Secondary Logo

Journal Logo

Original Research

Laboratory-Based Physical and Physiological Test Results That Serve as Predictors of Male, Amateur Road Cyclists' Performance Levels

Coetzee, Ben; Malan, Dawie

Author Information
Journal of Strength and Conditioning Research: October 2018 - Volume 32 - Issue 10 - p 2897-2906
doi: 10.1519/JSC.0000000000002619
  • Free

Abstract

Introduction

Competitive road cycling continues to grow and is becoming increasingly popular worldwide despite the great physiological demands that are placed on cyclists during participation in different road cycling events (41). Typical professional road races for men take place over distances between 40 and 245 km and may last from 1 to 6.5 hours (41). In view of these distances and the duration of road races, it is conceivable why physical and physiological components are highlighted as important contributors to road cycling performance (28,38). However, despite excellent performances of African cyclists, such as Robert Hunter and Louis Meintjes (South Africa) in the Tour de France, and Natnael Berhane (Eritrea), who became the first black African rider to win a professional stage race at the 2014 Tropica Amissa Bongo (41), limited research has focused on the laboratory-based physical and physiological parameters that may contribute to the performance of African road cyclists. Furthermore, a literature review of Lucia et al. (28) suggests that previous research on professional cyclists might not be directly extrapolated to elite, amateur cyclists. More so, only a few studies have thus far investigated the use of laboratory-based physical and physiological tests to predict the performance levels of road cyclists (7,11,27,38,40,42).

Various laboratory-based physical and physiological test variables are significantly related to endurance cycling performance. For example, researchers (5,9,19,30) observed a significant, direct relationship between the maximum power output values (Wmax) of road cyclists and time-trial performances in the 10-km (r = −0.59), 14 km (r = −0.90), 20-km (r = −0.91), and 40-km distances (r = −0.87), respectively. Absolute (W) and relative power outputs (W·kg−1), oxygen uptake (V̇o2), and percentage of maximum relative oxygen uptake (%V̇o2max) corresponding to the second ventilatory threshold (VT2) or lactate threshold (LT) have also been found to correlate significantly with performances (time or power output) over different cycling time-trial distances (2–4,30). Davison et al. (13) found that average power per unit of body mass as determined by the Wingate anaerobic test (WAnT) also served as the best predictor of simulated 1 km (r = −0.92) and 6 km (r = −0.90) hill climbing performance among competitive club cyclists (13).

However, the execution of power-related activities such as accelerating and sprinting maximally during road cycling are dependent on the level of muscle flexibility in the exercising areas. For example, Wilson et al. (43) found that stiffness in the maximal series elastic component of the muscle was reduced with an increase in muscle flexibility and the potential to store and release elastic strain to facilitate concentric muscle contractions increased. The Biomechanical Model of Exercise Physiology and Athletic Performance as proposed by Noakes (37) supports this contention by stating that elastic elements in the leg muscles (titin and collagen fibers) can act like a spring which stores elastic energy during continuous motion movements such as cycling.

Laboratory-based physical and physiological test variables may also serve as discriminators between different performance levels of road cyclists. In this regard, various researchers (7,11,27) indicated that absolute and relative power outputs as well as percentage of V̇o2max corresponding to the first ventilatory threshold (VT1), respiratory compensation point (RCP), or LT served as discriminators between successful and less successful road cyclists. In addition, significantly higher WAnT mean power outputs were reported for more successful (United States Cycling Federation category II) compared with less successful road cyclists (United States Cycling Federation category IV) (40). Another measure of peak anaerobic power output, namely the vertical jump height (32), also served as a statistically discriminating factor between international and category 1–3 cyclists (42).

Despite the availability of research that provided proof for the use of laboratory-based physical and physiological tests to predict the “real-life” performances and performance levels of road cyclists, contradictory research does exist. For example, Coyle et al. (11) found that the standard error of estimate (SEE) for the prediction of actual road racing 40-km time-trial performance was 1 minute and 48 seconds when laboratory-based 1-hour performance power was used. Balmer et al. (6) found a weak correlation between peak power output obtained during a maximal aerobic power test on a Kingcycle ergometer and 16.1-km time-trial performance on the road (r = 0.46, p = 0.07, SEE = 1 minute and 9 seconds). Furthermore, Jobsen et al. (22) showed that only 69.3% of the variation in road time-trial speed could be explained by laboratory time-trial speed and that the addition of body mass increased the explained variance to 78.3%. Therefore, they concluded that a large proportion of the difference between laboratory and field time-trial performance remains to be explained.

Above-mentioned findings would suggest that laboratory-based protocols do not always accurately predict “real-life” performances. Moreover, no current publications have focused on the power of different laboratory-based physical and physiological test results to predict the “real-life” performance levels of amateur road cyclists from Africa. Therefore, the purposes of this study were first, to determine the practical significant differences in laboratory-based physical and physiological test results between a selected group of successful and less successful male, amateur road cyclists from Africa; and second, to determine the significance, adequacy, accurateness, and usefulness of laboratory-based physical and physiological test results to serve as predictors of these male, amateur road cyclists' performance levels. Practitioners in the field of sport and conditioning need clarity concerning the use and applicability of laboratory-based physical and physiological test results to serve as predictors of “real-life” road cycling performances. The identification of significant, adequate, accurate, and useful laboratory-based predictors will enable practitioners to focus on the most relevant physical and physiological components for performance monitoring and assessment in road cycling.

Methods

Experimental Approach to the Problem

The design of this study was a selected group, quasi-experimental research design. The specific hypothesis under scrutiny was that most of the laboratory-based physical and physiological test results will serve as significant, adequate, accurate, and useful predictors of male, amateur road cyclists' performance levels. The hypothesis was tested by subjecting cyclists to a testing protocol that consisted of flexibility tests, the abdominal stage test (AST), the vertical jump test (VJT), the 30-second WAnT, and a graded V̇o2max exercise test. A binary forward stepwise logistic regression was performed to screen for the predictive value of different laboratory-based physical and physiological test (independent) variables in predicting group classification of subjects (dependent variables). Cyclists were categorized into successful and less successful cyclists based on competition and training performances as well as coaches' input.

Subjects

Subjects included 45 (n = 45) well-trained, male, amateur road cyclists (mean ± SD: age 21.3 ± 3.1 years; body stature 176.3 ± 6.1 cm; body mass 65.5 ± 6.5 kg) from 13 African countries, namely: South Africa, Egypt, Kenya, Namibia, Algeria, Mauritius, Tunisia, Seychelles, Libya, Burundi, Ethiopia, Zimbabwe, and Gabon. Cyclists were identified by the various cycling federations as their top amateur road cyclists and were attending a 3-month–long training camp of The African Continental Cycling Center during the testing period. Cyclists were ranked according to their training time trial and competition performances that they achieved during the 3-month period. The ranking list was then verified by coaches of the cycling group. The top 14 cyclists on the ranking list were categorized as successful cyclists (mean ± SD: age 22.4 ± 2.9 years; body stature 177.5 ± 5.5 cm; body mass 64.7 ± 6.3 kg) due to their considerable better performances in competitions and during training time trials compared with the rest of the cyclists who were classified as less successful cyclists (n = 31; mean ± SD: age 20.9 ± 3.1 years; body stature 175.7 ± 6.3 cm; body mass 65.9 ± 6.6 kg). The successful cycling group consisted of cyclists who obtained places among the top 30 cyclists during the Category B-World Championships. They also significantly outperformed the other cyclists during 20- and 40-km training time trials over the 3-month training period that was monitored.

The competitive participation experience of the road cyclists varied between 2 and 12 years with an average of 5.6 years. The average years of cycling experience among the successful group was 6.64 ± 2.74 years compared with the less successful group, which showed a value of 5.09 ± 2.51 years. The successful group obtained a medium practical significant higher experience value (d ∼0.6) than the less successful group. The training regimen of road cyclists during the precompetition period showed that they trained for 6 days a week that included road and resistance training. Training duration was between 18 and 24 hours a week. Daily training session distances were more than 50 km, which amounted to more than 300 km a week. Both the successful and less successful cyclists were subjected to the exact same precompetition training regimen.

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the ethical guidelines of the National Health Research Ethics Council of South Africa. The study was approved by the Health Research Ethics Committee of the North-West University where the research was conducted (NWU-00034-10-A1). All cyclists were informed of the benefits and risks of the investigation before signing an institutionally approved informed consent document to participate in the study. Subjects were requested to sleep at least 8 hours before the different testing sessions. They also had to abstain from ingesting any drugs or to participate in strenuous physical activity that may influence the physical or physiological responses of the body for at least 48 hours before the scheduled tests. Subjects were also requested to maintain the same diet during the week of testing. However, subjects' food intake was regarded to be the same during the testing period as subjects stayed together in a venue of the training center, which also provided all their meals. Subjects arrived at the laboratory in a rested and fully hydrated state. Fluid intake during testing was restricted to water, which was permitted ad libitum.

Procedures

Before the start of the testing period, researchers explained the study design, purpose, and possible risks of study participation in detail to the cyclists and coaches. Thereafter, written and signed informed consent was obtained. This was followed by the gathering of demographic, anthropometric (body mass and stature), and flexibility data. Subjects were then subjected to a warm-up period which consisted of a 10-minute cycling warm-up on their own bicycles which were mounted on a Cateye Cycle Simulator. The warm-up took place at a workload of 150 W with a cadence of 90 rpm. This was followed by the AST, the VJT, and completion of the graded V̇o2max exercise test. After a recovery period of 48 hours, the next testing session followed at the same time of day to minimize the effects of circadian variations on the different metabolic responses. Again, a warm-up was performed before the WAnT was executed.

Demographic and General Information Questionnaire

Cyclists' demographic information (nationality and age) was collected using a demographic and a general information questionnaire. The questionnaire was also used to determine cyclists' medical status, participation levels. and training habits.

Anthropometric Measurements

Body mass was recorded to the nearest 0.1 kg using a calibrated BFW 300 Platform scale (Adam Equipment Co., Ltd., United Kingdom). The body mass measurement was used for the calculation of the WAnT resistance and the power that was exerted during the VJT and the graded V̇o2max exercise test. Relative, explosive, jumping power; WAnT peak power, average power, power drop, total work, and power output during execution of the graded V̇o2max exercise test were all calculated and expressed relative to body mass to scale for different body dimensions between cyclists.

Flexibility Tests

The passive straight leg-raise test (PSLRT) was executed according to the method of Maud and Kerr (31). An intraclass correlation coefficient of 0.87 was reported for test-retest reliability of this test in a group of athletes (8). A standard plastic goniometer was centered over the greater trochanter of the right femur, with the mobile arm pointed toward the lateral epicondyle of the femur and the fixed arm aligned with the lateral midline of the pelvis. The same measurement was taken for the left leg. Each measurement was taken twice. If a difference of more than 5° was observed between the 2 measurements, a third measurement was taken. The lowest measurement was used as the final value.

The Modified Thomas Iliopsoas Test (MTIT) was executed according to the method of Harvey and Mansfield (18). An intraclass correlation coefficient (test-retest reliability) of 0.95 was reported for this test in a group of athletes (8). The angle of hip flexion was measured with a standard plastic goniometer centered over the greater trochanter of the right femur, with the fixed arm directed vertically using the spirit level and the mobile arm pointed toward the lateral knee joint line. The same measurement was taken for the left leg. Each measurement was taken twice. If a difference of more than 5° was observed between the 2 measurements, a third measurement was taken. The highest measurement was used as the final value.

The Modified Thomas Quadriceps Test (MTQT) was executed according to the method of Harvey and Mansfield (18). Test-retest reliability of this test as measured through the intraclass correlation coefficient was found to be 0.91 in a group of athletes (8). A standard plastic goniometer was centered laterally at the knee joint line, the fixed arm aligned with the length of the right femur toward the greater trochanter of the femur, and the mobile arm pointed towards the lateral malleolus of the fibula. The same measurement was taken for the left leg. Each measurement was taken twice. If a difference of more than 5° was observed between the 2 measurements, a third measurement was taken. The highest measurement was used as the final value.

Core Strength Test

The AST is a graded test for assessment of core strength and was conducted according to the method of Ellis et al. (14). A study on the validity of the test indicated intratrial differences of 0.3 for younger subjects (26). The AST is performed over 7 stages of which the starting position is a supine lying position on the floor with 90° bent knees, feet without shoes comfortably apart, in contact with the floor and not held. Different stages were performed one after another with 10 seconds in between the execution of each stage. Three attempts for each stage were permitted.

Explosive Power Test

The VJT is regarded to be an objective (r = 0.90) and valid (r = 0.93) test to determine the peak anaerobic power output of subjects (32). The VJT was executed according to the method of Harman et al. (17). The VJT was performed using the Vertec device (Power Systems, Knoxville, TN, USA). The difference between the standing touch height and jumping distance was calculated and recorded to the nearest 0.5 cm. Subjects performed a minimum of 2 trials with a 2-minute rest period between each trial. The better of the 2 trials was used in the final analysis of data recorded.

The Powertimer 300-Series testing system (Newtest Oy, Kiviharjuntie 11, FIN-90220, Oulu, Finland) was used during execution of the VJT to determine subjects' power output. The last-mentioned apparatus is regarded to be a reliable and valid apparatus to measure countermovement jumping height and power of athletes (15). Subjects stood on the Powertimer 300-Series testing system mat and performed 2 jumping trials. Relative jumping power was calculated in relation to body mass. The 2 trials were performed with a 15-second break in between and the better of the 2 trials was used in the final analysis.

Anaerobic Test

The WAnT is considered to be an objective (r = 0.84–0.88) and valid (r = 0.94–0.98) test to determine the anaerobic power and capacity of subjects (20). The test was conducted according to the method of Inbar et al. (20). The WAnT consisted of 30-second maximal pedaling on a Monark bicycle ergometer (model 834 E; Monark Exercise AB, Vansbro, Sweden) at a resistance of 0.1 g·kg−1·body mass for the duration of the test. The bicycle ergometer was modified with racing pedals, racing seat, and racing handlebars. Subjects prepared for the test with a 5-minute standardized submaximal warm-up. Subjects were instructed to sprint maximally from the start of the test and not to pace themselves through the testing period. Peak power, relative peak power, average power, relative average power, power drop, relative power drop, total work, relative total work, and fatigue rate for each subject were derived or calculated from the test.

Aerobic Test

Indirect calorimetry, open-circuit spirometry, and computerized instrumentation were used during a graded V̇o2max exercise test to measure cyclists' aerobic capacity or maximum aerobic power. Subjects performed the direct standard incremental V̇o2max test to the point of exhaustion on a Cateye Cycle Simulator (model CS-1000; Cateye, Higashi Sumiyoshi-ku, Osaka, Japan). This wind trainer is regarded to be a valid and reliable apparatus to assess V̇o2max using a speed-ramped protocol (12). Subjects' own bicycles were mounted on the Cateye Cycle Simulator. The rear tire was inflated to a pressure of 117.6 psi and put in contact within the rotating axle of the apparatus. The standard incremental test started at 135 W, after which the power output was increased by 45 W every minute. Expired air was continuously sampled using a portable wireless Oxygen Mobile Ergospirometry System (Version 5.0; Jeager, VIASYS Healthcare, Inc., Conshohocken, PA, USA) and the rate of oxygen consumption (V̇o2), carbon dioxide production (V̇Co2), minute ventilation (), and the respiratory exchange ratio (RER) were recorded and calculated every 5 seconds using an online computer system. Throughout the test, heart rate was recorded for each 5-second period using a Polar Heart Rate Transmitter Belt (Polar Electro, Kempele, Finland).

The maximum RER (RERmax) was defined as the highest RER value reached during any stage of the graded V̇o2max exercise test. Maximum relative power output (Wmax) was taken as the highest relative power output that was reached and maintained for at least 30 seconds during the graded V̇o2max exercise test. The Oxygen Mobile was calibrated according to the manufacturer's specifications every day before initiating testing. The test was stopped if the subject indicated the test must be stopped or if the criteria for reaching the V̇o2max value was achieved (e.g., an RER value higher than 1.15, oxygen consumption ceased to rise and reached a plateau or began to fall with an increase in work rate, or the maximum age-specific heart rate was reached) (33).

Ventilatory Threshold Point and Respiratory Compensation Point

VT1 was determined using the criteria of an increase in with no increase in and departure from the linearity of (7). The RCP was taken as the point that corresponds to an increase in both and (7). VT1 and RCP were visually detected by 2 independent experienced sport scientists. The relative power output and the V̇o2 corresponding to VT1 and RCP, respectively, were also determined.

Blood Lactate

Peripheral blood was analyzed from the finger tips 5 seconds before the end of each stage of the incremental test using a Simplified Blood Lactate Test Meter Lactate Pro LT-1710 (Arkray Factory, Inc., KDK Corporation, Shiga, Japan). The Lactate Pro Lactate Test Meter is a valid (r = 0.975–0.993) and reliable instrument (3% coefficient of variance) for the measurement of blood lactate (35). The lactate test meter was calibrated according to the manufacturer's specifications at the beginning of each test. The maximum blood lactate value (BLmax) was taken as the highest blood lactate concentration that was measured during any stage of the standard incremental test.

Statistical Analyses

Statistical analyses were conducted using the StatSoft Statistical Data Processing package (version 13.2; Dell, Inc., Tulsa, OK, USA). Descriptive statistics for each variable and each group were calculated, followed by the calculation of the Cohen's effect size (ES) for the determination of the practical significance of differences between the successful and less successful groups. Only practical significance was determined because a selected subject group was used in this study. Effect sizes (expressed as Cohen's d value) were interpreted as large when ES ∼0.8, moderate when ES ∼0.5, and small when ES ∼0.2 (24). Pearson product-moment correlation coefficients were then calculated between the different physical and physiological variables that were used to test for colinearity between variables. The researchers set the cut-off value for the detection of multicolinearity between variables at 0.70.

To screen for the predictive value of different laboratory-based physical and physiological (independent) variables in predicting group classification of subjects in successful and less successful groups (dependent variables), a binary forward stepwise logistic regression was performed. The significance of individual logistic regression coefficients for each independent variable was determined by making use of the Wald statistic. The level of significance was set at p ≤ 0.05. In the final model, the odds ratios (ORs) and 95% confidence intervals for all individual variables were calculated. The Hosmer and Lemeshow chi square goodness-of-fit test was used to test for the adequacy of the physical and physiological component-related logistic model with an adequately fitted logistic model being indicated by a nonsignificant χ2 value (p ≥ 0.05). The predicted probabilities of being in the successful or less successful cycling group were computed by making use of the logistic regression formula. The percentage correct value gave an indication of the accuracy of the logistic regression model. The usefulness of the model was determined by calculating the “hit rate” and “chance hit rate.” The model was deemed to be good if the observed “hit rate” was 25% better than the “chance hit rate.” The “better than change” index (I) was calculated by comparing the actual or observed “hit rate” with the “chance hit rate” to verify the last-mentioned result and also to determine the ES of the prediction model. I-values were categorized as being small when I < 0.1, as being medium if 0.15 < I < 0.25, and as being large if I > 0.3. The forward stepwise logistic regression analysis was performed by SAS for Windows 7 (version 9.7; SAS Institute, Inc., Cary, NC, USA).

Results

Descriptive statistics of the total, successful, and less successful groups as well as the practical significance of differences in physical and physiological variables between groups of male road cyclists are presented in Tables 1–3.

T1
Table 1.:
Descriptive statistics and associated d-values for the flexibility-related differences between groups (mean ± SD).*
T2
Table 2.:
Descriptive statistics and associated d-values for the abdominal stage test, vertical jump test, and Wingate anaerobic test–related differences between groups of subjects (mean ± SD).*
T3
Table 3.:
Descriptive statistics and associated d-values for the graded V̇o 2max exercise test–related differences between groups of subjects (mean ± SD).*

Flexibility-Related Variables

No large practical significant differences were found for the flexibility-related variables of the successful and less successful road cyclists.

Abdominal Stage Test, Vertical Jump Test, and Wingate Anaerobic Test–Related Variables

According to Table 2, almost all the WAnT-related variables were practical significant higher (d ∼0.8) for the successful compared with the less successful group except for fatigue rate (medium: d ∼0.5) and lowest relative power (small: d ∼0.2). No large practical significant differences were observed for AST- and VJT-related variables between groups.

The Standard Incremental Test–Related Variables

Large practical significant differences (d ∼0.8) between groups were observed for VT1 whether this threshold was expressed as a percentage of V̇o2max, percentage of Wmax, relative power output, or RER (Table 3). The second threshold, namely RCP, only showed large practical significant differences between groups when expressed as relative power output, blood lactate concentration, or RER. Several of the maximum physiological variables, namely Wmax, RERmax and BLmax, also displayed large practical significant values for the successful compared to the less successful group of road cyclists. No other large practical significant differences were noted with regards to the standard incremental test-related variables.

Colinearity Between Different Physical and Physiological Variables

Results of the Pearson product-moment correlation analysis to test for colinearity between different physical and physiological variables reduced the number of variables from 32 to 16. Variables that remained were: R-PSLRT, R-MTIT, R-MTQT, AST, VJT distance and relative power, WAnT lowest and peak relative power, VT1 (%HRmax, %V̇o2max), RCP (%HRmax, %V̇o2max, W·kg−1, RER), V̇o2max (ml·kg−1·min−1), and HRmax.

Forward Stepwise Logistic Regression Results

Table 4 presents the forward stepwise logistic regression results of the physical and physiological variables that acted as predictors between the 2 groups of road cyclists.

T4
Table 4.:
Summary of the odds ratios, 95% confidence intervals, and p-values obtained from the forward stepwise logistic regression analysis.*

Only 5 of the possible 16 physical and physiological variables were identified by the forward stepwise logistic regression to be primary predictors of successful and less successful amateur road cyclists, namely AST level, VJT distance, WAnT relative peak power, and RCP expressed as a percentage of V̇o2max or relative power output. According to Table 4, AST level, WAnT relative peak power, and RCP expressed as relative power output were the only significant predictors of successful and less successful road cyclists. The OR estimates indicate that VJT distance and RCP (%V̇o2max) did not obtain significance and therefore had no meaningful effect on group prediction.

The Hosmer and Lemeshow's goodness-of-fit test was used to assess the significance of the physical and physiological component-related logistic model (which is an indication of the adequacy of the model) and calculated a χ2 = 9.05 (p = 0.25), which indicates that the model's estimates fit the data at an acceptable level (p > 0.05). The predicted probabilities of being in the successful or less successful group of road cyclists were computed by making use of the logistic regression formula. Results of this analysis are presented in the form of a classification table (Table 5).

T5
Table 5.:
Classification table of the predicted probabilities of being in the successful or less successful road cycling group.

Results indicate that 82.84% of cyclists could again be classified into their respective groups by making use of the physical and physiological component-based logistic regression formula. For researchers to determine the accuracy of the percentage correct value, the “hit rate” and “chance hit rate” were calculated. The “hit rate” and “chance hit rate” were calculated to be 84.4% (38/45 × 100) and 57.04% {([14/45]2 × 100) + ([31/45]2 × 100)}, respectively. In this case, the “hit rate” was 27.36% higher than the “chance hit rate,” which indicates that the logistic model of identified physical and physiological variables is useful (“hit rate” was more than 25% better than the “chance hit rate”) in predicting different groups. The “better than chance” index (I) was also calculated to verify the last-mentioned result as well as to determine the ES of the prediction model. An I-value of 0.64 was found, which shows that the model has a large effect in predicting the classification of successful and less successful cyclists.

Discussion

To the best of the researchers' knowledge, this is the first study of its kind to evaluate the power of different laboratory-based physical and physiological test results to predict the “real-life” performance levels of amateur road cyclists. Furthermore, this is also the first study that has extensively evaluated the physical and physiological performance-related components of male, amateur road cyclists from Africa. Abdominal stage test level, VJT distance, WAnT relative peak power, and RCP expressed as a percentage of V̇o2max and relative power output were identified as primary predictors, whereas only AST level, WAnT relative peak power, and RCP expressed as relative power output were identified as significant predictors of performance levels in male, amateur road cyclists. However, the OR estimates indicated that VJT distance and RCP (%V̇o2max) had no meaningful effect on group prediction. Further analyses revealed that the logistic model could be regarded as an adequate, accurate, and useful model to predict amateur, male road cyclists' performance levels. This contention was further substantiated by the fact that 82.84% of cyclists could again be classified into their respective groups by making use of the physical and physiological component-based logistic regression formula.

Abdominal stage test level, which gives an indication of cyclists' abdominal muscle strength, is not a component that has drawn the attention of many researchers. Researchers are, however, of the opinion that the core of which the abdominal musculature forms part provides stability to the trunk and a solid foundation from where pedaling power can be generated (16). Furthermore, Abt et al. (1) demonstrated that knee sagittal and frontal plane as well as ankle sagittal plane motion (°) during cycling were significantly altered because of core fatigue. They also contended that a disruption in core stability would result in the reduction of pedaling effectiveness during the recovery phase (1). A strong core would, therefore, assist cyclists in maintaining a neutral pelvic position on the bicycle together with a good lower-extremity alignment (1). All these factors would probably allow cyclists to generate greater power due to the better leverage from which the psoas and gluteal muscles can contract (23).

Although no recent studies evaluated the VJT heights of road cyclists, existing findings suggest that the average VJT height of amateur road cyclists is a predictor of group classification (42). Others also found significant correlations between VJT height and sprint cycling performance (between r = −57 and −0.65; p ≤ 0.05) (39) and reported that VJT height served as a strong predictor of peak power and the total work done during a maximal-effort cycle ergometer test among amateur road cyclists (34). These findings may explain the identification of VJT height as a group classification predictor. The fact that WAnT relative power was also identified by the logistic regression as a predictor is probably related to the extremely high instantaneous power outputs of between 800 and 1000 W that road cyclists exert during attacks, counterattacks, hill climbing, and break-away attempts in a race (21).

Relationships between RCP (%V̇o2max and W·kg−1), which was also recognized in this study as a group classification predictor, time-trial performances, and Wmax (r = 0.93, p < 0.05) (36) give an indication of the value of this variable for the obtainment of road cycling success. Cyclists who want to be competitive on the international level should possess the ability to perform for extended periods (>60 minutes) at the anaerobic threshold exercise intensity (27). In this regard, Coyle et al. (10) identified percentage of V̇o2max at the LT to be the main contributor to the length of time that road cyclists could maintain 88% of V̇o2max during stationary ergometry. The anaerobic threshold of world-class road cyclists occurs at a high percentage of V̇o2max, which implies that their type I fibers are more fatigue resistant and have the ability to oxidize fat and reduce lactate accumulation at a given workload (7,27). In addition, Mora-Rodriguez and Aguado-Jimenez (36) concluded that RCP is a sensitive predictor of optimal pedaling cadence for road cycling performance. These findings may serve as possible explanations for the identification of RCP (%V̇o2max) and RCP (W·kg−1) as nonsignificant and significant predictors, respectively, of group membership.

In terms of differences between the laboratory-based physical and physiological test results of successful and less successful cyclists, VJT-related variables showed no practical significant differences between groups, which was surprising in view of research findings by White and Al-Dawalibi (42) that VJT height served as a statistically significant discriminator between international and other levels of cyclists. However, some of the WAnT results were consistent with those of Tanaka et al. (40) who also observed a statistically significant higher average WAnT mean power output for more successful compared with less successful road cyclists. In contrast to this study, researchers found no significant differences in WAnT relative peak power between more successful and less successful road cyclists (38,40). However, the nonsignificance of fatigue rate differences between the 2 groups of cyclists is in line with what Tanaka et al. (40) found when the WAnT results of above-mentioned groups of cyclists were evaluated.

VT1 (%V̇o2max,%Wmax, W·kg−1, RER) and RCP (W·kg−1, BL in mmol·L−1, RER) also displayed practical significant differences between groups. Some of these results are in agreement with previous findings, which indicated that absolute and relative power outputs as well as percentage of V̇o2max corresponding to VT1 or RCP, AT, and LT (7,9,27) are significantly different between successful and less successful road cyclists. Moreover, Lucia et al. (27) demonstrated that successful (professional) road cyclists displayed significantly higher RCP (RER) during a standard incremental test than their less successful colleagues (amateur road cyclists). However, they together with Lounana et al. (25) did not find significant differences in VT1 (RER) between professional and amateur road cyclists. It was not possible to compare the RCP results (BL in mmol·L−1) of this study with those of other studies because no studies have thus far investigated this.

With regard to the practical significant differences between the maximum physiological variables (Wmax, RERmax, and BLmax) of the 2 road cycling groups, other researchers have also observed significant differences between Wmax and BLmax of successful (professional) and less successful (amateur) road cyclists (7,27). Contradictory to this, no significant differences were reported for Wmax and BLmax between successful (professional) and less successful (amateur) road cyclists in other studies (25,29). Furthermore, no studies that have evaluated the RERmax between successful (professional) and less successful (amateur) road cyclists reported any significant differences (25,27).

In conclusion, although AST level, VJT distance, WAnT relative peak power, RCP (%V̇o2max), and RCP (W·kg−1) were the physical and physiological variables that served as adequate, accurate, and useful predictors of group membership, only AST level, WAnT relative peak power, and RCP (W·kg−1) were identified as significant predictors between the 2 groups of road cyclists. Therefore, the hypothesis that most of the laboratory-based physical and physiological test results will serve as significant, adequate, accurate, and useful predictors of amateur, male road cyclists' performance levels is rejected. Furthermore, only the following 13 of a possible 32 measured physical and physiological variables showed practical significant differences between different participation levels of amateur, male road cyclists: almost all the WAnT-related variables except for fatigue rate and lowest relative power, VT1 (%V̇o2max, %Wmax, W·kg−1, RER), RCP (W·kg−1, BL in mmol·L−1, RER), Wmax, RERmax, and BLmax.

Findings of this study provide insight into an area of research where uncertainty still reigns. However, results must be interpreted with caution because the prediction model was developed specifically for amateur, male road cyclists from Africa. Therefore, the accuracy of performance prediction functions and models of this nature should be tested and measured through longitudinal studies to evaluate its significance, adequacy, accurateness, and usefulness among different populations of road cyclists. Finally, it should be noted that components such as cycling efficiency and economy, which have also been recognized as important functional and physiological determinants of road cycling performance (28), were not measured in this study. As such, it can be recommended that further studies should focus on more elaborate testing protocols, which also include last-named components as part of their testing protocols.

Practical Applications

Results showed that AST level, WAnT relative peak power, and RCP (W·kg−1) were the only laboratory-based physical and physiological test results to significantly predict the “real-life” performance levels of male, amateur road cyclists. As such, tests that are used to evaluate these physical and physiological variables should be included in the laboratory-based testing protocols of road cyclists. Coaches, conditioners, and sport scientists would be able to only use 3 tests (AST, WAnT, and graded V̇o2max exercise test) to accurately monitor cycling performance, especially during the precompetition phase. A testing and monitoring protocol that consists of only a few tests will allow sport practitioners to test more frequently and ensure that results have real value in terms of accurate performance assessment. Finally, it would serve strength and conditioning coaches well to focus on training regimens that help improve cyclists' AST, WAnT, and V̇o2max exercise test results.

Acknowledgments

The authors thank the National Lottery Board of South Africa who provided funding for this work. There are no conflicts of interest. The authors received funding for this work from the National Lotteries Commission of South Africa.

References

1. Abt JP, Smoliga JM, Brick MJ, Jolly JT, Lephart SM, Fu FH. Relationship between cycling mechanics and core stability. J Strength Cond Res 21: 1300–1304, 2007.
2. Amann M, Subudhi A, Foster C. Influence of testing protocol on ventilatory thresholds and cycling performance. Med Sci Sports Exerc 36: 613–622, 2004.
3. Amann M, Subudhi AW, Walker J, Eisenman P, Shultz B, Foster C. An evaluation of the predictive validity and reliability of ventilatory threshold. Med Sci Sports Exerc 36: 1716–1722, 2004.
4. Amann M, Subudhi AW, Foster C. Predictive validity of ventilatory and lactate thresholds for cycling time trial performance. Scand J Med Sci Sports 16: 27–34, 2006.
5. Anton MM, Izquierdo M, Ibanez J, Asiain X, Mendigchia J, Gorostiaga EM. Flat and uphill climb time trial performance prediction in elite amateur cyclists. Int J Sports Med 28: 306–313, 2007.
6. Balmer J, Davison R, Bird SR. Peak power predicts performance power during an outdoor 16.1-km cycling time trial. Med Sci Sports Exerc 32: 1485–1490, 2000.
7. Chicharro JL, Hoyos J, Lucia L. Effects of endurance training on the isocapnic and hypocapnic hyperventilation phases in professional cyclists. Br J Sports Med 34: 450–455, 2000.
8. Cejudo A, Sainz de Baranda P, Ayala F, Santonja F. Test-retest reliability of seven common clinical tests for assessing lower extremity muscle flexibility in futsal and handball players. Phys Ther Sport 16: 107–113, 2015.
9. Costa V, Pertence L, Paton C, De Matos D, Neves MJ, De Lima J. Physiological correlates of 10-Km up-hill cycling performance in competitive cyclists. J Exerc Physiol Online 14: 26–33, 2011.
10. Coyle EF, Coggan AR, Hopper MK. Determinants of endurance in well-trained cyclists. J Appl Physiol 64: 2622–2630, 1988.
11. Coyle EF, Feltner ME, Kautz SA, Hamilton MT, Montain SJ, Baylor AM, Abraham LD, Petrek GW. Physiological and biomechanical factors associated with elite endurance cycling performance. Med Sci Sports Exerc 23: 93–107, 1991.
12. Dalziel W, Hannan S, Watts M, Coombes J. A new valid and reliable windtrainer protocol. J Sports Med Phys Fitness 47: 300–303, 2007.
13. Davison RCR, Swan D, Coleman D, Bird S. Correlates of simulated hill climb cycling performance. J Sports Sci 18: 105–110, 2000.
14. Ellis L, Giastin P, Lawrence S, Savage B, Buckeridge A, Stapff A, Tumilty D, Quinn A, Woolford S, Young W. Protocols for the physiological assessment of team sport players. In: Physiological Tests for Elite Athletes. Gore JC, ed. Champaign, IL: Human Kinetics, 2000. pp. 128–144.
15. Enoksen E, Tønnessen E, Shalfawi S. Validity and reliability of the Newtest Powertimer 300-series® testing system. J Sports Sci 27: 77–84, 2009.
16. Gambetta V. The critical body core. In: Athletic development. Champaign, IL: Human Kinetics, 2007. pp. 157–176.
17. Harman E, Garhammer J, Pandorf C. Administration, scoring and interpretation of selected tests. In: Essentials of Strength Training and Conditioning (2nd ed.). Baechle TR, Earle RW, eds. Champaign, IL: Human Kinetics, 2000. pp. 287–317.
18. Harvey D, Mansfield C. Measuring flexibility for performance and injury prevention. In: Physiological Tests for Elite Athletes. Gore JC, ed. Champaign, IL: Human Kinetics, 2000. pp. 98–113.
19. Hawley JA, Noakes TD. Peak power output predicts maximal oxygen uptake and performance time in trained cyclists. Eur J Appl Physiol Occup Physiol 65: 79–83, 1992.
20. Inbar O, Bar-Or O, Skinner JS. Description of the wingate anaerobic test. In: The wingate anaerobic test. Champaign, IL: Human Kinetics, 1996. pp. 8–24
21. Izquierdo M, Ibanez J, Häkkinen K, Kraemer WJ, Ruesta M, Gorostiaga EM. Maximal strength and power, muscle mass, endurance and serum hormones in weightlifters and road cyclists. J Sports Sci 22: 456–478, 2004.
22. Jobson S, Nevill A, Palmer G, Jeukendrup A, Doherty M, Atkinson G. The ecological validity of laboratory cycling: Does body size explain the difference between laboratory- and field-based cycling performance? J Sports Sci 25: 3–9, 2007.
23. Juker D, McGill S, Kropf P. Quantitative intramuscular myoelectric activity of lumbar portions of psoas and the abdominal wall during cycling. J Appl Biomech 14: 428–438, 1998.
24. Kline RB. Parametric Effect Size Indexes. In: Beyond Significance Testing: Reforming Data Analysis Method in Behavioral Research. American Psychological Association. Washington, DC: American Psychological Association, 2004. pp. 95–142.
25. Lounana J, Campion F, Noakes TD, Medelli J. Relationship between %HRmax, %HR reserve, %, and % reserve in elite cyclists. Med Sci Sports Exerc 39: 350–357, 2007.
26. Lubans DR, Morgan P, Callister R, Plotnikoff RC, Eather N, Riley N, Smith CJ. Test–retest reliability of a battery of field-based health-related fitness measures for adolescents. J Sports Sci 29: 685–693, 2001.
27. Lucia A, Pardo J, Durantez A, Hoyos J, Chicharro JL. Physiological differences between professional and elite road cyclists. Int J Sports Med 19: 342–348, 1998.
28. Lucia A, Hoyos J, Joyos J, Chicharro J. Physiology of professional road cycling. Sports Med 31: 325–337, 2001.
29. Lucia A, Hoyos J, Santalla A, Pérez M, Chicharro JL. Kinetics of in professional cyclists. Med Sci Sports Exerc 34: 320–325, 2002.
30. Mason A, Baldi J. Determinants of 40k cycling time trial performance. J Sci Med Sport 5: S109, 2002.
31. Maud PJ, Kerr KM. Static technique for the evaluation of joint range of motion and muscle length. In: Physiological Assessment of Human Fitness (2nd ed.). Maud PJ, Foster C, eds. Champaign, IL: Human Kinetics, 2006. pp. 227–251.
32. Maud PJ, Berning JM, Foster C, Cotter MS, Dodge C, DeKoning JJ, Hettinga MS, Lampen J. Testing for Anaerobic Ability. In: Physiological Assessment of Human Fitness (2nd ed.). Maud PJ, Foster C, eds. Champaign, IL: Human Kinetics, 2006. pp. 77–91.
33. McArdle WD, Katch FI, Katch VL. Individual Differences and Measurement of Energy Capacities. In: Nutrition and Human Performance (7th ed.). Baltimore, MD: Lippincott Williams and Wilkins, 2010. pp. 225–247.
34. McLean BD, Ellis L. Body mass, thigh volume and vertical jumping ability as predictors of short-term cycle ergometer performance in junior cyclists. Excel 8: 149–153, 1992.
35. Medbø JI, Mamen A, Holt Olsen O, Evertsen F. Examination of four different instruments for measurement blood lactate concentration. Scand J Clin Lab Invest 60: 367–380, 2000.
36. Mora-Rodriguez R, Aguado-Jimenez R. Performance at high pedaling cadences in well-trained cyclists. Med Sci Sports Exerc 38: 953–957, 2006.
37. Noakes TD. Oxygen transport. In: 4th ed. Lore of running. Cape Town, South Africa: Oxford University Press, 2001. pp. 28–114.
38. Sallet P, Mathieu R, Fenech G, Baverel GP. Physiological differences of elite and professional road cyclists related to competition level and rider specialization J. Sports Med Phys Fitness 46: 361–365, 2006.
39. Stone MH, Sands WA, Carlock J, Callan S, Dickie D, Daigle K, Cotton J, Smith SL, Hartman M. The importance of isometric maximum strength and peak rate-of-force development in sprint cycling. J Strength Cond Res 18: 878–884, 2004.
40. Tanaka H, Bassett DR, Swensen TC, Sampedro RM. Aerobic and anaerobic power characteristics of competitive cyclists in the United States Cycling Federation. Int J Sports Med 14: 334–338, 1993.
41. Union Cycliste Internationale. Road about 2017. Available at: http://www.uci.ch/road/about/. Accessed May 26, 2017.
42. White JA, Al-Dawalibi MA. Assessment of the power performance of racing cyclists. J Sports Sci 4: 117–122, 1986.
43. Wilson JW, Elliot BC, Wood GA. Stretch shorten cycle performance enhancement through flexibility training. Med Sci Sports Exerc 24: 116–123, 1992.
Keywords:

road cycling; performance; anaerobic capacity; aerobic capacity

© 2018 National Strength and Conditioning Association