Comparison between running performance in time trials on track and treadmill

Abstract

Few studies have investigated the influence of test environment (field vs. laboratory) on pacing strategy and on physiological variables measured during endurance running performance tests. The objective of this study was therefore to compare the behavior of mean velocity (MV), pacing strategy, heart rate (HR) and rating of perceived exertion (RPE) during one-hour running time trials conducted on an athletics track with the behavior of the same variables during one-hour running time trials conducted on a treadmill. Eighteen male recreational runners (25.4 ± 3.3 years) performed two one-hour time trials; the first running on a treadmill and the second on a 400 m athletics track. Rating of perceived exertion and HR were recorded every 10 minutes and MV was calculated every 15 minutes for analysis of pacing strategy (0-15min; 15-30min; 30-45min; and 45-60min). These performance variables were compared using Student's t test for paired samples. Figures for MV, HR and RPE measured at different points during the trials were compared using two-factor ANOVA for repeated measures, followed by Bonferroni's post hoc test. A significance level of P < 0.05 was adopted for all analyses. Mean velocity was higher for the trials on the running track (12.2 ± 0.8 km·h-1) than for the trials on the treadmill (11.8 ± 0.8 km·h-1). Additionally, there were also differences between the two test environments for mean and maximum heart rate, and in terms of pacing strategy. On the basis of these differences, it can be concluded that performance was influenced by the environment in which the one-hour time trials were conducted.

Full text

DOI: http://dx.doi.org/10.5007/19800037.2014v16n4p456
original article
Licence
Creative Commom
CC
BY
RBCDH
1 Universidade Estadual de Marin-
gá. Programa de Pós Graduaç ão
Associado em Educação Física
UEM/UEL. Maringá, PR. Brasil.
Received: 21 June 2013
Accepted: 11 February 2014
Comparison between running performance
in time trials on track and treadmill
Comparação entre desempenhos de corrida time trial
realizados em pista e esteira
Cecília Segabinazi Peserico1
Fabiana Andrade Machado1
Abstract – Few studies have investigated the inuence of test environment (eld vs.
laboratory) on pacing strategy and on physiological variables measured during endur-
ance running performance tests. e objective of this study was therefore to compare the
behavior of mean velocity (MV), pacing strategy, heart rate (HR) and rating of perceived
exertion (RPE) during one-hour running time trials conducted on an athletics track with
the behavior of the same variables during one-hour running time trials conducted on a
treadmill. Eighteen male recreational runners (25.4 ± 3.3 years) performed two one-hour
time trials; the rst running on a treadmill and the second on a 400 m athletics track.
Rating of perceived exertion and HR were recorded every 10 minutes and MV was calcu-
lated every 15 minutes for analysis of pacing strategy (0-15min; 15-30min; 30-45min; and
45-60min). ese performance variables were compared using Student’s t test for paired
samples. Figures for MV, HR and RPE measured at dierent points during the trials were
compared using two-factor ANOVA for repeated measures, followed by Bonferroni’s post
hoc test. A signicance level of P < 0.05 was adopted for all analyses. Mean velocity was
higher for the trials on the running track (12.2 ± 0.8 km·h-1) than for the trials on the
treadmill (11.8 ± 0.8 km·h-1). Additionally, there were also dierences between the two
test environments for mean and maximum heart rate, and in terms of pacing strategy.
On the basis of these dierences, it can be concluded that performance was inuenced
by the environment in which the one-hour time trials were conducted.
Key words: Athletic performance; Environment; Running.
Resumo – Poucos estudos vericaram a inuência do ambiente de teste (campo e labora-
tório) sobre o ritmo de corrida e variáveis siológicas obtidas durante o desempenho em
corrida de endurance. Portanto, o objetivo deste estudo foi comparar o comportamento da
velocidade média (VM), ritmo de corrida, frequência cardíaca (FC) e percepção subjetiva de
esforço (PSE) obtidas durante os desempenhos de corrida em provas time trial de uma hora
realizados em pista de atletismo e em esteira. Dezoito homens corredores recreacionais (25,4
± 3,3 anos) realizaram duas performances de uma hora de corrida: uma em esteira e outra
em pista de atletismo de 400 m. A PSE e a FC foram registradas a cada 10 minutos, e a VM
a cada 15 minutos para a determinação do ritmo de corrida. As variáveis relacionadas aos
desempenhos foram comparadas pelo teste t de Student para amostras pareadas. Os valores
de VM, FC e PSE obtidos durante diferentes momentos das provas foram comparados pela
Anova de dois fatores para medidas repetidas seguido do post hoc de Bonferroni. Para todas
as análises, foi adotado nível de signicância de P< 0,05. A VM da prova realizada em pista
(12,2 ± 0,8 km·h-1) foi superior à prova em esteira (11,8 ± 0,8 km·h-1). Além disso, foram
encontradas diferenças entre os dois desempenhos para os valores de frequência cardíaca
média e máxima, e para o ritmo de corrida. A partir dessas diferenças, conclui-se que os de-
sempenhos foram inuenciados pelo ambiente onde as provas de uma hora foram realizadas.
Palavras-chave: Corrida; Desempenho atlético; Meio ambiente.

Rev Bras Cineant ropom Desempenho Hum 2014, 16(4):456-46 4 457
INTRODUCTION
Long distance (endurance) runners’ performance is usually evaluated in
time trials, in which participants either attempt to cover a xed distance
in the shortest time possible or attempt to cover the greatest distance pos-
sible in a xed time1. Assessing runners allows researchers and trainers
to simulate sporting performance and/or investigate elements related to
performance in a controlled manner, making it possible to choose certain
variables and use them to monitor athletes’ progress, with the objective of
setting targets for performance improvements2.
In this context, many studies have employed a one-hour test to assess
performance in endurance running because it is representative of a range
of the dierent competitions in which long distance runners compete, and
because it is a test that has demonstrated a high degree of reproducibility
for assessment of endurance runners3-6.
With regard to test environments, time trials can be conducted in
the eld (athletics track) or in laboratories (treadmill). Studies that have
investigated dierences in running patterns between the two have found
dierences between trials conducted on a track or treadmill in relation to
biomechanical aspects, maximum velocity attained and perceived veloc-
ity7-10. Nummela et al.9 and Morin and Seve8 analyzed maximal sprint tests
conducted on treadmill or track and found that maximum velocity attained
was statistically higher on the track than on the treadmill. However, a
large proportion of published studies have assessed performance in short
duration tests, i.e. sprints, and there has so far been little study of the dif-
ference between performance in endurance time trials conducted in the
eld and in the laboratory.
Another important aspect to be taken into account when assessing
performance in endurance tests is the pace or race strategy adopted by
the runner. Studies that have analyzed pacing strategy and the behavior of
physiological variables during endurance running have found that the best
strategy that would make best performance possible is not established11-13.
Additionally, factors such as the capacity of energy systems, the runner’s
experience, duration of the trial and environmental conditions all have
an eect on the choice of pacing strategy12, 14. However, no studies have
investigated the inuence of test environment (eld or laboratory) on
pacing strategy and on the behavior of physiological variables recorded
during endurance running.
e objective of this study was therefore to compare the behavior
of mean velocity (MV), pacing strategy, heart rate (HR) and rating of
perceived exertion (RPE) during one-hour time trials run on an athletics
track and on a treadmill. Our hypothesis is that the MV for the trial per-
formed on the track will be greater than for the treadmill trial and that
pacing strategy and the behavior of HR and RPE will be inuenced by
test environment.

458
Running per formance on track and trea dmill Peserico & Machado
METHODOLOGICAL PROCEDURES
Participants
Eighteen male recreational endurance runners volunteered to take part in
the study. ey had experience of local and/or regional 10 km competitions
and were on a systematic training program (age 25.4 ± 3.3 years; height
178.0 ± 0.7 cm; body mass 76.2 ± 8.6 kg; body mass index (BMI) 24.1 ± 2.3
kg·m-2 and body fat percentage (%G) 13.9 ± 3.0 %). ey had started running
an average of 3.8 ± 3.3 years previously, were training an average of 3.2 ±
1.1 days week-1, covering an average distance of 26.9 ± 16.6 km·week-1. Ad-
ditionally, 15 of the 18 volunteers trained in open areas such as parks and
streets. Before any experimental procedures were conducted, participants
signed free and informed consent forms and the research project was ap-
proved in advance by the Research Ethics Committee at the Universidade
Estadual de Maringá, under protocol number 719/2010.
Experimental Design
e volunteers were already accustomed to conducting trials both on tread-
mills and on running tracks. ey undertook two performance tests in the
form of one-hour time trials, the rst running on an automatic program-
mable ergometric treadmill (INBRAMED Super ATL, Porto Alegre – Brazil)
and the second on a 400 m outdoor athletics track. e sequence of trials
was dictated by the availability of venues. An interval of one week was le
between tests. Participants were instructed to present for the tests in a well-
hydrated state and to continue eating their usual diet, eating breakfast as
normal before all tests, and to abstai n from consuming alcohol or caeine for
24 hours before assessments. Participants were also asked to suspend their
training routines during the test period. All assessments were conducted
in the mornings with the objective of maintaining similar experimental
conditions throughout. Additionally, treadmill trials were conducted in an
air-conditioned laboratory (with temperature set at 20 °C to 24°C and relative
humidity bet ween 50 and 60%), while track tria ls were conducted at tempera-
tures ranging from 16°C to 26°C, with humidity ranging from 60 to 80%.
Performance of one-hour time trials (track and treadmill)
Both one-hour time trials were preceded by a ve-minute warm-up period,
at 6 km·h-1 on the treadmill and as participants preferred on the track. For
both types of trial, participants were requested to attempt to run as far as
possible in one hour and the total distance achieved was recorded. e over-
all MV for each trial was calculated by dividing the total distance covered
by the trial duration. Additionally, partial MVs were calculated for each 15
minutes in order to prole their pacing strategy at four points during the
trial (0-15min; 15-30min; 30-45min; 45-60min). Before the tests, participants
were familiarized with the Borg 0-20 scoring scale15 used to determine their
rating of perceived exertion (RPE) during the trials. Additionally, heart rate
was monitored throughout all trials using a heart rate monitor (Polar RS800).

Rev Bras Cineant ropom Desempenho Hum 2014, 16(4):456-46 4 459
e values of HR and RPE were recorded every 10 minutes. In treadmill
trials, RPE was measured without pauses, whereas on the track RPE was
recorded when participants passed points chosen by the evaluator, which were
spaced approximately 10 minutes apart in time. Mean heart rate (HRmean)
was calculated as the mean of HR values recorded and maximum heart
rate (HRmax) was the highest rate recorded at any point during the trial. e
RPE value recorded at the end of the trial was taken as nal RPE (RPEnal).
e information given to participants during the trials was limited in
order to reduce the inuence of any other variables on results1,3. During
the track time trial, participants were only informed of the time that had
passed every 5 minutes, by the researcher, while the treadmill provided
participants with visual feedback consisting of time and velocity, with
inclination set at 1%16. Before the tests, participants had been taught how
to control the velocity of the treadmill. Each trial was initiated with the
treadmill velocity set at 8 km·h-1 and thereaer participants self-selected
their pace until the end of the trial. Mineral water was provided ad libi-
tum in cups throughout both types of trial, so that runners could hydrate
themselves as they are used to doing in long-distance races.
Statistical analysis
Data were expressed as mean ± standard deviation (SD); normality of data
was veried using the Shapiro-Wilk test. Variables related to performance
on track and treadmill were compared using Student’s t test for paired
samples. Results for MV, HR and RPE recorded at dierent points during
the track and treadmill trials were compared using two-factor ANOVA
for repeated measures followed by the Bonferroni post hoc test for mul-
tiple comparisons. e assumption of sphericity was veried using the
Mauchly test and when violated the degrees of freedom were corrected
using Greenhouse-Geisser sphericity estimates. Analyses were conducted
with the aid of the Statistical Package for the Social Sciences version 17.0
(SPSS). For all analyses, a signicance level of P < 0.05 was adopted.
RESULTS
Table 1 lists the results for variables related to performance in the one-hour
time trials on treadmill and track. e MV for the treadmill trial was sta-
tistically slower than the MV for the track trial (P = 0.001). Additionally,
values of HRmed (bpm) and HRmax (bpm) were statistically dierent between
the two trial types (P = 0.016 and 0.030, respectively). ere was no dier-
ence in RPEnal (6-20) between the two types of test (P > 0.05).
Pacing strategy was analyzed by calculating MV every 15 minutes,
i.e. for 0-15min, 15-30min, 30-45min and 45-60min. Two-factor ANOVA
for repeated measures detected a main eect on mean velocities calcu-
lated every 15 minutes from the test environment (P = 0.001) and from
stage of time trial (time elapsed) (P = 0.021). e gures shown in Table 2
demonstrate that there was a dierence between treadmill and track trials

460
Running per formance on track and trea dmill Peserico & Machado
in terms of mean velocity during the rst quarter (0-15min) of the trials.
Additionally, during the treadmill time trial, the pacing strategy was to
increase MV progressively, whereas on the track the behavior of MV was
constant. Furthermore, there was also an interaction between test environ-
ment and stage of time trial (time elapsed) (P < 0.0 01).
Tab le 1. Compari sons between result s for one-hour time tri als on treadmill and running tr ack. (n=18)
Treadmill Track
MV (km·h-1) 11.8 ± 0.8 12.2 ± 0.8*
HRmean (bpm) 175 ± 7.8 178 ± 7.5*
HRmax (bpm) 188 ± 7.1 184 ± 8.1*
RPE
nal
(6-20) 19 ± 0.8 19 ± 1.1
MV, mean velocity; HRmean, averag e heart rate; HRmax, maximum he art rate; RPEnal, maximum ratin g of perceived
exertion; * P < 0.05 compared to the t readmill trials.
Tab le 2. Mea n velocities (MV), calculated for 15-min ute intervals, during the o ne-hour time trials on tr eadmill
and track (n=18).
0-15 m in 15- 30 mi n 30-45 min 45-60 min
MV Treadmill (km·h-1) 10.9 ± 1.0 11.8 ± 1. 0a11.9 ± 0.8a12.5 ± 0.9a
MV Track (km·h-1) 12.6 ± 1.0* 12.2 ± 0.9 11.9 ± 0.9 12.1 ± 1.0
* Dierence (P < 0.05) bet ween track and treadmill. a Di erence (P < 0.05) with relationshi p to 0-15 min of
treadmill trial.
Table 3 lists the results recorded for HR and RPE every 10 minutes dur-
ing the one-hour time trials on treadmill and track. Two-factor ANOVA for
repeated measures demonstrated that both the environment in which trials
were conducted (P = 0.016) and stage of time trial (time elapsed) (P < 0 .0 01)
had a main eect on HR recorded every 10 minutes and also showed that
there was an interaction (P < 0.001) between the two independent factors.
e RPE scores were also aected by test environment (P = 0.005) and stage
of time trial (P < 0.001), and these factors also exhibited an interaction (P
= 0.013). Additionally, it was observed that HR was statistically dierent
between track and treadmill at the 10th minute, while RPE was dierent
at the 40th. Furthermore, in the treadmill trials HR increased constantly
over time and RPE exhibited the same behavior in both types of trial.
Tab le 3. Result s for heart rate (HR) and ratin g of perceived exertion (RPE ) recorded every 10 minutes dur ing
one-hour ti me trials on treadmill and trac k (n=18).
Time (min)
10 20 30 40 50 60
HR (bpm)
Treadmill 160 ± 12.7 169 ± 11.0a174 ± 8.9a178 ± 8.3a,b 182 ± 6.7a-d 188 ± 7.1a-e
Track 174 ± 9.2* 176 ± 8.9 177 ± 9.1 180 ± 8.1 181 ± 7.5 183 ± 8.7
RPE (6-20)
Treadmill 8 ± 1.8 11 ± 2.0a12 ± 1.8a,b 14 ± 1.9a-c 17 ± 1.8 a-d 19 ± 0.8 a-e
Track 9 ± 1.4 11 ± 1.4a13 ± 1.7 a,b 15 ± 2.0* a-c 18 ± 1.6 a-d 19 ± 1.0 a-e
* Dierence (P < 0.05) bet ween track and treadmill; a dierence (P < 0.05) with re lation to 10 min point of
same trial; b dierence (P < 0.05) in relatio n to 20 min point of same trial; c dierence (P < 0.05) in relation to
30 min point of same t rial; d dierence (P < 0.05) in relation to 40 min point of sam e trial; and dierence (P <
0.05) in relation to 50 min po int of same trial.

Rev Bras Cineant ropom Desempenho Hum 2014, 16(4):456-46 4 461
DISCUSSION
e objective of the present study was to compare the behavior of mean
velocity (MV), pacing strategy, heart rate (HR) and rating of perceived
exertion (RPE) during one-hour running time trials conducted either on
an athletics track or on a treadmill. e principal nding of this study was
that performance was inuenced by the environment in which time trials
were conducted, since MV for the trial conducted on track was faster than
MV for the treadmill trial. Additionally, runners exhibited dierences
in the variables HRmea n, HRm ax, and pacing strategy when tested on track,
compared to the trials conducted on the treadmill.
ese dierences observed between one-hour time trial performance
(track and treadmill) add weight to the results of other studies which have
compared short-duration and high-intensity running (sprinting) and
found that performance on a treadmill was slower than performance on a
running track8,9. In the light of this, some studies have suggested that the
fact that the test environment appears more monotonous and less attrac-
tive in a laboratory may result in worse performance in treadmill tests9,10.
Additionally, Milgrom et al.17 claim that whereas running on a treadmill
involves repetition of the same body kinematics, running on a track involves
frequent changes of direction, rhythm and stride, which makes running on
the track more motivating than running on a treadmill. Specically with
relation to our study, it is possible that the fact that the runners investigated
here are more familiar with the track than the treadmill, since they train
in open spaces, may have inuenced the results.
Another important element that may have had an inuence on our
ndings is the dierence in perceived running velocity when being tested
on a treadmill or on a track7,18 . In a study conducted by Kong et al.7, par-
ticipants rst ran on a track at their preferred velocity (self-selected) and
then immediately aerwards attempted to reproduce the same velocity on
a treadmill for three minutes, blinded to the velocity shown on the display.
e results showed that average velocity was 27.1% slower and that the
need for greater balance and coordination, fear of falling o and increased
demands on attention and vision may all be related to the perception of
higher velocity on a treadmill.
In addition to MV, both HRma x and HRmean were also statistically
dierent between trials conducted in the two dierent test environments.
e fact that HRmax was greater for the treadmill trials may be because the
participants increased in their pace at the end of the test, oen described as
a sprint nish. Studies involving one-hour time trials on a treadmill with
constant visua l feedback of time elapsed have shown that there is a tendency
for runners to distribute their energy reserves along the 60-minute run in
such a way as to be able to increase velocity, i.e. to sprint, during the nal
minutes of the trial3,19.
Both our observations and reports from the volunteers who took part in
our study suggest that the fact that they were able to see both their velocity

462
Running per formance on track and trea dmill Peserico & Machado
and elapsed time throughout the test when running on the treadmill, but
were only informed of elapsed time every ve minutes when running on
track, may have made them more cautious when deciding at which point
to increase the treadmill velocity, which they generally le until the end of
the trial. On this basis, one limitation of this study is the fact that partici-
pants were only able to monitor elapsed time constantly when running on
the treadmill, which in turn may have had an inuence on their dierent
behavior at the end of the two types of trial.
It was observed that HRmean was greater when time trials were run on
the track, primarily because of the greater intensity of track trials (in terms
of MV). Additionally, temperature appears to have inuenced the behavior
of HR during the trials, since, in contrast to the treadmill trials which were
conducted under temperature-controlled laboratory conditions, when
running on the track participants were exposed to the sun and to heat,
which may have physiologically altered their HR response, accentuating
cardiovascular dri, and increased their HR f urther still as a consequence20.
Another important nding was that the pacing strategy, analyzed in
terms of mean velocity at dierent points during the one-hour time trial,
diered between treadmill and track. When running on a treadmill, the
participants adopted a pacing strategy that was progressive throughout
the trial. A similar strategy has been described by Schabort et al.6, who
conducted a study with eight trained runners (27 years old and VO2max of
66 mL·kg-1 min-1) who performed one-hour time trials on a treadmill and
showed that MV increased over the rst 30 minutes, then stabilized at that
intensity until the 50th minute, before once more increasing progressively
up to the end of the trial. In contrast, Rollo et al.3 conducted one-hour time
trials on a treadmill with 10 experienced runners (32 years old and VO2max
of 61 mL·kg-1 min-1), observing a constant pacing strategy in which MV
remained similar from the second to the 59th minute of the test. e pacing
strategy adopted by the participants in the present study when running
on the athletics track was dierent to the strategy they employed on the
treadmill, and ts the pattern that Abbis and Laursen 12 have described
as a parabolic strategy. In other words, one in which the start of the time
trial is run at high velocity, followed by a progressive decrease during the
run, followed by an increase in velocity towards the end. is strategy has
been observed in studies analyzing pacing strategy in recreational runners
and also among competitors in top-level 10 km races13,21.
It therefore appears that factors related to the environment in which
trials are conducted may aect the choice of pacing strategy adopted in
performance tests. However, it is also possible that other factors may have
had an eect on the dierent pacing strategies adopted by the participants.
ese factors include the constant visual feedback of velocity and elapsed
time which was only available during the treadmill trial and the failure to
randomize conditions, i.e., the fact that the trials on the track were con-
ducted aer the treadmill trials, possibly leading to familiarization with
the time trial test format.

Rev Bras Cineant ropom Desempenho Hum 2014, 16(4):456-46 4 463
With regard to the results of the analysis of HR and RPE, measured
every 10 minutes during the trials, it was demonstrated that only HR at the
10th minute and RPE at the 40th minute were statistically dierent between
track and treadmill. It was also observed that over the course of the time
trials RPE increased signicantly irrespective of test environment and HR
increased signicantly during treadmill trials. e few studies that have
analyzed these variables during endurance runs have also found signicant
increases between the start and end of the test11,13 .
Finally, certain limitations of this study should be borne in mind,
including the failure to randomize the order of trials, the constant visual
feedback of velocity and elapsed time only available during treadmill trials;
the fact that wind speed was not measured during track tests; the minor
dierences in temperature and humidity between test environments; and
the dierences in warm-up protocols before the two types of trial.
CONCLUSIONS
Concluding, the differences observed between one-hour time trials
conducted on treadmill or on the running track in terms of MV, HRmed,
HRmax and pacing strategy show that the test environment (in the eld or
in a laboratory) had an inuence on the results. In general, the variables
measured during the running performance tests conducted here, such as
trial MV and HR (particularly HRmax), are parameters used to control and
prescribe training intensity, which in turn shows that the fact that these
variables exhibit dierent responses in dierent test environments implies
that changing environment will change the prescription and, consequently,
lead to dierent physiological adaptations. erefore, once it is known that
MV and HR may dier between tests conducted on the track or on tread-
mills, when coaches and athletes are choosing the environment in which
they will conduct tests, they must take into account both their objectives
and the environments in which the athletes train.
Acknowledgements
is research received support from the Coordenação de Aperfeiçoamento
de Pessoal de Nível Superior - CAPES, Brazil.
REFERENCES
1. Laursen PB, Francis GT, Abbiss CR, Newton MJ, Nosaka K. Reliability of time-
to-exhaustion versus time-trial running tests in runners. Med Sci Sports Exerc
2007;39(8): 1374-9.
2. Currell K, Jeukendrup AE. Validity, Reliability and Sensitivity of measures of
sporting performance. Sports Med 2008;38(4):297-316.
3. Rollo I, Williams C, Nevill A. Repeatability of scores on a novel test of endurance
running performance. J Sports Sci 2008;26(13):1379-86.
4. Loin M, Sothern M, Koss C, Tuuri G, Vanvrancken C, Kontos A, et al. Energy
expenditure and inuence of physiologic factors during marathon running. J
Strength Cond Res 2007;21(4):1188-91.

464
Running per formance on track and trea dmill Peserico & Machado
464
Corresponding author
Fabiana Andrade Machado.
Depto. de Educaç ão Física – Bloco M
06, Sala 6.
Campus Universitário
Universidade Estadual de Maringá
Av. Colombo, 5.790
CEP 87.020-900 – Maringá, PR. Brasi l.
Email: famachado_uem@hotmail.com
5. Gamelin FX, Coquart J. Ferrari N, Vodugnon H, Matran R, Leger L, et al. Predic-
tion of one hour performance using Constant duration tests. J Strength Cond Res
2006;20(4): 735-9.
6. Schabort EJ, Hopkins WG, Hawley JA. Reproducibility of self-paced treadmill
performance of trained endurance runners. Int J Sports Med 1998;19(1):48-51.
7. Kong PW, Koh TMC, Tan WCR, Wang YS. Unmatched perception of speed when
running overground and on a treadmill. Gait Posture 2012;36(1):46-8.
8. Morin JB, Se ve P. Spri nt runn ing performance: c omparison b etween treadmi ll and
eld conditions. Eur J Appl Physiol 2011;111(8):1695-703.
9. Num mela A, Hamalai nen I, Rusko H. Comparison of ma ximal anaerobic runn ing
tests on a treadmill and track. J Sports Sci 2007;25(1):87-96.
10. Schache AG, Blanch PD, Rath DA, Wrigley TV, Starr R, Be nnel KL. A compar ison of
overground and t readmill r unning for measu ring the thre e-dimensional k inematics
of the lumbo– pelvic–hip complex. Clin Biomech 2001;16(8):667-80.
11. Loin M, Sothern M, Tuuri G, Tompkins C, Koss C , Bonis M. Gender comparison
of physiologic and perceptual responses in recreational marathon runners. Int J
Sports Physiol Perform 2009;4(3):307-16.
12. Abbiss CR, Laursen PB. Describing and understanding pacing strategies during
athletic competition. Sports Med 2008;38(3):239-52.
13. Bertuzzi RSM, Nakamura FY, Rossi LC, Kiss MAPD, Franchini E. Independência
temporal das respostas do esforço percebido e da frequência cardíaca em relação à
velocidade de corrida na simu lação de uma prova de 10 km. Rev Bras Med Esporte
2006; 21(4):179-83.
14. Tucker R, Noakes TD.e physiological regulation of pacing strategy during exer-
cise: a critical review. Br J Sports Med 2009;43(6):1-9.
15. Borg GA. Psyc hophysical ba ses of perceived exertion. Med S ci Sports Exerc 1982;14
(5):377-81.
16. Jones AM, Doust JH. A 1% treadmill grade most accurately reects the energetic
cost of outdoor running. J Sports Sci 1996;14(4):321-7.
17. Milgrom C, Finestone A, Segev S, Olin C, Arndt T, Ekenman I. Are overground
or treadmill runners more likely to sustain tibial stress fracture? Br J Sports Med
2003;37(2): 160-3.
18. Kong PW, Candelaria NG, Tomaka J. Perception of self-selected running speed
is influenced by the tread mill but not footwear. Sports Biomech 2009;8(1):52-9.
19. Whitham M, Mckinney J. Eect of a carbohydrate mouthwash on running time
trial performance. J Sports Sci 2007;25(12):1385-92.
20. Lamber t MI, Mbambo ZH, Gibson S C. Heart rate dur ing traini ng and competition
for long-distance running. J Sports Sci 1998;16:S85-90.
21. iel C, Foster C, Banzer W, Koning J. Pacing in Olympic track races: competitive
tactics versus best performance strategy. J Sports Sci 2012;30 (11):1109-15.