Validity of using functional threshold power and intermittent power to predict cross-country mountain bike race outcome

Article (PDF Available) · January 2014with 1,153 Reads
Abstract Validity of using functional threshold power and intermittent power to predict cross-country mountain bike race outcome. Purpose : Field tests are important for athletes and sport practitioners as they offer valuable information on performance without demanding the time and cost to visit a laboratory. This study tested the ability of relative functional threshold power (FTP) and intermittent power (IP) field-tests to be used as predictors of cross-country mountain bike (XC-MTB) race finishing time (RT). Methods : Eleven well-trained male XC-MTB cyclists (mean age: 35.8 ± 8.2 yr; mean mass: 80.8 ± 13.4 kg) volunteered for this study. Relative (W/kg) FTP and relative IP were collected from field tests with the mean of all work intervals was recorded as IP and FTP calculated from 95% of mean maximal 20-minute power. RT was collected during a mass-start 17.4 km simulated XC-MTB race. Results : Both IP (r 2 =0.786) and FTP (r 2 =0.736) models were able to significantly predict RT (p < 0.001). However, the prediction errors were less when using Relative IP than Relative FTP (273.5 s versus 303.6 s). Conclusion : A field-based IP test can be used as a benchmark for the determination of XC-MTB athlete ability and preparedness. Athlete improvements can be tracked using an IP test.
J Sci Cycling. Vol. 3(1), 16-20
© 2014 Miller; licensee JSC. This is an Open Access article distributed under the terms of the Creative Commons Attribution License
(, which permits unrestricted use, distribution, and reproduction in any medium, provided the original
work is properly cited.
Open Access
Validity of using functional threshold power
and intermittent power to predict cross-
country mountain bike race outcome
Matthew C Miller 1, 2, Gavin L Moir 1 and Stephen R Stannard2
Validity of using functional threshold power and intermittent power to predict cross-country mountain bike race
outcome. Purpose: Field tests are important for athletes and sport practitioners as they offer valuable information on
performance without demanding the time and cost to visit a laboratory. This study tested the ability of relative
functional threshold power (FTP) and intermittent power (IP) field tests to be used as predictors of cross-country
mountain bike (XCO-MTB) race finishing time. Methods: Eleven well-trained male XCO-MTB cyclists (mean age:
35.8 ± 8.2 yr; mean mass: 80.8 ± 13.4 kg) volunteered for this study. Relative (W/kg) FTP and relative IP were
collected from field tests with the mean of all intermittent work intervals recorded as IP and FTP calculated from 95%
of mean maximal 20-minute power. Race performance time was collected during a mass-start 17.4 km simulated
XCO-MTB race in the field. Results: Both IP (r2=0.786) and FTP (r2=0.736) models were able to significantly predict
race performance (p < 0.001). However, the prediction errors were less when using Relative IP than Relative FTP
(273.5 s versus 303.6 s). Conclusion: A field-based IP test can be used as a benchmark for the determination of
XCO-MTB athlete ability and preparedness. Considering IP can be measured on a stationary trainer in any location
and independent of expensive equipment, coaches can easily use this model to track athlete training.
Keywords: cycling, mountain bike, off-road, functional threshold power, cross country, intermittent power
Contact email: (MC. Miller)
1 East Stroudsburg University, East Stroudsburg, PA, USA
2 Massey University, Palmerston North, New Zealand
Received: 14 April 2014. Accepted: 6 June 2014.
Cross-country mountain bike racing (XCO-MTB) is a
popular sport among both elite and recreational cyclists
alike. The physiological demands of competition and
thus predictors of performance are less understood in
XCO-MTB than in road cycling. Research available
shows that power output during XCO-MTB racing to
have a high degree of oscillation due to terrain and race
course conditions (Impellizzeri & Marcora, 2007;
MacDermid & Stannard, 2012; Stapelfeldt et al., 2004).
Accordingly, the literature suggests XCO-MTB
requires high rates of aerobic and anaerobic energy
production with an average heart rate during
competition to be 90% of maximum corresponding
with 84% of maximum oxygen uptake (Impellizzeri et
al., 2002). Some 82% of total race time is spent above
the lactate threshold (LT) (Impellizzeri & Marcora,
2007) and 22% of power produced is supra-maximal
(Stapelfeldt et al., 2004). Taken simplistically, these
data tend to suggest that success in the sport will be
related to a high sustainable power output and may be
closely related to the LT.
The LT is well accepted as an important endurance
performance indicator (Gavin et al., 2012; Allen &
Coggan, 2006; Sjödin & Karlsson, 1981; Coyle et al.,
1995; Bassett & Hawley, 2000; McNaughton et al.,
2006). Good correlations between the LT and road
cycling performance (Coyle et al., 1995; Bishop et al.,
1998) as well as the LT and XCO-MTB performance
have been expressed (Costa et al., 2008; Impellizzeri et
al., 2005). With the increasing availability of on-the-
bike personal power measuring devices, cycling
coaches developed functional threshold power (FTP) as
a field test to estimate the LT (Allen & Coggan, 2006).
FTP can easily be measured during a twenty-minute
maximal-power time-trial and tracked throughout a
training program with training intensity zones and
athlete ability subsequently estimated (Allen &
Coggan, 2006). Despite FTP being historically not
well understood, it has recently been shown to be
equivalent to the onset of blood lactate concentration of
4.0 mMol∙L-1 (Gavin et al., 2012). No study was found
that compared the FTP field test to any cycling
performance even though some cycling publications
recommend testing it regularly to track fitness
(Carmichael & Rutberg, 2009).
However well the LT has been shown to correlate with
cycling, previous literature has suggested that
intermittent performances are not best predicted from
continuous tests (Morton & Billat, 2003). With relation
to the variables of interest in this study, the LT as
Miller (2014). Validity of using functional threshold power and intermittent power to predict cross-country mountain bike race outcome.
Journal of Science and Cycling, 3(1): 16-20
Page 17
indicated by the FTP test can be considered a
continuous test and XCO-MTB deemed an intermittent
Interestingly, there is limited research comparing the
relationship between intermittent laboratory tests and
XCO-MTB performance (Inoue et al., 2012; Prins et
al., 2007), and there were no intermittent field tests
located that explore these variables. Due to the high
intensity and intermittent nature of XCO-MTB and the
given physiologic parameters surrounding current field
test standards, it seems sensible to tailor a field test
specific to the demands of XCO-MTB for use in the
prediction of performance and determination of athlete
ability. Notably, Prins et al. (2007) made a call for the
design of a XCO-MTB-specific test. In such a test,
cyclists would perform a discontinuous effort during
which the majority of the time is spent above the LT
and recovery periods are well below the LT. The
purpose of this study therefore is to investigate the
validity of using continuous (i.e. FTP) and intermittent
power (IP) field tests to predict XCO-MTB race
Materials and methods
Participants and procedures
Eleven regionally competitive male XCO-MTB cyclists
(mean age: 35.8 ± 8.2 yr; mean mass: 80.8 ± 13.4 kg)
volunteered to participate in this study. All participants
reported to be healthy and free of injury and had been
cycling regularly (≥ 4 sessions per week). Ethical
approval was granted after review from the Institutional
Review Board for the Protection of Human Subjects of
East Stroudsburg University. All participants
completed a written consent form and received
notification of potential risks and benefits of the study.
Participants completed three separate testing sessions in
random order and were asked to arrive to testing three-
hours post-absorptive and with no heavy exercise in the
previous 24 hours. During one session the participants
performed a FTP test following the procedure
explained by Allen and Coggan (2006) with a modified
warm-up. The second session was an intermittent test
(IP) with 20 intervals of 45 seconds work and 15
seconds recovery. The final testing session was a
mass-start mountain bike race. Subjects completed all
sessions within a maximum of 14 days and no less than
72 hours between tests.
Data collection
Prior to testing, body mass was recorded for each
subject. All participants used their own bikes for all
testing sessions. Each bike was fitted with a CycleOps
PowerTap rear wheel (G3 or Pro XCO-MTB Disc) and
mounted on a stationary trainer (CycleOps
Supermagneto Pro) for both FTP and IP testing. Data
was collected onto a mobile recording unit (CycleOps
Joule 2.0) and analyzed using PowerAgent software.
Race time was recorded using a stopwatch to the
nearest second. Each testing session began with a
shortened warm-up protocol derived from that
explained by Allen & Coggan (2006) and included ten
minutes of easy pedaling followed by five minutes of a
steady self-determined ‘hard’ effort and culminating
with ten minutes of easy pedaling.
Two elite cyclists were selected to gather anecdotal
pilot data from actual international and regional race
performances. All data were gathered on the
participants’ own XCO-MTB using a PowerTap Pro
XCO-MTB hub and recorded on a Joule 2.0. Data
analysis was done in PowerAgent software. During
these races it was determined that approximately 25%
of race time was spent either coasting or in recovery
power zone based on zones relative to FTP and
outlined by PowerAgent software. Throughout the
race, many efforts were completed above FTP and
lasted approximately less than one minute before
coasting or easy pedalling of varying duration. It is
from this data that work:rest ratios were determined for
Table 1. Values for relative FTP, IP, and race time. Values are means ± standard deviations.
Relative FTP (W/kg)
Race Time (s)
3.32 ± 0.74
4153 ± 561
Note: Relative FTP = Functional threshold Power; Relative IP = Intermittent Power
* Both models p < 0.001
Table 2. Linear regression models for predicting race time from Relative FTP and Relative IP.
Relative FTP
Race time = 6317.224 + (-655.688 Relative FTP)*
Relative IP
Race Time = 6662.768 + (-598.752 Relative IP)*
Note: Relative FTP = Relative Functional Threshold Power (W/kg); Relative IP = Relative Intermittent Power (W/kg); MSE = mean square error, calculated
as the residual sum of squares divided by the degrees of freedom; Error = estimation error, calculated as the square root of MSE.
* Both models p < 0.001
Table 3. Multiple regression models for predicting Race time from Relative IP and a combination of Relative IP and Relative FTP.
Relative IP
Race Time = 6662.768 + (-598.752 Relative IP)*
Race Time = 6662.768 + (-537.759 Relative IP) + (-71.950 Relative FTP)*
Note: Relative IP = Relative Intermittent Power (W/kg); Combined = regression model combining Relative FTP and Relative IP; MSE = mean square
error, calculated as the residual sum of squares divided by the degrees of freedom; Error = estimation error, calculated as the square root of MSE.
* Both models p < 0.001
J Sci Cycling. Vol. 3(1), 16-20
Miller et al.
Page 18
IP testing. Duration of total time for IP testing was set
at 20 minutes to have an equal duration as FTP testing.
From their own training, all participants were familiar
with the FTP test that required a sustained maximal
effort for 20 minutes. Athletes were instructed to
perform at their highest sustainable power for the
duration of the FTP test. FTP was recorded as 95% of
mean power across the duration of the test. The IP
protocol consisted of 20 intervals of 45 seconds work
and 15 seconds rest. Participants were told to
‘visualize covering the most distance possible’ during
each work interval. Participants could sit or stand
whenever necessary and maintain any cadence or
gearing throughout testing. The beginning and end of
each work bout was indicated by telling the subject to
either start or stop and measured by pressing the
‘INTERVAL’ button on the Joule 2.0 head unit.
Feedback related to elapsed time and power output
were available to the participants at any time during
testing just as they would be during self-intended field
testing. During recovery intervals in IP testing,
participants were instructed to pedal at an easy rate or
coast; these intervals were not recorded.
The race was conducted across nine laps consisting of
approximately 40% grass fields and 60% moderately
difficult singletrack trails on rolling terrain with 43 m
of elevation change each. Total distance covered for all
participants was 17.4 km.
Calculation of variables
Mean power was recorded during all work bouts. FTP
was calculated as 95% of mean 20-minute power and
was recorded relative to body mass. The mean of all
work bouts during IP was recorded relatively. Race
time was recorded to indicate performance and
measured in seconds after the completion of all laps.
Linear regression models (SPSS 19.0) were used to
evaluate the coefficient of determination and standard
error when using relative measures (W/kg) of FTP and
IP in prediction of XCO-MTB race performance. Mean
square error (MSE) was calculated as the residual sum
of squares divided by the degrees of freedom and used
to determine estimation error. Estimation error (Error)
was calculated as the square root of MSE and
expressed in unit time (s).
The data for the relative FTP, relative IP and the race
time are shown in Table 1. The correlation coefficient
between the values achieved in the Relative FTP and
those in the Relative IP was 0.964 (r2 = 0.929).
Table 2 shows the linear regression models created
using relative FTP and relative IP to predict race time.
Both models were able to significantly predict race
performance (p < 0.001). However, the prediction
errors were less when using relative IP than relative
FTP (273.5 s versus 303.6 s).
Multiple regression models were developed to assess
the prediction of race performance from Relative FTP
and Relative IP combined. The effect of combining
relative FTP and relative IP (r2=.887) did not enhance
the variance in XCO-MTB performance explained by
the relative IP (r2=.886) model substantially.
Furthermore, the error associated with the combined
model was in excess of that associated with the model
containing only relative IP (289.6 s and 273.5 s,
To the best of our knowledge, this is the first study to
use field-based tests to predict XCO-MTB
performance. Field tests are particularly practical to
athletes and coaches who do not have access to
laboratory equipment. The development of power-
based field tests that are highly correlated with actual
performance could help those with access only to a
portable power meter gain valuable insight into
potential for performance. Considering this study
presents strong correlations between XCO-MTB
performance and tests completed on a stationary
trainer, the repeatability of these tests stands
independent of weather conditions, expensive
equipment and laboratory practitioners.
Indeed, the primary purpose of this study was to
determine the validity of using FTP and IP field tests to
predict XCO-MTB race performance. Based on the
intermittent nature of XCO-MTB, it was questioned
whether a steady-state physiologic indicator was the
best option for assessing athletes of the sport. The FTP
test was chosen for a condition of comparison based on
the previous research relating the LT and XCO-MTB
performance (Costa et al., 2008; Impellizzeri et al.,
2005). Given the findings of Gavin et al. (2012)
indicating FTP to be equal to a commonly used
laboratory LT indicating the onset of blood lactate (4
mmol∙L-1), and the strong face validity of the FTP field
test among cyclists in general, this relationship
suggested FTP was suitable for means of use in this
Our pilot data and revealed that XCO-MTB cyclists
spend approximately 25% of race time either coasting
or pedaling in a recovery zone during a race. This is
relatable to the finding of Stapelfeldt et al. (2004), who
determined 39% of XCO-MTB race power to be less
than the aerobic threshold. With this information, the
IP test protocol for this study entailed work: rest ratios
set at 45 s: 15 s and was designed to be completed in
the same amount of time as the FTP field test (20
minutes). This ensured the IP field test was effective
means of blending the demands of XCO-MTB racing
found during pilot testing with the time constraint of
the FTP field test.
After a thorough search of published peer-reviewed
literature on cycling, this is the first intermittent field
test found that was designed to assess XCO-MTB. One
of our first findings was that IP is strongly correlated
with FTP. This can be explained by previous work
showing that the ability to complete intermittent
exercise to be reliant on aerobic metabolism and
oxygen uptake (Bogdanis et al., 1996; Gaitanos et al.,
1993; Bishop, 2012, Bishop et al., 2004), and that the
LT is strongly correlated to the percent of Type I
Miller (2014). Validity of using functional threshold power and intermittent power to predict cross-country mountain bike race outcome.
Journal of Science and Cycling, 3(1): 16-20
Page 19
muscle fibers (Coyle et al., 1992). This suggests that
IP performance is at least in part based on the
parameters surrounding an athlete’s FTP. Moreover,
interval training can improve aerobic-dependent time-
trial performance (Stepto et al., 1999; Lindsay et al.,
1996; Padilla et al., 1999). While the training of the
participants in this study was not recorded, it can be
postulated that the strong relationship between these
two tests is at least in part due to the combination of
training benefits of off-road cycling and the upper limit
of intermittent exercise capacity as constrained by
aerobic efficiency.
The main finding of this study is that when using
relative power output to predict XCO-MTB, IP has a
stronger correlation than FTP. This agrees with the
finding of Inoue et al. (2012) where a 5 x 30-s Wingate
(at 50% Wingate load) could predict XCO-MTB
performance. Interestingly, when Prins et al., (2007)
had athletes perform a time trial on the same course as
race performances were recorded, a similar correlation
was found as that in this study when comparing IP with
race time. In the same study, variable intensity and 1-
km time trials were performed and compared with the
same XCO-MTB performance. While these trials were
completed with a high degree of control in the
laboratory, the results shown here suggest IP field test
can better predict race performance. This supports the
IP field test as it suggests that performance can be
predicted at least as well as any other previously used
model and independent of weather conditions and
expensive laboratory equipment.
The error associated with the regression models used in
this study indicates that IP can predict XCO-MTB
performance with less error than FTP. This is
important given that finish rankings were often
determined by smaller time margins than the error
associated with both tests, and could mean the
difference between finishing first and third or getting
pulled from UCI races where the 80% rule is in effect.
Two potential criticisms of the IP test from a laboratory
practitioner could be lack of control for the workload
and the selection of the work:rest ratios. However, we
feel that given the relationship of IP and race
performance, the ease of execution of the IP field test,
and the suggestion that the IP field test is at least as
good at predicting XCO-MTB performance as other,
more difficult field and laboratory measures, future
research can potentially point towards fine-tuning a
similar test to the one used in this study. This area
needs more research.
Practical applications
This is the first study to relate power-based field tests
of FTP and IP to XCO-MTB performance. While
the FTP test can explain much of the variation in
performance, a field-based IP test is a better
predictor of XCO-MTB race time than FTP. The
astute coach should use this test as a criterion when
determining the ability and preparedness of an
athlete for competition.
The authors would like to thank Saris Cycling Group,
USA for supplying PowerTap power meters, Joule 2.0
head units and Supermagneto Pro stationary trainers for
all test sessions.
1. Allen, H. & Coggan, A. (2006). Training and racing
with a power meter. Boulder, CO: VeloPress.
2. Amann M, Subudhi A, Foster C. (2006). Predictive
validity of ventilatory and lactate thresholds for cycling
time trial performance. Scandinavian Journal of
Medicine & Science In Sports. February 2006;16(1):27-
3. Bishop, D. J. (2012). Fatigue during intermittent-sprint
exercise. Clin Exp Pharmacol Physiol, 39(9), 836-841.
4. Bishop, D., Edge, J., & Goodman, C. (2004). Muscle
buffer capacity and aerobic fitness are associated with
repeated-sprint ability in women. European Journal of
Applied Physiology, 92(4-5), 540-547. doi:
5. Bishop, D., Jenkins, D. G., & Mackinnon, L. T. (1998).
The relationship between plasma lactate parameters,
Wpeak and 1-h cycling performance in women.
Medicine and Science in Sports and Exercise, 30(8),
6. Bogdanis, G. C., Nevill, M. E., Boobis, L. H., &
Lakomy, H. K. (1996). Contribution of phosphocreatine
and aerobic metabolism to energy supply during
repeated sprint exercise. Journal of Applied Physiology
(1985), 80(3), 876-884.
7. Carmichael, C and Rutberg, J. The Time-Crunched
Cyclist: Fit, Fast, and Powerful in 6 Hours a Week.
Boulder, CO: VeloPress, 2009.
8. Ciric, IM., Stojiljkovic, S., Stefanovic, N., Djurdjevic,
S., Bjelakovic, L., Pirsl, D. (2012). Anaerobic threshold
determination by direct blood lactate measurement with
and without warm up protocol in female athletes.
HealthMed, Vol. 6 Issue 6, p2152-2157.
9. Costa, V., & Fernando, D.O. (2008). Physiological
variables to predict performance in cross-country
mountain bike races. Journal of Exercise Physiology
Online, 11(6).
10. Coyle, E. F., Coggan, A. R., Hopper, M. K., & Walters,
T. J. (1988). Determinants of endurance in well-trained
cyclists. Journal of Applied Physiology, 64(6), 2622-
11. Coyle, E. F., Sidossis, L. S., Horowitz, J. F., & Beltz, J.
D. (1992). Cycling efficiency is related to the percentage
of Type I muscle fibers. Medicine and Science in Sports
and Exercise, 24(7), 782-788.
12. Davis, J. A., Rozenek, R., DeCicco, D. M., Carizzi, M.
T., & Pham, P. H. (2007). Comparison of three methods
for detection of the lactate threshold. Clinical
Physiology & Functional Imaging, 27(6), 381-384.
13. Faude, O., Kindermann, W., Meyer, T. (2009). Lactate
threshold concepts: how valid are they? Sports
Medicine , Vol. 39, No. 6., pp. 469-490.
14. Gaitanos, G. C., Williams, C., Boobis, L. H., & Brooks,
S. (1993). Human muscle metabolism during
intermittent maximal exercise. Journal of Applied
Physiology (1985), 75(2), 712-719.
15. Gavin, T. P., Van Meter, J. B., Brophy, P. M., Dubis, G.
S., Potts, K. N., & Hickner, R. C. (2012). Comparison of
a field-based test to estimate functional threshold power
and power output at lactate threshold. Journal Of
Strength & Conditioning Research (Lippincott Williams
& Wilkins), 26(2), 416-421.
J Sci Cycling. Vol. 3(1), 16-20
Miller et al.
Page 20
16. Gregory, J., Johns, D., Walls, J. Relative vs. absolute
physiological measures as predictors of mountain bike
cross-country race performance. (2007). Journal Of
Strength & Conditioning Research (Allen Press
Publishing Services Inc.). February 2007;21(1):17-22.
17. Impellizzeri, F. M., Rampinini, E., Sassi, A., Mognoni,
P., & Marcora, S. (2005). Physiological correlates to off-
road cycling performance. Journal of Sports Sciences,
23(1), 41-47.
18. Impellizzeri, F.M., and Marcora, S.M. (2007).
Physiology of mountain biking. Sports Medicine. 2007,
19. Impellizzeri, F., Sassi, A., Rodriguez-Alonso, M.,
Mognoni, P., & Marcora, S. (2002). Exercise intensity
during off-road cycling competitions. Medicine &
Science in Sports & Exercise, 34(11), 1808-1813.
20. Inoue, A., SaFilho, A.S., Mello, F.C.M., Santos, T.M.
(2012). Relationship Between Anaerobic Cycling Tests
and Mountain Bike Cross-Country Performance. Journal
of Strength and Conditioning Research, 26(6)/1589-
21. Lee, H., Martin, D., Anson, J., Grundy, D., Hahn, A.
(2012). Physiological characteristics of successful
mountain bikers and professional road cyclists. Journal
of Sports Sciences. December 2002;20(12):1001-1008.
22. Lindsay, F. H., Hawley, J. A., Myburgh, K. H.,
Schomer, H. H., Noakes, T. D., & Dennis, S. C. (1996).
Improved athletic performance in highly trained cyclists
after interval training. Medicine & Science in Sports &
Exercise, 28(11), 1427-1434.
23. Macdermid, P., Stannard, S. (2012). Mechanical work
and physiological responses to simulated cross country
mountain bike racing. Journal Of Sports Sciences.
October 2012;30(14):1491-1501.
24. McNaughton, L.R., Roberts, S., Bentley, D.J. (2006).
The relationship among peak power output, lactate
threshold, and short-distance cycling performance:
effects of incremental exercise test design. Journal of
Strength & Conditioning Research, Vol. 20 Issue 1,
p157-161. 5p.
25. Morton, R. H., & Billat, L. V. (2004). The critical power
model for intermittent exercise. European Journal of
Applied Physiology, 91(2-3), 303-307.
26. Padilla, S., Mujika, I., Orbananos, J., & Angulo, F.
(2000). Exercise intensity during competition time trials
in professional road cycling. Medicine & Science in
Sports & Exercise, 32(4), 850-856.
27. Prins, L., Terblanche, E., & Myburgh, K. H. (2007).
Field and laboratory correlates of performance in
competitive cross-country mountain bikers. Journal of
Sports Sciences, 25(8), 927-935.
28. Stapelfeldt, B., Schwirtz, A., Schumacher, Y.O., &
Hillebrecht, M. (2004). Workload demands in mountain
bike racing. International Journal of Sports Medicine,
25(4), 294300.
29. Stepto, N. K., Hawley, J. A., Dennis, S. C., & Hopkins,
W. G. (1999). Effects of different interval-training
programs on cycling time-trial performance. Medicine &
Science in Sports & Exercise, 31(5), 736-741.
30. Sjodin, B., & Jacobs, I. (1981). Onset of blood lactate
accumulation and marathon running performance.
International Journal of Sports Medicine, 2(1), 23-26.
  • Article
    Full-text available
    AbstractAdvances in technology have made the use of a variety of power meters ubiquitous in road cycling along with an ever-increasing popularity during mountain biking. This study compared data from one bicycle using three power meters: Stages (non-driveside crank arm); Quarq (chainring spider); and Powertap (rear-wheel hub). While no differences (p?>?.05) between power meters were present during treadmill riding at high or low cadences, dissimilarities for both power (W) and cadence (rpm) were apparent during actual cross-country mountain bike riding. Frequency distribution and analysis of coasting indicate that the Stages records more time (p?<?.001) at zero watts (6.9?±?3.3 s) and zero cadence (6.9?±?3.3 s) compared with Quarq (W?=?3.3?±?1.5 s, rpm?=?.8?±?.7 s) and Powertap (W?=?1.1?±?.8 s, rpm?=?3.0?±?1.2 s). Consequently, significant interactions (power meter???terrain, p?=?.0351) and main effects (power meter p?<?.0001, and terrain p?<?.0001) for power output were present and include
  • Article
    Full-text available
    Real-world cycling performance depends not only on exercise capacities, but also on efficiently traversing the bicycle through the terrain. The aim of this study was to determine if it was possible to quantify the braking done by a cyclist in the field. One cyclist performed 408 braking trials (348 on a flat road; 60 on a flat dirt path) over 5 days on a bicycle fitted with brake torque and angular velocity sensors to measure brake power. Based on Newtonian physics, the sum of brake work, aerodynamic drag and rolling resistance was compared with the change in kinetic energy in each braking event. Strong linear relationships between the total energy removed from the bicycle-rider system through braking and the change in kinetic energy were observed on the tar-sealed road (r² = 0.989; p < 0.0001) and the dirt path (r² = 0.952; p < 0.0001). T-tests revealed no difference between the total energy removed and the change in kinetic energy on the road (p = 0.715) or dirt (p = 0.128). This study highlights that brake torque and angular velocity sensors are valid for calculating brake power on the disc brakes of a bicycle in field conditions. Such a device may be useful for investigating cyclists’ ability to traverse through various terrains.
  • Article
    Functional threshold power is defined as the highest power output a cyclist can maintain in a quasi-steady state for approximately 60 min (FTP60). In order to improve practicality for regular evaluations, FTP60 could theoretically be determined as 95% of the mean power output in a 20-min time trial (FTP20). This study tested this assumption and the validity of FTP20 and FTP60 against the individual anaerobic threshold (IAT). Twenty-three trained male cyclists performed an incremental test to exhaustion, 20- and 60-min time trials, and a time to exhaustion at FTP20. Power output, heart rate and oxygen uptake representing FTP20, FTP60 and IAT were not different (p>0.05), and large to very large correlations were found (r=0.61 to 0.88). Bland-Altman plots between FTP20, FTP60 and IAT showed small bias (-1 to -5 W), but large limits of agreement ([-40 to 32 W] to [-62 to 60 W]). Time to exhaustion at FTP20 was 50.9±15.7 min. In conclusion, FTP20 and FTP60 should not be used interchangeably on an individual basis and their validity against IAT should be interpreted with caution.
  • Article
    Full-text available
    This study aimed to cross validate previously developed predictive models of mountain biking performance in a new cohort of mountain bikers during a 4-h event (XC4H). Eight amateur XC4H cyclists completed a multidimensional assessment battery including a power profile assessment that consisted of maximal efforts between 6 and 600 s, maximal hand grip strength assessments, a video-based decision-making test as well as a XC4H race. A multiple linear regression model was found to predict XC4H performance with good accuracy (R² = 0.99; P < 0.01). This model consisted of relative to total cycling mass (body mass including competition clothing and bicycle mass), maximum power output sustained over 60 s relative to total cycling mass, peak left hand grip strength and two-line decision-making score. Previous models for Olympic distance MTB performance demonstrated merit (R² = 0.93; P > 0.05) although subtle changes improved the fit, significance and normal distribution of residuals within the model (R² = 0.99; P < 0.01), highlighting differences between the disciplines. The high level of predictive accuracy of the new XC4H model further supports the use of a multidimensional approach in predicting MTB performance. The difference between the new, XC4H and previous Olympic MTB predictive models demonstrates subtle differences in physiological requirements and performance predictors between the two MTB disciplines.
  • Article
    Full-text available
    Using a brake power meter, experienced and inexperienced mountain bikers were tested on an isolated, controlled off-road cycling descent with a turn to determine how riding experience affects the pattern of braking behaviour. Overall braking measurements such as absolute and relative brake work and brake power, as well as brake time, were significantly related to performance time on the track used in this study. Inexperienced mountain bikers displayed greater absolute and relative brake work and brake time, but had lower absolute and relative brake power when compared with experienced mountain bikers, which resulted in a significant performance decrement for inexperienced riders. Experienced mountain bikers concentrated braking efforts to later in the track, which meant that they spent less time at lower speeds. Inexperienced riders displayed a greater reliance on the rear brake, which likely contributed to their overall increased braking variables. The results of this study highlight that differences in braking magnitude and behaviour are attributable to reduced performance on an isolated off-road track with a corner. Inexperienced mountain bike riders may be able to improve their performance by learning braking patterns similar to those of experienced mountain bikers.
  • Article
    Full-text available
    Purpose: This study aimed to analyze the relationship between the Functional Threshold Power (FTP) and the Lactate Threshold (LT). Methods: 20 male cyclists performed an incremental test in which the LT was determined. At least 48 h later, they performed a 20-minute time trial and 95% of the mean power output (P20) was defined as FTP. Participants were divided into recreational (Peak Power Output [PPO] < 4.5 W∙kg-1, n=11) or trained cyclists (PPO > 4.5 W∙kg-1, n=9) according to their fitness status. Results: The FTP (240 ± 35 W) was overall not significantly different (effect size[ES]=0.20, limits of agreement [LoA]=-2.4 ± 11.5%) from the LT (246 ± 24 W), and both markers were strongly correlated (r=0.95, p<0.0001). Accounting for the participants’ fitness status, no significant differences were found between FTP and LT ([ES]=0.22; LoA=2.1 ± 7.8%) in TC, but FTP was significantly lower than the LT (p=0.0004, ES=0.81; LoA=-6.5 ± 8.3%) in RC. A significant relationship was found between relative PPO and the bias between FTP and the LT markers (r=0.77, p<0.0001). Conclusions: The FTP is a valid field test-based marker for the assessment of endurance fitness. However, caution should be taken when using the FTP interchangeably with the LT as the bias between markers seems to depend on the athletes’ fitness status. Whereas the FTP provides a good estimate of the LT in trained cyclists, in recreational cyclists FTP may underestimate LT.
  • Article
    Full-text available
    There is currently a dearth of information describing cycling performance outside of propulsive and physiological variables. The aim of the present study was to utilise a brake power meter to quantify braking during a multi-lap cross-country mountain bike time trial and to determine how braking affects performance. A significant negative association was determined between lap time and brake power (800.8 ± 216.4 W, mean ± SD; r = −0.446; p < 0.05), while the time spent braking (28.0 ± 6.4 s) was positively associated with lap time (314.3 ± 37.9 s; r = 0.477; p < 0.05). Despite propulsive power decreasing after the first lap (p < 0.05), lap time remained unchanged (p > 0.05) which was attributed to decreased brake work (p < 0.05) and brake time (p < 0.05) in both the front and rear brakes by the final lap. A multiple regression model incorporating braking and propulsion was able to explain more of the variance in lap time (r² = 0.935) than propulsion alone (r² = 0.826). The present study highlights that riders’ braking contributes to mountain bike performance. As riders repeat a cross-country mountain bike track, they are able to change braking, which in turn can counterbalance a reduction in power output. Further research is required to understand braking better.
  • Article
    Full-text available
    Sports performance testing is one of the most common and important measures used in sport science. Performance testing protocols must have high reliability to ensure any changes are not due to measurement error or inter-individual differences. High validity is also important to ensure test performance reflects true performance. Time-trial protocols commonly have a coefficient of variation (CV) of <5%, however, familiarization, well-trained subjects and/or conducting the trial outdoors in the athlete’s most familiar environment can lead to CVs of < 1%. Long duration time-trials or the inclusion of sprints within a time-trial appears to not negatively influence reliability. Few studies have assessed the validity of endurance performance tests, and as such more research should evaluate different ways of simulating outdoor performances in the laboratory. The use of warm-up, simulation of convection load, and implementation of race specific hydration practices are important considerations for researchers regarding test validity.
  • Article
    Full-text available
    Objectives Traditional performance tests in mountain bike marathon (XCM) primarily quantify aerobic metabolism and may not describe the relevant capacities in XCM. We aimed to validate a comprehensive test protocol quantifying its intermittent demands. Methods Forty-nine athletes (38.8±9.1 years; 38 male; 11 female) performed a laboratory performance test, including an incremental test, to determine individual anaerobic threshold (IAT), peak power output (PPO) and three maximal efforts (10 s all-out sprint, 1 min maximal effort and 5 min maximal effort). Within 2 weeks, the athletes participated in one of three XCM races (n=15, n=9 and n=25). Correlations between test variables and race times were calculated separately. In addition, multiple regression models of the predictive value of laboratory outcomes were calculated for race 3 and across all races (z-transformed data). Results All variables were correlated with race times 1, 2 and 3: 10 s all-out sprint (r=−0.72; r=−0.59; r=−0.61), 1 min maximal effort (r=−0.85; r=−0.84; r=−0.82), 5 min maximal effort (r=−0.57; r=−0.85; r=−0.76), PPO (r=−0.77; r=−0.73; r=−0.76) and IAT (r=−0.71; r=−0.67; r=−0.68). The best-fitting multiple regression models for race 3 (r²=0.868) and across all races (r²=0.757) comprised 1 min maximal effort, IAT and body weight. Conclusion Aerobic and intermittent variables correlated least strongly with race times. Their use in a multiple regression model confirmed additional explanatory power to predict XCM performance. These findings underline the usefulness of the comprehensive incremental test to predict performance in that sport more precisely.
  • Article
    Full-text available
    This research aimed at pointing out the determination of the anaerobic threshold by the direct blood lactate measurement with and without prior administration of the warm up protocol in female athletes. Resesearch sample was drawn from 50 female athletes subjects divided into four groups: 15 nontrained subjects, prior subjected to the warm up protocol, 15 well- trained subjects prior subjected to the warm up protocol, 10 non-trained subjects, who were not prior subjected to the warm up protocol, 10 well- trained subjects who were not prior subjected to the warm up protocol. Results obtained and presented in this paper show that AT, determined by the direct blood lactate measurement, is statistically significant (p <0,001) in higher values of the heart rate in well- trained subjects compared to the non-trained subjects, equally for those subjected and not subjected to the warm up protocol before the workload test.
  • Article
    Full-text available
    Costa VP, De-Oliveira FR. Physiological variables to predict performance in crosscountry mountain bike races. JEPonline 2008;11(6):14-24. Mountain bike (MTB) is a recent sport derived from cycling with little information about the athletes and races. The aim of this study was to identify the morph-physiological characteristics in Elite MTB athletes and the physiological variables associated in performance during Crosscountry Olympic (XCO) races. Six Elite mountain bikers (26.5 ± .6 years; 69.1 ± 2.1 kg; 174.0 ± 1.2 cm; 5.9±0.9 % body fat estimated; 9.0 ± 1.3 years of training) were included in this study. The participants were submitted to the Wingate test and an incremental progressive exercise. Then they were evaluated during the XCO World Cup and XCO Brazilian National Championship. The results indicate that riders presented similar morphologic characteristics to the international athletes. However, the sub-maximal and maximal power outputs are lower. The maximal power output (W max) relative to body mass was significantly associated with performance in two races. The power at second lactate threshold (WLL 2) was only significantly correlated in XCO World Cup when normalized to exponent of mass 0.79. Therefore, the results of this study provide the support to the use of the W max and WLL 2 in the physiological assessments of mountain bikers. Furthermore, the body size should be taken into account to evaluate off road cyclists.
  • Article
    Full-text available
    Abstract The purpose was to assess the mechanical work and physiological responses to cross country mountain bike racing. Participants (n = 7) cycled on a cross country track at race speed whilst [Formula: see text]O(2), power, cadence, speed, and geographical position were recorded. Mean power during the designated start section (68.5 ± 5.5 s) was 481 ± 122 W, incurring an O(2) deficit of 1.58 ± 0.67 L - min(-1) highlighting a significant initial anaerobic (32.4 ± 10.2%) contribution. Complete lap data produced mean (243 ± 12 W) and normalised (279 ± 15 W) power outputs with 13.3 ± 6.1 and 20.7 ± 8.3% of time spent in high force-high velocity and high force-low velocity, respectively. This equated to, physiological measures for %[Formula: see text]O(2max) (77 ± 5%) and % HR(max) (93 ± 2%). Terrain (uphill vs downhill) significantly (P < 0.05) influenced power output (70.9 ± 7.5 vs 41.0 ± 9.2% W(max)),the distribution of low velocity force production, [Formula: see text]O(2) (80 ± 1.7 vs 72 ± 3.7%) and cadence (76 + 2 vs 55 ± 4 rpm) but not heart rate (93.8 ± 2.3 vs 91.3 ± 0.6% HR(max)) and led to a significant difference between anaerobic contribution and terrain (uphill, 6.4 ± 3.0 vs downhill, 3.2 ± 1.8%, respectively) but not aerobic energy contribution. Both power and cadence were highly variable through all sections resulting in one power surge every 32 s and a supra-maximal effort every 106 s. The results show that cross country mountain bike racing consists of predominantly low velocity pedalling with a large high force component and when combined with a high oscillating work rate, necessitates high aerobic energy provision, with intermittent anaerobic contribution. Additional physical stress during downhill sections affords less recovery emphasised by physiological variables remaining high throughout.
  • 1. There is a reversible decline in force production by muscles when they are contracting at or near their maximum capacity. The task-dependent nature of fatigue means that the mechanisms of fatigue may differ between different types of contractions. This paper examines how fatigue manifests during whole-body, intermittent-sprint exercise and discusses the potential muscular and neural mechanisms that underpin this fatigue. 2. Fatigue is defined as a reversible, exercise-induced reduction in maximal power output (e.g. during cycling exercise) or speed (e.g. during running exercise), even though the task can be continued. 3. The small changes in surface electromyogram (EMG), along with a lack of change in voluntary muscle activation (estimated from both percutaneous motor nerve stimulations and trans-cranial magnetic stimulation), indicate that there is little change in neural drive to the muscles following intermittent-sprint exercise. This, along with the observation that the decrease in EMG is much less than that which would be predicted from the decrease in power output, suggests that peripheral mechanisms are the predominant cause of fatigue during intermittent-sprint exercise. 4. At the muscle level, limitations in energy supply, including phosphocreatine hydrolysis and the degree of reliance on anaerobic glycolysis and oxidative metabolism, and the intramuscular accumulation of metabolic by-products, such as hydrogen ions, emerge as key factors responsible for fatigue.
  • Article
    It has been proposed that field-based tests (FT) used to estimate functional threshold power (FTP) result in power output (PO) equivalent to PO at lactate threshold (LT). However, anecdotal evidence from regional cycling teams tested for LT in our laboratory suggested that PO at LT underestimated FTP. It was hypothesized that estimated FTP is not equivalent to PO at LT. The LT and estimated FTP were measured in 7 trained male competitive cyclists (VO2max = 65.3 ± 1.6 ml O2·kg(-1)·min(-1)). The FTP was estimated from an 8-minute FT and compared with PO at LT using 2 methods; LT(Δ1), a 1 mmol·L(-1) or greater rise in blood lactate in response to an increase in workload and LT(4.0), blood lactate of 4.0 mmol·L(-1). The estimated FTP was equivalent to PO at LT(4.0) and greater than PO at LT(Δ1). VO2max explained 93% of the variance in individual PO during the 8-minute FT. When the 8-minute FT PO was expressed relative to maximal PO from the VO2max test (individual exercise performance), VO2max explained 64% of the variance in individual exercise performance. The PO at LT was not related to 8-minute FT PO. In conclusion, FTP estimated from an 8-minute FT is equivalent to PO at LT if LT(4.0) is used but is not equivalent for all methods of LT determination including LT(Δ1).
  • Article
    Despite its apparent relevance, there is no evidence supporting the importance of anaerobic metabolism in Olympic crosscountry mountain biking (XCO). The purpose of this study was to examine the correlation between XCO race time and performance indicators of anaerobic power. Ten XCO riders (age: 28 ± 5 years; weight: 68.7 ± 7.7 kg; height: 177.9 ± 7.4 cm; estimated body fat: 5.7 ± 2.8%; estimated ·VO₂max: 68.4 ± 5.7 ml·kg⁻¹·min⁻¹) participating in the Lagos Mountain Bike Championship (Brazil) completed 2 separate testing sessions before the race. In the first session, after anthropometric assessments were performed, the cyclists completed a single 30-second Wingate (WIN) test and an intermittent tests consisting of 5 × 30-second WIN tests (50% of the single WIN load) with 30 seconds of recovery between trials. In the second session, the riders performed a maximal incremental test. A significant correlation was found between race time and maximal power on the 5× WIN test (r = -0.79, IC(95%) -0.94 to -0.32, p = 0.006) and the mean average power on the 5× WIN test normalized by body mass (r = -0.63, IC(95%) -0.90 to -0.01, p = 0.048). The finding of the study supports the use of anaerobic tests for assessing mountain bikers participating in XCO competitions and suggests that anaerobic power is an important determinant of performance.
  • Article
    Full-text available
    During the last nearly 50 years, the blood lactate curve and lactate thresholds (LTs) have become important in the diagnosis of endurance performance. An intense and ongoing debate emerged, which was mainly based on terminology and/or the physiological background of LT concepts. The present review aims at evaluating LTs with regard to their validity in assessing endurance capacity. Additionally, LT concepts shall be integrated within the 'aerobic-anaerobic transition' - a framework which has often been used for performance diagnosis and intensity prescriptions in endurance sports. Usually, graded incremental exercise tests, eliciting an exponential rise in blood lactate concentrations (bLa), are used to arrive at lactate curves. A shift of such lactate curves indicates changes in endurance capacity. This very global approach, however, is hindered by several factors that may influence overall lactate levels. In addition, the exclusive use of the entire curve leads to some uncertainty as to the magnitude of endurance gains, which cannot be precisely estimated. This deficiency might be eliminated by the use of LTs. The aerobic-anaerobic transition may serve as a basis for individually assessing endurance performance as well as for prescribing intensities in endurance training. Additionally, several LT approaches may be integrated in this framework. This model consists of two typical breakpoints that are passed during incremental exercise: the intensity at which bLa begin to rise above baseline levels and the highest intensity at which lactate production and elimination are in equilibrium (maximal lactate steady state [MLSS]). Within this review, LTs are considered valid performance indicators when there are strong linear correlations with (simulated) endurance performance. In addition, a close relationship between LT and MLSS indicates validity regarding the prescription of training intensities. A total of 25 different LT concepts were located. All concepts were divided into three categories. Several authors use fixed bLa during incremental exercise to assess endurance performance (category 1). Other LT concepts aim at detecting the first rise in bLa above baseline levels (category 2). The third category consists of threshold concepts that aim at detecting either the MLSS or a rapid/distinct change in the inclination of the blood lactate curve (category 3). Thirty-two studies evaluated the relationship of LTs with performance in (partly simulated) endurance events. The overwhelming majority of those studies reported strong linear correlations, particularly for running events, suggesting a high percentage of common variance between LT and endurance performance. In addition, there is evidence that some LTs can estimate the MLSS. However, from a practical and statistical point of view it would be of interest to know the variability of individual differences between the respective threshold and the MLSS, which is rarely reported. Although there has been frequent and controversial debate on the LT phenomenon during the last three decades, many scientific studies have dealt with LT concepts, their value in assessing endurance performance or in prescribing exercise intensities in endurance training. The presented framework may help to clarify some aspects of the controversy and may give a rationale for performance diagnosis and training prescription in future research as well as in sports practice.
  • Article
    We determined that the variability in the oxygen cost and thus the caloric expenditure of cycling at a given work rate (i.e., cycling economy) observed among highly endurance-trained cyclists (N = 19; mean +/- SE; VO2max, 4.9 +/- 0.1 l.min-1; body weight, 71 +/- 1 kg) is related to differences in their % Type I muscle fibers. The percentage of Type I and II muscle fibers was determined from biopsies of the vastus lateralis muscle that were histochemically stained for ATPase activity. When cycling a Monark ergometer at 80 RPM at work rates eliciting 52 +/- 1, 61 +/- 1, and 71 +/- 1% VO2max, efficiency was determined from the caloric expenditure responses (VO2 and RER using open circuit spirometry) to steady-state exercise. Gross efficiency (GE) was calculated as the ratio of work accomplished.min-1 to caloric expenditure.min-1, whereas delta efficiency (DE) was calculated as the slope of this relationship between approximately 50 and 70% VO2max. The % Type I fibers ranged from 32 to 76%, and DE when cycling ranged from 18.3 to 25.6% in these subjects. The % Type I fibers was positively correlated with both DE (r = 0.85; P less than 0.001; N = 19) and GE (r = 0.75; P less than 0.001; N = 19) during cycling. Additionally, % Type I fibers was positively correlated with GE (r = 0.74; P less than 0.001; N = 13) measured during the novel task of two-legged knee extension; performed at a velocity of 177 +/- 6 degrees.s-1 and intensity of 50 and 70% of peak VO2 for that activity.(ABSTRACT TRUNCATED AT 250 WORDS)
  • Article
    Fourteen competitive cyclists who possessed a similar maximum O2 consumption (VO2 max; range, 4.6-5.0 l/min) were compared regarding blood lactate responses, glycogen usage, and endurance during submaximal exercise. Seven subjects reached their blood lactate threshold (LT) during exercise of a relatively low intensity (group L) (i.e., 65.8 +/- 1.7% VO2 max), whereas exercise of a relatively high intensity was required to elicit LT in the other seven men (group H) (i.e., 81.5 +/- 1.8% VO2 max; P less than 0.001). Time to fatigue during exercise at 88% of VO2 max was more than twofold longer in group H compared with group L (60.8 +/- 3.1 vs. 29.1 +/- 5.0 min; P less than 0.001). Over 92% of the variance in performance was related to the % VO2 max at LT and muscle capillary density. The vastus lateralis muscle of group L was stressed more than that of group H during submaximal cycling (i.e., 79% VO2 max), as reflected by more than a twofold greater (P less than 0.001) rate of glycogen utilization and blood lactate concentration. The quality of the vastus lateralis in groups H and L was similar regarding mitochondrial enzyme activity, whereas group H possessed a greater percentage of type I muscle fibers (66.7 +/- 5.2 vs. 46.9 +/- 3.8; P less than 0.01). The differing metabolic responses to submaximal exercise observed between the two groups appeared to be specific to the leg extension phase of cycling, since the blood lactate responses of the two groups were comparable during uphill running. These data indicate that endurance can vary greatly among individuals with an equal VO2 max.(ABSTRACT TRUNCATED AT 250 WORDS)