.1
Weight (kg) 71.3 ⫾ 6.5
Ppeak (W) 390 ⫾ 28
Ppeak/kg (W䡠kg⫺1) 5.5 ⫾ 0.3
PLT (W) 248 ⫾ 32
PLT/kg (W䡠kg⫺1) 3.5 ⫾ 0.2
PLT⫹1 (W) 326 ⫾ 38
PLT⫹1/kg (W䡠kg⫺1) 4.6 ⫾ 0.3
HRLT (bpm) 137 ⫾ 13
HRLT⫹1 (bpm) 163 ⫾ 10
Values are mean ⫾ SD. Ppeak, maximum power output; Ppeak/kg, maximum power
output per kilogram of body weight; PLT, power output at the lactate threshold; PLT/kg,
[2] power output per kilogram of body weight at the lactate threshold; PLT⫹1, power output
at a lactate increase of 1 mM above the LT; PLT⫹1/kg, power output per kilogram of body
weight at a lactate increase of 1 mM above the LT; HRLT, heart rate at the lactate
threshold; HRLT⫹1, heart rate at a lactate increase of 1 mM above the LT.
http://www.acsm-msse.org
TABLE 2. Results during competition.
Heart rate Power output Power output per Energy Length Altitude difference
Stage (bpm) (W) kilogram (W䡠kgⴚ1) expenditure (MJ) (km) (m)
1 141 ⫾ 7 190 ⫾ 9 2.7 ⫾ 0.3 11.5 ⫾ 0.8 167 650
2 140 ⫾ 10 205 ⫾ 32 2.9 ⫾ 0.2 13.4 ⫾ 2.1 164 1500
3 147 ⫾ 4 271 ⫾ 34 3.8 ⫾ 0.3 7.4 ⫾ 1.0 84 700
4 169 ⫾ 7 392 ⫾ 55 5.5 ⫾ 0.6 2.6 ⫾ 0.4 13 450
5 140 ⫾ 7 227 ⫾ 32 3.2 ⫾ 0.2 13.1 ⫾ 2.0 170 400
6 144 ⫾ 4 235 ⫾ 8 3.4 ⫾ 0.2 14.1 ⫾ 0.5 160 200

Table 2. Stage 4 was an uphill time trial; the other five
stages were carried out as mass start stages.
Power and heart rate during competition. During
the five mass start stages, a mean power output of 220 ⫾ 22
W (3.1 ⫾ 0.2 W䡠kg⫺1) was measured. The exercise intensity
ranged 28 W above the LT and 107 W under the LT ⫹ 1.
During these five stages an average energy expenditure of
11.9 MJ was measured. The average heart rate was 142 ⫾
5 bpm, 5 bpm above the LT and 21 bpm under the LT ⫹ 1.
Uphill time trial. Stage 4 was an uphill time trial with
a higher recorded power output and heart rate compared
with the mass start stages. The average power output was
392 ⫾ 60 W, corresponding to 5.5 ⫾ 0.4 W䡠kg⫺1. Heart rate
averaged 169 ⫾ 3 bpm during this stage.
Comparison of competition and laboratory test-
ing. For the mass start stages, the average distribution of
exercise time spent in different intensities calculated for
power output was 58% for zone 1, 14% for zone 2, and 28%
for zone 3. Calculated for heart rate, the cyclists spent 38%
in zone 1, 38% in zone 2, and 24% in zone 3 (Figure 1).
DISCUSSION
The purpose of this study was to determine the direct
power output and thereby estimate the workload demands
during a professional multistage road cycling race. Direct
power output most precisely describes performance in cy-
cling (3). Very few data investigating the direct power
output during cycling are available. One study analyzed the
direct power output during a time trial performed by ama-
teurs (26), another during a cyclo-cross race (8), and a third
during mountain bike races (27).
Our study is the first to evaluate the workload sustained
by professional elite road cyclists during a multistage race
using direct power measurements. To date, the intensity of
bicycle racing was mainly described through heart rate
profiles (4,15,17,18,22,23). However, this indirect method
to describe the intensity of exercise is somewhat imprecise
and problematic; heart rate might be significantly influenced
by the “cardiovascular drift” caused by hypohydration and
hyperthermia (9,20).
The participants of the present study are characterized by
a very high aerobic capacity, as shown by the power output
at the LT and LT ⫹ 1 (Table 1). These features have been
demonstrated for pro cyclists in various other investigations
(4,5,11,14,16,18,19,22,23). Compared with the above-men-
tioned studies, Ppeak of the participants in the present study
was lower because the incremental cycling test was not
performed until exhaustion. As shown by Urhausen et al.
(28), the determination of the LT and LT ⫹ 1 does not rely

FIGURE 1—Average distribution of exercise

time spent in different intensities calculated
for power output and heart rate.

POWER OUTPUT IN PROFESSIONAL ROAD CYCLING Medicine & Science in Sports & Exercise姞 149
on exhaustion. Furthermore, slight differences in the power
output data might be due to the different exercise testing
protocols between the present and the above-mentioned
studies.
Power Output and Heart Rate during Competition
Mass start stages. The average power output mea-
sured during all stages, including the uphill time trial, was
245 ⫾ 33 W (3.4 ⫾ 0.3 W䡠kg⫺1). For the five mass start
stages, an average power output of 220 ⫾ 22 W (3.1 ⫾ 0.2
W䡠kg⫺1) was measured. Hansen et al. (8) found higher
power readings in three cyclo-cross riders during a cyclo-
cross competition. The mean power output of these three
cyclists was 374, 316, and 309 W (4.7, 4.1, and 4.9
W䡠kg⫺1). The difference to the data in the present study
might be explained by the fact that cyclo-cross races are
shorter than road races. The studied cyclo-cross race had a
duration of 60 min versus a mean duration of 215 min for
the stages in the present investigation.
A study of Smith et al. (26) determined the direct power
output of eight amateur cyclists during a 40-km time trial.
The mean power output during approximately 60 min was
312 ⫾ 23 W. Stapelfeldt et al. (27) determined the direct
power output in 15 mountain bike races of 11 elite cyclists
(9 male, 2 female). Average power output during racing was
246 ⫾ 12 W (3.5 ⫾ 0.2 W䡠kg⫺1) for the male cyclists and
193 ⫾ 1 W (3.1 ⫾ 0.2 W䡠kg⫺1) for the female athletes for
a mean race duration of 128 min (male) and 108 min
(female).
A mean power output of 240 W measured with the SRM
Training System during a 6-h stage of the Tour de France in
a cyclist was slightly higher than our findings (10). How-
ever, the power output in that study was registered in a
mountain stage and was consequently obtained in a more
strenuous stage than the present data. Fernandez-Garcia et
al. (4) showed that competitive cyclists spent longer time
periods at higher intensities in mountain stages compared
with flat stages.
For indirect measurements, Padilla et al. (23) estimated
intensities of 246 ⫾ 44 W in high mountain stages of the
3-wk races (Giro d’Italia, Tour de France, Vuelta a Espana),
234 ⫾ 43 W in semimountainous stages, and 192 ⫾ 45 W
in flat stages through a heart rate–power output relationship.
Our directly measured data thereby confirm the previous
indirect estimations of mean power output during cycling
mass start races.
Uphill time trial. Padilla et al (22) investigated the
exercise intensity in several professional time trial races via
a heart rate–power relationship. The authors described mean
power outputs during prologue time trials (⬍10 km) of 380
⫾ 62 W, during short time trials (⬍40 km) of 362 ⫾ 59 W,
and during long time trials (⬎40 km) of 347 ⫾ 46 W.
During uphill time trials of 40.6 ⫾ 4.8 km and mean race
duration of 75 ⫾ 8.4 min, an average power output of 342
⫾ 32 W was calculated.
The uphill time trial in the present investigation had a
length of 13 km, an altitude difference of 450 m, and a mean
150 Official Journal of the American College of Sports Medicine
duration of 23 min. The mean power output during this
stage was 391 ⫾ 60 W (5.5 ⫾ 0.4 W䡠kg⫺1). Compared
with the study of Padilla et al. (22), the power output was
higher; however, this might be due to the difference in
race duration and variations in the stage profile or envi-
ronmental conditions.
The estimated power outputs of Lance Armstrong and Jan
Ullrich at the Alpe d’Huez uphill time trial during the Tour
de France 2004 were 445 and 440 W (6.3 and 6.0 W䡠kg⫺1).
The record holder of this 13.8-km climb with an altitude
difference of 1110 m is Marco Pantani, who had an esti-
mated power output of 388 W, corresponding to 6.9 W䡠kg⫺1
(14). These power outputs were estimated through the
length of the climb, race time, weight of rider and bike, and
altitude difference.
During the uphill time trial in the present investigation,
the measured mean power output was 66 W above the LT ⫹
1. The average energy expenditure on this stage was 2.6 ⫾
0.4 MJ. The mean heart rate on this stage was 169 bpm and
was approximately 30 bpm higher than during the mass start
stages. These data indicate that the uphill time trial is the
type of stage with the highest mean exercise intensity in
competitive road cycling.
Comparison of Competition and Laboratory
Testing
To date, heart rate is mostly used to describe exercise
intensity during competition (4,15,17,18,22,23). But, as
mentioned above, heart rate is influenced by several phys-
iological and psychological factors. Power output measured
directly on the bike using a mobile crank dynamometer
system (SRM crankset) seems to be less influenced by
internal and external factors and represents a most precise
description of performance delivered by the athlete (6,12).
To compare exercise intensity determined through the
traditional heart rate measurement with exercise intensity
determined by direct measurement of power output, exercise
intensity was divided into three zones (zones 1–3) based on
heart rate and power- and lactate measurement from labo-
ratory testing.
Figure 1 shows the distribution of power output and heart
rate in the three zones during the six stages. It seems that the
description of exercise intensity through heart rate tends to
underestimate the time spent in zones 1 and 3 and to over-
estimate the time spent in zone 2 (Fig. 1).
A reason for this might be that the regulation of the
cardiovascular system is slower to adapt to the quick
changes in high and low power output in decisive race
situations. During descents, for instance, heart rate can be in
zone 2 while power output is already in zone 1. Or, during
intermittent sprints, power output may be in zone 3 while
heart rate is still in zone 2. For the daily training routine, this
finding might indicate that because of the delay in the
response of heart rate to rapid changes in workload, work-
load prescriptions in watts might be more suitable to mon-
itor the pace of shorter intervals or sessions with frequent
pace changes.
http://www.acsm-msse.org
 The cardiovascular drift can be another explanation for
the underestimation of zone 1 using heart rate measure-
ments: heart rate tends to increase with ongoing exercise
because of hyperthermia and hypohydration. This increase
can reach up to 20 bpm compared with euhydration and
regular temperature (9,20). Furthermore, long-lasting exer-
cise can lead to a decline of the maximal heart rate, as
demonstrated in many investigations of overreaching and
overtraining (7,29).
CONCLUSION
In the present study, performance was measured directly
using a mobile crank dynamometer (SRM crankset) during
a professional multistage road cycling competition. Most of
REFERENCES
1. BIJKER, K. E., G. DE GROOT, and A. P. HOLLANDER. Delta efficien-
cies of running and cycling. Med. Sci. Sports Exerc. 33:1546–
1551, 2001.
2. BROUNS, F., W. H. SARIS, J. STROECKEN, et al. Eating, drinking, and
cycling. A controlled Tour de France simulation study, part I. Int.
J. Sports Med. 10 Suppl:S32–S40, 1989.
3. COYLE, E. F., M. E. FELTNER, S. A. KAUTZ, et al. Physiological and
biomechanical factors associated with elite endurance cycling
performance. Med. Sci. Sports Exerc. 23:93–107, 1991.
4. FERNANDEZ-GARCIA, B., J. PEREZ-LANDALUCE, M. RODRIGUEZ-
ALONSO, and N. TERRADOS. Intensity of exercise during road race
pro-cycling competition. Med. Sci. Sports Exerc. 32:1002–1006,
2000.
5. GARCIA-ROVES, P. M., N. TERRADOS, S. FERNANDEZ, and A. M.
PATTERSON. Comparison of dietary intake and eating behavior of
professional road cyclists during training and competition. Int.
J. Sport Nutr. Exerc. Metab. 10:82–98, 2000.
6. GARDNER, A. S., S. STEPHENS, D. T. MARTIN, E. LAWTON, H. LEE,
and D. JENKINS. Accuracy of SRM and power tap power monitor-
ing systems for bicycling. Med. Sci. Sports Exerc. 36:1252–1258,
2004.
7. HALSON, S. L., M. W. BRIDGE, R. MEEUSEN, et al. Time course of
performance changes and fatigue markers during intensified train-
ing in trained cyclists. J. Appl. Physiol. 93:947–956, 2002.
8. HANSEN, E. A., K. JENSEN, and K. KLAUSEN. The workload demands
in cyclo-cross. Cycle Coaching. 2:23–24, 1999.
9. HEAPS, C. L., J. GONZALEZ-ALONSO, and E. F. COYLE. Hypohydra-
tion causes cardiovascular drift without reducing blood volume.
Int. J. Sports Med. 15:74–79, 1994.
10. JEUKENDRUP, A., and A. VAN DIEMEN. Heart rate monitoring during
training and competition in cyclists. J. Sports Sci. 16:91–99, 1998.
11. JEUKENDRUP, A. E., N. P. CRAIG, and J. A. HAWLEY. The bioener-
getics of World Class Cycling. J. Sci. Med. Sport. 3:414–433,
2000.
12. JONES S. M., AND L. PASSFIELD. The dynamic calibration of bicycle
power measuring cranks. In: The Engineering of Sport. H. S. J.
(Ed.). Oxford, 1998, pp. 265–274.
13. KUIPERS, H., F. T. VERSTAPPEN, H. A. KEIZER, P. GEURTEN, and G.
VAN KRANENBURG. Variability of aerobic performance in the lab-
oratory and its physiologic correlates. Int. J. Sports Med. 6:197–
201, 1985.
14. LUCIA, A., C. EARNEST, and C. ARRIBAS. The Tour de France: a
physiological review. Scand. J. Med. Sci. Sports. 13:275–283,
2003.
15. LUCIA, A., J. HOYOS, A. CARVAJAL, and J. L. CHICHARRO. Heart rate
response to professional road cycling: the Tour de France. Int.
J. Sports Med. 20:167–172, 1999.
POWER OUTPUT IN PROFESSIONAL ROAD CYCLING
the competition time during the mass start stages was spent
in intensities below the LT ⫹ 1. The mean power output
ranged near the LT. When compared with power output,
heart rate measurement underestimated the time spent in
intensity zones situated below the LT and above the LT ⫹
1 (zones 1 and 3), and overestimated the time spent in the
moderate-intensity zone between LT and LT ⫹ 1 (zone 2).
The benefits of measuring workload through power
output include better quantification of intensity compared
with heart rate measurements. Training specificity can
therefore be improved through an optimized monitoring
of performance.
The authors thank the professional cycling Team Telekom for
their help and cooperation in carrying out this study.
16. LUCIA, A., J. HOYOS, and J. L. CHICHARRO. Physiology of profes-
sional road cycling. Sports Med. 31:325–337, 2001.
17. LUCIA, A., J. HOYOS, M. PEREZ, and J. L. CHICHARRO. Heart rate and
performance parameters in elite cyclists: a longitudinal study.
Med. Sci. Sports Exerc. 32:1777–1782, 2000.
18. LUCIA, A., J. HOYOS, A. SANTANA, C. EARNEST, and J. L. CHICHARRO.
Tour de France versus Vuelta a Espana: which is harder? Med. Sci.
Sports Exerc. 35:872–878, 2003.
19. LUCIA, A., J. HOYOS, A. SANTANA, C. P. EARNEST, and J. L. CHI-
CHARRO. Giro, Tour, and Vuelta in the same season. Br J Sports
Med. 37:457–459, 2003.
20. MONTAIN, S. J., and E. F. COYLE. Influence of graded dehydration
on hyperthermia and cardiovascular drift during exercise. J. Appl.
Physiol. 73:1340–1350, 1992.
21. MOSELEY, L., and A. E. JEUKENDRUP. The reliability of cycling
efficiency. Med. Sci. Sports Exerc. 33:621–627, 2001.
22. PADILLA, S., I. MUJIKA, J. ORBANANOS, and F. ANGULO. Exercise
intensity during competition time trials in professional road cy-
cling. Med. Sci. Sports Exerc. 32:850–856, 2000.
23. PADILLA, S., I. MUJIKA, J. ORBANANOS, J. SANTISTEBAN, F. ANGULO,
and J. JOSE GOIRIENA. Exercise intensity and load during mass-start
stage races in professional road cycling. Med. Sci Sports Exerc.
33:796–802, 2001.
24. ROECKER, K., O. SCHOTTE, A. M. NIESS, T. HORSTMANN, and H. H.
DICKHUTH. Predicting competition performance in long-distance
running by means of a treadmill test. Med. Sci. Sports Exerc.
30:1552–1557, 1998.
25. ROECKER, K., H. STRIEGEL, and H. H. DICKHUTH. Heart-rate recom-
mendations: transfer between running and cycling exercise? Int. J.
Sports Med. 24:173–178, 2003.
26. SMITH, M. F., R. C. DAVISON, J. BALMER, and S. R. BIRD. Reliability
of mean power recorded during indoor and outdoor self-paced 40
km cycling time-trials. Int. J. Sports Med. 22:270–274, 2001.
27. STAPELFELDT, B., A. SCHWIRTZ, Y. O. SCHUMACHER, and M. HILL-
EBRECHT. Workload demands in mountain bike racing. Int.
J. Sports Med. 25:294–300, 2004.
28. URHAUSEN, A., B. COEN, B. WEILER, and W. KINDERMANN. Individ-
ual anaerobic threshold and maximum lactate steady state. Int.
J. Sports Med. 14:134–139, 1993.
29. URHAUSEN, A., H. H. GABRIEL, and W. KINDERMANN. Impaired
pituitary hormonal response to exhaustive exercise in overtrained
endurance athletes. Med. Sci. Sports Exerc. 30:407–414, 1998.
30. WASSERMAN, K., J. E. HANSEN, D. Y. SUE, R. CASABURI, AND B. J.
WHIPP. Principles of Exercise Testing and Interpretation. Third
Edition. Baltimore: Lippincott Williams & Wilkins, 1999.
Medicine & Science in Sports & Exercise姞 151