
Vol. 12, Issue 2 (2023)
Special Issue of the 2023 Science & Cycling Congress | Bilbao - June
October 2023
Pages 1–109
Science & Cycling Conference, Bilbao 2023
-
Effects of test duration on pressure distribution, pelvic stability, and subjective perception in laboratory investigations for evaluating bicycle saddles and seat pads
The aim of this study was to investigate the consistency of typical evaluation characteristics used to assess seating systems in cycling for both male and female cyclists, over a ride time of up to 60 minutes. Pressure distribution, pelvic stability, and subjective perception were measured in a laboratory test using a stationary bicycle. Results showed that mean pressure load increased with the duration of the test, particularly in the first 30 minutes, leading to a decrease in overall subjective comfort. The analysis of pelvic motion yielded inconclusive findings for both genders. Based on these results, it is recommended that future tests should last at least 15 minutes per condition to obtain reliable and valid data. -
Gender related differences in pressure distribution, pelvis movement and subjective perception during cycling
This research aims to examine and compare the saddle pressure distribution, pelvis movement and subjective feedback regarding perceived saddle pressure and stability of male and female cyclists. For this corresponding data are recorded after three minutes of sporty cycling. The findings Indicate notable differences, which can be attributed to anatomical variations between genders. These results have practical Implications for the design of saddles and seat pads, as they highlight the need for gender-specific considerations In cycling equipment.
-
Anterior saddle conflict in women cycling: a case study
Women's cycling is growing rapidly, but as it is a recent development, women do not always find products that are suitable for them. It is well known that women are particularly sensitive about their seating position on the bike and have difficulties in finding the right saddle. The choice of saddle is often made by default. We have noticed that an unsuitable saddle can cause the pelvis tilt, as if to avoid resting on an irritating spine. The aim of our case study was to investigate the effects of a custom-made saddle against an anterior saddle conflict that had been going on for several years. Our subject is an international female cyclist who rides for a UCI Women Continental team. She had unilateral vulvar lymphoedema and knee pain on the same side. She was already wearing orthopaedic corrections to stabilise her position on the bike. The stability gain was partial. Knee and seat pain remained. We compared the athlete's original saddle with the custom-made saddle without foam and with foam. Kinematics, kinetics, saddle pressure and comfort perception were measured while riding at 70-75% of estimated MAP and with 2 different cadences: 90-95rpm (power) and 70-75rpm (strength). The bearing surface of the saddle increased significantly. The maximum saddle pressures were reduced. The distribution of lateral and anteroposterior pressures was clearly rebalanced. The saddle was perceived to be clearly comfortable. However, no changes could be detected in the kinematics and kinetics of pedalling. This case study seems to verify that material-specific saddle conflict is a reality. The custom-made saddle could be a solution to this indication. The advantages and disadvantages of the different customised techniques are however still to be studied, as well as the effect they could have on a larger group.
-
A laboratory treadmill for simulation of road surface induced vibrations in cycling
Within the framework of a systematic approach, a novel test setup for the simulation of road surface induced vibrations was developed. This can be used for various questions concerning the oscillatory system cyclist-bicycle. The innovation compared to the state of the art is the separate introduction of vibrations at the front and rear wheel while at the same time enabling a natural riding behavior during pedaling and coasting, even at higher speeds. -
Reliability and Sensitivity of the Notio Device and Aer-oscale Service to Quantify Cyclists' Drag Coefficients in Outdoor Conditions
Notio is a device based on a wind sensor which offers estimates of the CdA (drag coefficient multiplied by the area) of the pair cyclist and bike. Notio is used with specific analysis software, which computes CdA estimates after a ride. The Aeroscale Company proposes a half-day service with their own wind sensor and experimental protocol, to also deliver estimates of the CdA. In both cases, the main objective of a wind sensor is to give estimates in outdoor conditions. The Aeroscale specificity is that all experiments are done without any power sensor, in freewheel. In our study, we experimented Notio device and software as well as Aeroscale Service through an incremental protocol with increasing disks, which led us to obtain sensitivity measures of 4.5% for Notio and 0.5% for Aeroscale, with a reliability of 8% for Notio and 1% for Aeroscale.
-
Automated PTZ framing of track cyclists using timing loops
Track cycling coaches are mostly busy focussing on what their athletes are doing live. Therefore, they often lack the time to capture videos of the performances of their athletes during practice or competition from which the athletes can learn a lot. We propose a system that can automatically capture track cyclists using a PTZ camera and timing loops while limiting the amount of work for the coaches and technical staff.
-
Experimental validation of a method for virtual testing of bike settings
Based on our previous works, we propose a method for virtual testing of bike settings. This allows the biomechanical and aerodynamic properties of a bike setting to be assessed without having to test it in real life. We also perform an experimental validation of the method by comparing measured and simulated values of different indicators: the frontal area, the maximum knee flexion, the maximum knee extension, the hip angle closed and the hip angle open.
-
The effect of symmetry monitoring system on lower limb muscle activation asymmetry in bike field test
Purpose: The aim of this study was to access the effect of lower limb asymmetry in field test when symmetry measurement system intervenes. Methods: This study recruited 14 participants (height: 172.1 3.9 cm, weight: 64.2 5.3kg, age: 21.9 1.6 year-old). The wireless EMG was applied to observe the bilateral muscle activation in lower limb (rectus femoris, bicep femoris, tibialis anterior, gastrocnemius) with and without symmetry monitoring system in outdoor cycling. The asymmetry index (AI) was calculated by the percentage difference between right (R) and left (L) sides. The repeated measurement ANOVA was used. The significance level was set as =.05. Results: The intervention of symmetry monitoring system would improve the bilateral symmetry of rectus femoris and gastrocnemius. Conclusion: The intervention of symmetry monitoring system could improve the asymmetry phenomenon in lower limb, possibly postpone the onset of fatigue and reducer injury rate.
-
The Correlation Between Pedaling Rate and Gross Efficiency of Road Bike Cycling
Abstract: To investigate the relation between the gross efficiency (GE) and the pedaling cadence (rpm) under different intensities of cycling. Methods: Ten amateur male cyclists were recruited. Each participant completed five tests, which were the maximal incremental test, functional threshold power (FTP) test and three multi-cadence ride tests. The three multi-cadence ride tests were conducted under different intensities of personal 90% FTP, 100% FTP and 105% FTP. A repeated measures 2-way ANOVA was adopted to compare the effects of intensity and gear ratio on GE. Pearson’s correlation was conducted to investigate the relation between gross efficiency and cadence (rpm). Results: For the main effect of intensity, no statistically significant difference was shown. For the main effect of gear ratio, GE’s values among gear 1 to gear 4 or 5 didn’t show significant difference while the values in gear 6 or gear 7 mostly demonstrated significant difference to other gear ratio. By splitting the data according to cadence (e.g. 80, 85, 90 and 95 rpm), the original data of GE versus rpm would be divided into two groups. Through linear regression, the rpm data of 90 and lower showed no significant correlation with GE. Conversely, the rpm data higher than 90 were negatively correlated with GE. In the current study, we found that there is a turning point of rpm corresponding to a drop of exercise efficiency. And it seems not a certain cadence, but a range of cadence demonstrates an equivalent and personal optimal GE. Whether the cyclists should pedal at a rpm close to the turning point is valuable for further study.
Keywords: metabolic energy expenditure, cadence, work rate,
-
Acute and chronic effects of training with a fixed gear on pedaling technique
- Introduction
In cycling, the round pedaling technique is characterized by applying force as uniformly as possible throughout the entire pedal cycle (2). This involves applying more force in areas of the pedal stroke such as dead spots or the recovery phase. Although there is still controversy as to which type of pedaling technique is more metabolically efficient (1, 8), the round pedaling reduces the load on the most important muscles for propulsion in cycling, i.e. the knee extensors (3). In this way, the work necessary to maintain the intensity of the exercise is distributed among other leg muscles (3), which would translate into a greater potential to use the knee extensors in the crucial moments of the competition.
Traditionally, fixed-gear training has been considered as one of the main methods to improve the round pedaling technique (5, 6). However, this type of gear does not force the rider to pull up on the pedal during the upstroke, so its effectiveness in improving round pedaling is questionable.
Therefore, the aim of the present investigation was to evaluate the acute and chronic adaptations in the pedaling technique of a group of track and road cyclists while pedaling with both fixed gear and a freewheel.
- Materials and Methods
22 cyclists (13 track and 9 road cyclists) participated in the study (height: 179.5 ± 5.8 cm and 182.3 ± 3.4 cm; weight: 75.7 ± 6.6 kg and 68.0 ± 4.0 kg; age: 21.7 ± 2.4 years and 21.7 ± 2.5 years, respectively) after giving informed consent. Track cyclists were used to train with both fixed gear and a freewheel, while road cyclists had never pedaled before with a fixed gear.
A track bike was equipped with a system for measuring the tangential and radial forces relative to the crank applied to both pedals (PowerForce, O-Tec, Germany). Two identical freewheel sprockets were used for the tests, one of which was modified to remain fixed. These sprockets were mounted on a double sprocket track wheel, so that variability due to the material used was minimized.
With each of the two sprockets, subjects pedaled on a traditional Elite E-motion roller at three cadences (60, 80 and 100 rpm) and two different resistances (low and high) for one minute. Road cyclists were tested first with the freewheel, whilst track cyclists were tested in random order. Forces applied to both pedals were recorded at 1000 Hz and the last 20 seconds of each trial were used for subsequent analysis. To quantify pedaling technique, the following variables were measured: positive impulse (IMP+), negative impulse (IMP-), positive impulse proportion (PIP), index of effectiveness during the whole pedal stroke (IE360), downstroke (IE0-180) and upstroke (IE180-360).
- Results
Due to the overall symmetry existing between the force applied with both legs, only the data relative to the left leg are presented. The low resistance setup of the roller was weight dependent, and increased proportionally with the riders’ weight. There were no significant differences between the fixed gear and freewheel conditions in any condition or group. However, road cyclists showed smaller left leg power and IMP+ on the low resistance conditions. Table 1 shows the results at 80 rpm, as representative of the other cadences.
Table 1. Means ± SD for the variables in Track and Road cyclists for the left leg.
TRACK cyclists
80 rpm, low resistance
80 rpm, high resistance
Fixed Gear
Freewheel
Fixed Gear
Freewheel
Left Power (W)
* 86.7 ± 7.5
* 87.3 ± 7.6
165.9 ± 9
163.5 ± 10.9
IMP+ (N·s)
* 56.6 ± 5.1
* 56.59 ± 5.2
89.72 ± 5.8
89.51 ± 5.9
IMP- (N·s)
-11.7 ± 2.3
-11.92 ± 2.6
-4.27 ± 1.9
-4.85 ± 2
PIP (%)
82.9
82.6
95.5
94.9
IE360
0.45 ± 0
0.44 ± 0
0.65 ± 0
0.64 ± 0
IE0-180
0.76 ± 0
0.76 ± 0
0.78 ± 0
0.78 ± 0
IE180-360
-0.36 ± 0.1
-0.36 ± 0.1
0.03 ± 0.1
-0.02 ± 0.1
ROAD cyclists
80 rpm, low resistance
80 rpm, high resistance
Fixed Gear
Freewheel
Fixed Gear
Freewheel
Left Power (W)
76.7 ± 5.6
76.9 ± 4.8
161.8 ± 6.1
159.4 ± 5.9
IMP+ (N·s)
51.13 ± 5.2
51.81 ± 4.1
87.84 ± 4.1
86.63 ± 3.1
IMP- (N·s)
-11.44 ± 3.2
-11.73 ± 3.1
-3.82 ± 3.1
-3.69 ± 3.2
PIP (%)
81.7
81.5
95.8
95.9
IE360
0.43 ± 0.1
0.42 ± 0.1
0.66 ± 0.1
0.66 ± 0.1
IE0-180
0.76 ± 0
0.76 ± 0
0.79 ± 0
0.79 ± 0
IE180-360
-0.32 ± 0.1
-0.33 ± 0.1
0.12 0.2
0.14 ± 0.2
* Statistically significant between track and road cyclists (p < 0.05).
- Discussion
There is no an acute adaptation to fixed gear pedaling, as both groups had the same values between the two pedaling systems in every condition. This is especially meaningful in road cyclists, as they had never pedaled with a fixed gear before. The differences between track and road cyclists found in this study can be attributed to the slightly higher power at low resistance induced by the higher body mass of track cyclists. The increase in PIP, IE360, IE0-180 and IE180-360 in the high resistance condition is proportional in both groups, and in line with previous research (5). Therefore, it is quite unlikely that track cyclists have modified their pedaling technique due to a long term training with the fixed gear.
- Practical Applications.
Cyclists willing to improve their round pedaling technique should avoid using the fixed gear and seek for other active training strategies like pedaling with independent cranks (3, 7).
Conflicts of Interest: The authors declare no conflict of interest.
References
- Edwards, L.M., Jobson, S.A., George, S.R., Day, S.H., & Nevill AM. (2009). Whole-body efficiency is negatively correlated with minimum torque per duty cycle in trained cyclists. Journal of Sports Sciences, 27(4), 319-25. doi: 10.1080/02640410802526916
- Faria, IE. & Cavanagh, PR. (1978). The physiology and biomechanics of cycling. New York, USA: John Wiley and Sons.
- Fernández-Peña, E., Lucertini, F., & Ditroilo, M. (2009). Training with independent cranks alters muscle coordination pattern in cyclists. Journal of Strength and Conditioning Research, 23(6), 1764-72. doi: 10.1519/JSC.0b013e3181b3e094
- García-López, J., Díez-Leal, S., Ogueta-Alday, A., Larrazabal, J. & Rodríguez-Marroyo, J.A. (2016). Differences in pedalling technique between road cyclists of different competitive levels. Journal of Sports Sciences, 34(17), 1619-26. doi: 10.1080/02640414.2015.1127987
- Hinault, B. & Genzling, C. (1987). Ciclismo con Bernard Hinault. Barcelona: Ediciones Martinez Roca.
- Hottenrott, K. & Zülch, M. (2000). Entrenamiento de resistencia: Preparación física para la bicicleta. Bilbao: Dorleta, S.A.
- Hug, F., Boumier, F. & Dorel, S. (2009). Altered muscle coordination when pedaling with independent cranks. Frontiers in Physiology, 4, 1-7. doi: 10.3389/fphys.2013.00232
- Leirdal, S. & Ettema, G. (2011). Pedaling technique and energy cost in cycling. Medicine & Science in Sports & Exercise, 43(4), 701-5. doi: 10.1249/MSS.0b013e3181f6b7ea
-
Power loss of the chain drive in tandem bicycles
Abstract:
Introduction: Tandem cycling is a Paralympic sport, in which two cyclists ride on one tandem bicycle. The front rider (pilot), is responsible for the steering, braking, changing gears, and tactics. The back rider (stoker), has a visual impairment and is responsible for responding to changes in cadence and generating power. As the average difference in finish time between obtaining a medal and becoming 4th was 0.5% at the Tokyo 2020 Paralympics for tandem cycling track and road time trial events, much time and effort in the competitive preparation is focused on enhancing individual and team performance. One way to increase performance is by minimizing power losses of the chain drive, which consists of two chains in tandem cycling, the primary and secondary chain. This study aimed to determine the mechanical power loss due to the chain drive of a race tandem bicycle and whether power losses in the chain drive are different for the pilot and stoker. Also, the effect of chain wheel size on power loss was studied.Methods: The tests were performed on a tandem bicycle to assess power losses in the chain drive by applying a fixed power input to of the cranks with an electric motor and measuring the power output at the location of the rear wheel. This was done at the rear crank without the secondary chain to simulate a solo bicycle, and with the primary and secondary chain attached to the rear crank to simulate power input only by the stoker on a tandem. The power input was also delivered to the front crank using both primary and secondary chains to simulate power input only by the pilot. In each situation, gears 53-11 and 53-13 were used.
Results: Power losses in the chain drive were significantly affected by bicycle type and the position of the tandem cyclist (p<0.001). Results showed that the power loss in the solo bicycle (mean power loss = 1.9% ±1.5%) was significantly less than at the tandem stoker (3.7% ±2.4%, p<0.001) and pilot (3.0% ±1.8%, p<0.001), with a significant difference between pilot and stoker (p<0.001). However, these differences between pilot and stoker were only present at a power output below ~250W. Results also showed that in tandem bicycles a gear ratio of 53-11 (mean power loss = 4.0% ±2.8%) resulted in a higher power loss than 53-13 (2.9% ±1.7%; p<0.001).
Conclusion: Power loss due the chain drive in tandem cycling is higher than in solo cycling. It is affected by the chain wheel size, with a larger chain wheel resulting in a lower power loss. It is also affected by the location of the power input, with the location of the pilot resulting in a lower power loss than that of the stoker, although only for the lower power zones.
Keywords: transmission efficiency, para-cycling, power output
-
Chain wear and elongation in Elite cycling: a case study
In modern cycling, chain drives are commonly used since they are reliable and very efficient (1). However, their efficiency is known to be a function of chain wear, being highest after a few break-in kilometres and then slightly decreasing with the accumulated kilometres (2). Bicycle mechanics usually use mechanical gauges in order to check the chain elongation and therefore decide when to replace the chain. However, chain elongation can result from both material loss due to friction and mechanical elongation caused by a plastic deformation of the chain elements, which would be independent of material loss. Moreover, the wear of chain link elements is not uniform (3). The aim of this study is to determine the part of plastic deformation on chain elongation. The hypothesis is that plastic deformation is highly responsible for chain elongation. Identical chain spans were tested either on fatigue on a traction machine or in real riding conditions, on a same number of chain cycles (150 000). Since the traction machine acts only in the longitudinal axle, there is no friction occurring during these tests which therefore cause mechanical elongation but no wear. On the other hand, real riding conditions combine both traction and friction. Chain spans elongation was measured throughout the tests. For real riding conditions, parameters such as power output (W), pedalling cadence (rpm), gear ratio, weather and lubrication were controlled. They allowed to determine the chain tension Fn (N), the number of chain cycles and the mechanical work W endured by the chain roller (Nm).
-
Longitudinal bending stiffness of cycling footwear - What is stiff enough?
To investigate the influence of a cycling shoe’s longitudinal bending stiffness on the kinematics of the cyclist’s foot, three identical leisure-cycling shoes for flat pedals with different longitudinal bending stiffness were tested during indoor trials with twelve male subjects riding at five intensities in seated and standing cycling, respectively. Using an infrared-based 3D motion capture system and power output measurement with a crank-based power meter it was shown that power output does not differ significantly between shoes with different sole stiffness, that metatarsophalangeal angle between forefoot and rearfoot is increasing with increasing power and mostly decreasing with increasing bending stiffness of the shoe and that the pedal angle over crank-cycle shows individual differences that can be clustered in three types. Ultimately it can be said that Ethylenvinylacetat (EVA) inlays show excessive bending whereas there is only a small difference between nylon and nylon-carbon inlays.
-
Performances of bicycle tyres - effect of vertical load and inflation pressure
Tyre characteristics can strongly affect bicycle dynamics, therefore the overall bicycle performances. We already know the importance of longitudinal tyre characteristics (commonly known as rolling resistance), but the lateral characteristics are mostly ignored. Proper testing machines for bicycle tyres are required to measure the lateral characteristics, in order to have a full tyre characterization. The paper presents the results from an experimental campaign on road racing bicycle tyre. Vertical load may affect a lot the tyre parameters, as well as inflation pressure. Specifically, a variation in vertical load of 50 N may results also in 10% increase in cornering stiffness, the main parameter used to assess tyre lateral characteristics.
-
Good Vibrations? An investigation examining the effects of speed, tyre pressure and wheel choice on whole-body vibration during road cycling
Whole-body vibration and hand-arm vibration can be detrimental to health leading to numerous musculoskeletal disorders among other issues. The aim of this study was to investigate the effects that speed, tyre pressure and wheel choice had on whole-body vibration (WBV) and hand-arm vibration (HAV) while road cycling. With added analysis provided by the implementation of a power spectral density (PSD). The study was performed around a 1-kilometre loop on varying road surface by one participant. The wheelsets used were provided by Hunt Wheels (Hunt 48 and Hunt 34); each of these were tested at speeds of 20,30, and 40 kilometres per hour (kph), each at a pressure of 60,70,80 and 90 pounds per square inch (psi). Each combination was repeated 7 times. The bike was fitted with six accelerometers to record the vibration; rear hub, saddle and lumbar to measure WBV; front hub, handlebar, and left wrist to measure HAV. The recorded data was then processed using custom MATLAB script to calculate the average exposure over an eight-hour day (A(8)) and the fourth-power vibration dose value (VDV) outlined in ISO-2631. Additionally, a PSD of the vibration was plotted using Welch’s Method which the max power, the frequency where the maximum occurred, and the absolute power were extracted. The VDV occupational limit was exceeded at all speeds with both tyre pressure (p>0.075) and changing wheelsets (p > 0.793) having no significant effect on reducing the value at the rear hub. The A(8) limit value was not passed and the time to reach it was over 3 hours at 40 kph. The max power and absolute power both increased with speed with the wheelset having no significant effect (p>0.108), however pressure did have significant effect, but only between 60 psi and higher pressures at lower speeds. The frequency at which the max power occurred at the lumbar was between 2.77-5.48 Hz which aligns with the destructive frequency of the lower back. WBV while cycling surpasses the occupational limits and changing speed, tyre pressure and wheelsets has no effect on reducing it to safe levels. Further research into products advertised to reduce vibration while cycling would be beneficial to the cycling community.
-
Low cost and personalized highlight generation during cycling races
Our smart Sensor-driven Tripod focuses on supplying a unified sensor / video solution to capture personalized highlights for participants of cycling races. It does so by capturing video and sensor data simultaneously. The clipping mechanism is driven by sensing nearby signals of ANT+ capable devices. The hardware is accompanied by an online interface where participants can manage their sensors and discover for which events they can generate clips. The device has been deployed and tested in two amateur level cycling races in which it provided clips for all but one of the registered participants without interruptions.
-
Coconut water; a sports drink alternative?
Coconut water is used in replacement of sports drinks for hydration during endurance cycling, however evidence to support it is limited. This study determined if drinking coconut water compared to a sports drink altered cycling performance and physiology. In a randomized cross-over trial, 19 experienced male (n=15) and female (n=4) cyclists (age 30±9 years, body mass 79±11 kg, V̇O2 peak 55±8 mL.kg-1.min-1) completed two experimental trials consuming either a commercially available sports drink or iso-calorific coconut water during 90 minutes of sub-maximal cycling at 70% of peak power output, followed by a simulated, variable gradient, 20-km time trial. Blood glucose, lactate, sweat loss, and heart rate were monitored throughout 90 minutes of sub-maximal cycling, as well as time trial performance (seconds) and average power (watts). A repeated measures analysis of variance and effect sizes (Cohen’s d) analysis were applied. There were no significant differences (p ≥ 0.05) between treatments for any measured physiological or performance variables. Additionally, effect size analysis showed only trivial (d ≤ 0.2) differences between treatments for all measured variables, except blood glucose, which was lower in the coconut-water trial compared to the sports drink trial (d = 0.31). Consuming coconut water has a similar effect on cycling time trial performance and physiological responses to consuming a commercially available sports drink.
-
Creatine supplementation during a training camp in young professional cyclists: a randomized controlled trial
Abstract: Background: Growing evidence suggests that creatine supplementation can provide beneficial effects on exercise performance and recovery, particularly in strength/power sports. However, its effects on endurance athletes remain unclear. We aimed to assess the effects of short term creatine supplementation in professional cyclists during a training camp.
Methods: Twenty-three professional U23 cyclist (19±1 years, peak oxygen uptake: 73.0 ± 4.6 ml/kg/min) participated in a six-day training camp. Participants were randomized to consume after each training session either a recovery drink along with a creatine supplement (20 g) (n = 11) or just the recovery drink (n = 12). Indicators of fatigue/recovery (Hooper index, Recovery-Stress Questionnaire for Athletes (RESTQ), countermovement jump), body composition, and performance (1-, 6-, and 12-minute time trials, as well as the estimated critical power) were assessed as study outcomes.
Results: The training camp resulted in a significant (p<0.001) increase of training loads (e.g., +50% increase in training time and +61% Increase In training stress score compared with the preceding month), which induced an increase in fatigue indicators (e.g., time effect for delayed-onset muscle soreness, fatigue, and total Hooper index, p<0.001) and a decrease in performance (e.g., time effect for critical power, p=0.020). A significant group by time effect (p<0.05) was observed for different recovery items (i.e., success, social recovery, and physical recovery) of the RESTQ, but no additional between-group differences were found for any of the analyzed outcomes.
Conclusion: Short-term creatine supplementation seems to exert no consistent beneficial effects strenuous training period in professional cyclists.
-
Continuous Glucose Monitoring of Non-Diabetic Professional Cyclists during a Training Camp
Cyclists undergo demanding efforts both in training and competition that may perturb the glucose metabolism. Recent advances in wearable technology have produced a variety of devices which can be used to monitor biometric parameters in athletes like interstitial glucose. The aim of this study was to describe daily glucose fluctuations using a continuous glucose monitoring device in non-diabetic professional cyclists during a training camp. Values showed stability of interstitial glucose concentration and a relatively low occurrence of overnight hypoglycemia. These results suggest that the athletes maintained adequate glucose levels and control across the training camp.
-
The higher fraction of V̇O2max during interval training, the greater gains in performance
The fraction of maximal oxygen consumption (VO2max) achieved during a high-intensity interval training (HIT) session and the time it is sustained ≥ 90% of VO2max have been suggested as good criterions for judging the effectiveness of the HIT session. However, scientific findings supporting this assumption are surprisingly still unavailable. To our knowledge, the present study is the first who continuously has measured participants oxygen consumption (VO2) during every HIT session throughout an entire exercise training intervention. Our finding that higher fractions of VO2max elicited during HIT sessions translates into greater endurance performance improvements covers a great gap in the existing literature of applied exercise physiology.
-
Effects of acute hypoxia at rest exposure on time trial performance of national category cyclists.
The aim of this study was to analyze the effects of rest exposure to hypoxia at performance on a simulated cycling time trial (TT) in normoxia cycling. Nine cyclists of national class participated in the study. A single blind, cross-over randomised study in which participants performed two test sessions on two different days in the laboratory was conducted. First, they were exposed to normobaric hypoxia simulating an altitude of 4500 m above sea level (FiO2=0.09) or normoxia for 30 min. After hypoxia or normoxia exposure, participants performed a 20-min simulated TT. Mechanical performance variables, perceived exertion and economy and gross efficiency (GE) amongs other physiological variables were measured. The results showed that hypoxia exposure did not provoke any changes neither in physiological variables, mechanical performance variables and perception of effort (p > 0.05, ES = -0.03 – -0.25, trivial to small), nor in economy and GE (p > 0.05, ES = 0.450, small, -0.139, trivial, respectively). The results suggest that a 30-min rest hypoxia exposure does not affect performance in a 20-min simulated TT in a cycle ergometer under normoxic conditions.
t
-
Physiological Characteristics Associated with W′ and W′bal Used During Intermittent Exercise Task to Failure
The critical power (CP) and W′ concept has become more integrated within applied cycling performance assessment. Both parameters can be tested for in the laboratory and field and have thus become a useful tool for coaches and athletes. The development of the Wbal concept allows us to mathematically model the depletion and recovery of W′ (Skiba et al., 2012, Skiba and Clarke, 2021). When exercising above CP, there is a proposed linear utilisation of W′. When exercising below CP, W′ is reconstituted (W′rec). When W′ has been fully utilised and Wbal (the balance between utilisation and reconstitution) reaches 0, it is assumed exhaustion will occur.
Despite the attention given to the underpinning physiological determinants of CP which are well established (Poole et al., 2016; Mitchell et al., 2018 and Chorley et al., 2020) the physiological determinants of W′ are not as well understood. W′ correlates with muscle volume in elite track cyclists (Kordi et al., 2021) yet there are no correlations with the muscle fibre composition (Vanhatalo et al., 2016; Mitchell et al., 2018), muscle capillarity (Mitchell et al., 2018) or mitochondrial content and function (Rogers, Ferguson et al., unpublished). It has, however, been suggested that so-called muscle typology influences W′rec following high intensity exercise (Lievens et al., 2020). Furthermore W′rec differs between intensity domains (Lievens et al., 2021) with Caen et al. (2021) concluding that W′rec following exhaustion is based on parameters of aerobic fitness due to the faster onset of V̇O2 kinetics, and thus W′rec has been demonstrated to follow a two-phased exponential time course. Despite this, more work is needed to understand the factors influencing W′ and W′rec so we can have greater confidence in its use as a performance and training tool.
As part of a wider PhD research project the associations between physiological performance characteristics, W′ and W′rec were examined. The initial aim of the study was to assess the relationship between physiological performance characteristics and W′. The second aim was to then determine which characteristics were associated with the total amount of work done above CP (W′total) during three different intermittent exercise protocols. Finally, W′rec parameters were modelled, using different Tau calculations to assess which model would predict a Wbal of 0 at exhaustion.
Thirteen endurance trained cyclists (10 males, 3 females; age: 23[8] y, height; 177 [7] cm, body mass; 69.2 [8.3] kg, V̇O2max; 54.3 [11.6] mL·min-1·kg-1, MAP; 371 [70] W, PMAX 1214 [307] W, CP; 270 [49] W, W′ 20.5 [6.2] kJ, mean [SD]) were recruited for this laboratory-based investigation. Participants attended the laboratory on seven separate occasions for the determination of V̇O2max and maximal aerobic power (MAP), lactate threshold (first LT [LT1], baseline + 0.4 mMol.L-1); second LT [LT2, fixed blood lactate concentration of 4 mMol·L-1), CP and W′ (3 and 12-minute method, Simpson and Kordi, 2017; peak lactate [Bla-]Peak and lactate clearance rate [BLa-]CLR and maximal sprint power ([Pmax; peak 1s power). All performance tests were performed on the participants own racing bicycle attached to a Wahoo Kickr with self-selected power chosen for the CP trials and main trials completed in ERG mode. Three intermittent exercise trials were also completed. Intervals were performed at a power output equivalent to the calculated 6-minute power (P6) + 50% of the difference between P6 and CP. Recovery power was 50% of LT1. The three intermittent trial protocols involved: 3 x 60s efforts with 30s recovery proceeded by a sustained time to exhaustion (TTE); 3 x 20s with 10s recovery proceeded by a TTE; continued 20s efforts with 10s recovery until exhaustion, examining total work done above CP (W′total). Different W′bal Tau calculations were used (Skiba at al., 2012; Bartram et al., 2018 and Pugh et al., 2021 (Reg & Nat). Individualised Tau (TauINDV) using a single value to give Wbal 0 at the point of task failure were also assessed. To establish relationship between W′, W′total, and TauINDV, Person′s product-moment correlations were performed.
W′ was associated with Pmax (r=.822 p=<0.001), MAP (r=.735, P=0.004), peak [BLa-] (r=.645, p= 0.017) and V̇O2max (r=.645, p=0.026). W′total was correlated with CP·kg-1 (r=.766, p=0.002), LT1·kg-1 (r=.771, p=0.006) and PMax·kg-1 (r=.792, p=0.001). W′bal at the end of the 20:10 until exhaustion trial using the following Tau values were, Skiba, -20.3 [12.6] kJ; Bartram, 8 [3.5] kJ; Pugh Nat, -3.8 [4.3] kJ; and Pugh Reg, -2.6 [4.7]. TauINDV was correlated with CP·kg-1 (r=0.696 p=0.012), and there tended to be a correlation with LT1·kg-1 (r=0.564, p=0.056).
These data suggest that W′ has many contributing factors, with Pmax and MAP being of particular importance, likely reflecting the extent of muscle mass usually associated with high absolute power outputs. We also demonstrate that CP, LT1 and Pmax normalised to body mass are important factors for W′total during severe intensity intermittent exercise, likely supported by associated metabolic adaptions such as large capillary networks enhancing oxygen extraction.
Implications for applied practitioners.
Our findings confirm the possibilities to utilise the W′ and W′bal concepts within applied cycling performance assessment. We suggest Pmax should be part of the CP testing protocol.
We also propose the use of Tau as a training metric, allowing the assessment of whether an individual is improving their rate of recovery, something that has yet to be considered. However, due to the large variances in the W′bal remaining at the point of exhaustion and task failure, as has previously been suggested, the current models need further improvement with a more standardised protocol to establish individualised Tau values. Nice!
Bartram, Jason C., Dominic Thewlis, David T. Martin, and Kevin I. Norton. 2018. ‘Accuracy of W’ Recovery Kinetics in High Performance Cyclists-Modeling Intermittent Work Capacity’. International Journal of Sports Physiology and Performance 13(6):724–29. doi: 10.1123/IJSPP.2017-0034.
Caen, Kevin, Gil Bourgois, Charles Dauwe, Laura Blancquaert, Kobe Vermeire, Eline Lievens, J. O. Van Dorpe, Wim Derave, Jan G. Bourgois, Lauren Pringels, and Jan Boone. 2021. ‘W ′ Recovery Kinetics after Exhaustion: A Two-Phase Exponential Process Influenced by Aerobic Fitness’. Medicine and Science in Sports and Exercise 53(9):1911–21. doi: 10.1249/MSS.0000000000002673.
Chorley, Alan, and Kevin L. Lamb. 2020. ‘The Application of Critical Power, The Work Capacity above Critical Power (W´), and Its Reconstitution: A Narrative Review of Current Evidence and Implications for Cycling Training Prescription’. Sports 8(9). doi: 10.3390/sports8090123.
Kordi, Mehdi, Campbell Menzies, and Len Parker Simpson. 2018. ‘Relationship between Power–Duration Parameters and Mechanical and Anthropometric Properties of the Thigh in Elite Cyclists’. European Journal of Applied Physiology 118(3):637–45. doi: 10.1007/s00421-018-3807-1.
Lievens, Eline, Malgorzata Klass, Tine Bex, and Wim Derave. 2020. ‘Muscle Fiber Typology Substantially Influences Time to Recover from High-intensity Exercise’. Journal of Applied Physiology 128(3):648–59. doi: 10.1152/japplphysiol.00636.2019.
Lievens, Maarten, Kevin Caen, Jan A. N. G. Bourgois, Kobe Vermeire, and J. A. N. Boone. 2021. ‘W ′ Reconstitution Accelerates More with Decreasing Intensity in the Heavy- versus the Moderate-Intensity Domain’. Medicine and Science in Sports and Exercise 53(6):1276–84. doi: 10.1249/MSS.0000000000002574.
Mitchell, Emma A., Neil R. W. Martin, Stephen J. Bailey, and Richard A. Ferguson. 2018. ‘Critical Power Is Positively Related to Skeletal Muscle Capillarity and Type i Muscle Fibers in Endurance-Trained Individuals’. Journal of Applied Physiology 125(3):737–45. doi: 10.1152/japplphysiol.01126.2017.
Poole, David C., Mark Burnley, Anni Vanhatalo, Harry B. Rossiter, and Andrew M. Jones. 2016. ‘Critical Power: An Important Fatigue Threshold in Exercise Physiology’. Medicine and Science in Sports and Exercise 48(11):2320–34. doi: 10.1249/MSS.0000000000000939.
Pugh, Charles F., C. Martyn Beaven, Richard A. Ferguson, Matthew W. Driller, Craig D. Palmer, and Carl D. Paton. 2022. ‘Critical Power, Work Capacity, and Recovery Characteristics of Team-Pursuit Cyclists’. International Journal of Sports Physiology and Performance 17(11):1606–13. doi: 10.1123/ijspp.2021-0478.
Simpson, Len Parker, and Mehdi Kordi. 2017. ‘Comparison of Critical Power and W′ Derived from 2 or 3 Maximal Tests’. International Journal of Sports Physiology and Performance 12(6):825–30. doi: 10.1123/ijspp.2016-0371.
Skiba, Philip F., Weerapong Chidnok, Anni Vanhatalo, and Andrew M. Jones. 2012. ‘Modelling the Expenditure and Reconstitution of Work Capacity above Critical Power’. Medicine and Science in Sports and Exercise 44(8):1526–32. doi: 10.1249/MSS.0b013e3182517a80.
Skiba, Philip F., and David C. Clarke. 2021. ‘The W′ Balance Model: Mathematical and Methodological Considerations’. International Journal of Sports Physiology and Performance 16(11):1561–72.
-
Modelling Human Endurance: Power laws vs Critical power
The power–duration relationship describes the time to exhaustion for exercise at different intensities. It is generally believed to be a “fundamental bioenergetic property of living systems” that this relationship is hyperbolic. Indeed, the critical- power model which formalises this belief is the dominant tool for studying endurance exercise, e.g. in cycling, running, rowing, or swimming. However, this model is now the focus of a heated debate because it unrealistically represents efforts that are short (< 2 minutes) or long (> 15 minutes). We contribute to this debate by demonstrating that the power--duration relationship is not hyperbolic (and thus cannot be adequately described by a “critical power’’ parameter). Rather, the power–duration relationship is more adequately represented by an alternative power-law model. In particular, the oftenobserved good fit of the critical-power model to exercise durations between 2 and 15 minutes should not be taken as proof that the power–duration relationship is hyperbolic. Rather, in this range, a hyperbolic function just happens to approximate a power law fairly well. Additionally, we show that the power-law model sheds light on the strengths and limitations of many popular rules of thumb for performance prediction such as Jack Daniel’s “VDOT” tables in running and FTP in cycling. For instance, for the latter, it explains why the 95-% rule for estimating a cyclist’s 60-minute power output from a 20-minute effort is too optimistic.
-
Predicting performance in sub-10s f200 m male track sprint cyclists
Maximal power output (Pmax) and the ability to sustain power output close to Pmax are primary determinants of track sprint cycling performance. Given Pmax is achieved at a specific optimal cadence (Fopt) the importance of gear selection is paramount. To optimise gear selection for a specific event, individualised fatigue rates (i.e., fatigue rate per maximal pedal stroke) and field-derived Torque- & Power-Cadence profiles can be used in combination with physics-based model of track cycling. The aim of this investigation was to produce a model of track sprint cycling that can accurately predict performance times of the f200-m. The model utilised mechanical profiles derived from laboratory and field testing as the input variable to optimise f200-m gear selection and performance in elite and world-class track sprint cyclists. Six elite male track-sprint cyclists (Pmax = 2146 ± 423W) completed two testing sessions to identify Torque-Cadence (T-C) and Power-Cadence (P-C) profiles, while fatigue rates were identified during maximal sprints performed at Fopt. The P-C and fatigue profiles were utilised to predict power output during a f200-m event and, in conjunction with a physics-based model of track cycling, predict performance times. This physics model was also utilised to simulate the f200-m performance with different gear ratios. The gear ratio that resulted in the fastest f200-m time was deemed to be theoretically optimal for each athlete. There was no significant difference between the modelled (10.23 ± 0.60s) and actual (10.23 ± 0.53s) f200-m times (p=0.9254). The model predicted that three of the athletes could theoretically improve f200-m performance by increasing the gear ratio, while three could theoretically benefit from a lower gear ratio. P-C and fatigue rate profiles in combination with a physics-based model of track cycling can be used to accurately predict f200-m times.
-
Optimizing the high-intensity interval session through workload alterations in well-trained female and male cyclists
High-intensity interval training (HIT) sessions with a variable workload during the work periods has been considered more effective compared to HIT sessions performed at a constant workload; the optimal HIT design is unknown. This study investigated the isolated effect of power output distribution during three specific HIT sessions, not influenced by factors such as different training duration and different mean power output. 19 well-trained female and male cyclists (age, 22.6 (4.1); maximal oxygen consumption (VO2max), 70.3 (11.0) mL∙min-1∙kg-1) performed three different 6x8-minutes HIT sessions at a mean power output corresponding to the individual participants´ 40-minute maximal power output (FTP) on three consecutive days, in a randomized order: Session 1) 8-minute work periods with 30-seconds work periods at 118% of FTP separated by 15-seconds recovery periods at 60% of FTP (30/15), Session 2) 8-minute work periods performed at a steady power output corresponding to 100% of FTP (CON), Session 3) 8-minute work periods performed as two continuously changing 60-seconds work periods at 110 and 90% of FTP, respectively (60/60). During the HIT sessions, both mean oxygen consumption (%VO2maxHIT) and time ≥90% of VO2max were higher in 30/15 compared to 60/60 (p=0.045 and p=0.042, respectively) and CON (p=0.001 and p=0.001, respectively). In conclusion, 30/15 led to a higher %VO2maxHIT and more accumulated time ≥90% of V̇O2max than both 60/60 and CON, suggesting that this HIT protocol provides the most effective training stimulus.
-
Cycling performance after accumulated load: does durability change during a cycling season?
Introduction:
Durability is the ability to produce high power outputs after accumulated load (at the end of a race), which has been shown to be an important success factor in professional road cycling (van Erp et al. 2020; Leo et al. 2021; Mateo-March et al. 2022; Muriel et al. 2022). Gross efficiency (GE), fat oxidation (FATox) and carbohydrate oxidation (CHOox) have been proposed as underlying physiological factors for durability, as changes have been shown to be related to a decrease in performance after accumulated load (Noordhof et al. 2020; Passfield & Doust, 2000; Stevenson et al. 2022), although there is much debate on this topic. Therefore, the aim of this study is to investigate durability and the underlying physiological factors in semi-professional cyclists. Secondly, this study investigates how durability and certain physiological factors changed during a cycling season.Materials and methods:
In total, 16 semi-professional cyclists (10min PO: 379±22 W) visited the lab on 3 occasions: Start of pre-season (PRE), start of race season (START), and halfway into the race season (IN). Testing days included: a warm-up at 55% of VO2peak power at which GE, FATox and CHOox were determined, followed by a 1min and 10min time trial to measure power output (TT1minfresh, TT10minfresh). After 3hours of endurance training (31±5 KJ/kg) the same protocol was repeated to investigate the influence of accumulated load on performance (TT1minfatigued and TT10minfatigued) and physiological parameters (GE, FATox and CHOox). Differences between PRE vs START vs IN and fresh vs fatigued were investigated using a mixed-effects multilevel model. When significant (p<0.05) main effects occurred, Bonferroni post-hoc was executed for pairwise comparisons.Results:
No differences were found between TT1minfresh at PRE, START and IN and TT1minfatigued at PRE, START and IN, however TT1minfresh was significantly higher than TT1minfatigued at PRE, START and IN (figure 1). TT10minfresh did not differ from TT10minfatigued on all occasions, however, both TT10minfresh and TT10minfatigued were lower at PRE compared to START and IN (figure 1). CHOox (PRE and START) and GE (START) were lower in fatigued compared to fresh state, while FATox was higher in fatigued compared to fresh (PRE and START). In addition, GEfresh was higher at START and IN compared to PRE, while there was a decrease in FAToxfatigued from PRE and START to IN. Lastly, an increase was noted from PRE to START in CHOoxfatigued and PRE to IN in GEfatigued and CHOoxfatigued.Discussion:
This study is to the best of the authors knowledge, the first to present the changes in power output, gross efficiency and substrate oxidation before and after accumulated load during a full cycling season.Performance over short efforts (TT1min) is affected by fatigue but doesn’t improve from PRE to IN. Performance over longer efforts (TT10 min) is not significantly affected by accumulated load, however, improves throughout the season (from PRE to IN). As the total load at the beginning of the fatigued test was ± 40 KJ/kg, it could be suggested that the protocol in the current study was not fatiguing enough, with decline in performance in professional cyclists found after 50 KJ/kg (van Erp et al. 2020). However, other studies already found an earlier decline in performance (Leo et al. 2021; Mateo-March et al. 2022; Muriel et al. 2022). Also, previous studies are based on race data. As not only the accumulated load, but also the intensity of that load plays a role in an impaired performance after exercise (Vermeire et al. 2023), it could be suggested that the intensity of the races was higher as compared to the controlled environment in the current study.
Not only the load, but also intensity plays a role in an impaired performance after exercise
Gross efficiency was not affected by the accumulated load in the current study, which is in contrast with previous research (Noordhof et al. 2020; Passfield & Doust, 2000) . As the level of the cyclist in the current study is higher as compared to previous research, it could be argued that the protocol was not fatiguing enough to result in a decreased efficiency. However, GE seems to improve throughout the season (from PRE to IN).
Substrate oxidation changes after accumulated load with FATox increasing and CHOox decreasing, as in line with previous research (Stevenson et al. 2022). However, this difference diminishes from PRE to IN, because of higher CHOox and lower FATox after accumulated load. It could be suggested that this shows a better durability during the season compared to the pre-season and that the ability to maintain CHOox high in fatigued state results in better performance (although not significant).
-
MMP and Torque – duration profiles from u19 cyclists and WT cyclists
Performance in cycling is often defines by the power produced by an athlete (i.e., power profile). Studies have tried to identify the differences between categories and levels. It has been showed that world tour cyclists have better qualities than the others. Moreover recently, it has been showed that power profile is mainly depend on the torque production capacities. The aim of our study was to investigate the torque at Mean Maximal Power – duration relationship by i) testing the fitting of the relative torque used in MMP from different durations ii) determine the differences between world tour cyclists (WT) and under nineteen cyclists (U19), because of the development strakes of this category. 17 u19 cyclists and 20 world tour participated in this study. Relative Power, cadence and relative torque data from one full season were used to determine the relative mean maximal power and relative mean torque relationships, for 10s, 1min, 2Min, 5min, 10min, 20min and 30min, and critical power was determined with the 3-parameter cp model. The goodness of fit was excellent (r² = .98 [.91-1). Significant differences were found from 5min to 30min and cp were found between the groups for relative power and relative torque data. No differences were found for the cadence. Torque duration relationship can be modelled and use to track performance of cyclists. Differences between world tour cyclists and under nineteen cyclists are mainly due to differences in torque capacities.
-
A Novel Submaximal Field Test of Fatigue Resistance in Professional Cyclists
Induced by a given amount of prior accumulated work, it should be possible to quantify fatigue resistance by measuring changes in perception of effort, not requiring a maximal effort from the athletes. The aim of the study was to investigate the impact of prior accumulated work on perception of effort during submaximal exercise in professional cyclists. Seven male professional cyclists performed a submaximal field test consisting of three 5-minute bouts at three different power outputs in both fresh and fatigued states. Heart rate (HR), ratings of perceived exertion (RPE), and power output (PO) data were collected. Before the submaximal test in the fatigued state total work done was 3244 ± 83 kJ. RPE and HR were higher in the fatigued state than in the fresh state (p < 0.05). Absolute and relative PO during the submaximal test were not different between fresh and fatigued states (p > 0.05). The main findings of the study are that both RPE and HR during submaximal exercise are sensitive indicators of the state of fatigue induced by more than 3000 kJ of work done in professional cyclists. Therefore, we proposed a feasible submaximal exercise protocol to test fatigue resistance in the field.
-
Estimation of Anaerobic Threshold with dynamic transfer function models based on heart rate and power in cycling
The anaerobic threshold is an important marker commonly used in cycling to identify training zones and monitor training progress. However, monitoring of the anaerobic threshold is not performed on a regular basis due to the high costs, invasive blood sampling and time-intensive test protocols. Mathematical modelling might form a suitable alternative for regular monitoring of performance. Furthermore, modelling allows for identifying the effect of certain influential variables on the performance, resulting in performance prediction. They might enable more practical applications, especially in combination with wearable technology. In cycling, the heart rate, power and cadence are already continuously measured and monitored during training and competition, making them particularly suitable for integration in a modelling technique. This study attempts to estimate the anaerobic threshold with linear time-invariant and linear time-varying models based on the heart rate, power.
-
FTP in cycling: functional or fictional threshold power?
Functional threshold power (FTP) is a very popular metric in recreational and competitive cycling to evaluate, monitor and predict performance. This “threshold” is defined as the power output that a rider can maintain for one hour and is mostly determined by calculating 95% of the mean power output performed during a 20-min time trial. The popularity of FTP mainly originates from its practical feasibility, however, the scientific basis of this concept can be questioned. In this presentation, the physiological meaning of FTP and its relationship to other threshold concepts, such as critical power and the maximal lactate steady state, will be critically evaluated. Based on the most recent insights from literature, both the advantages and disadvantages of utilizing FTP in research and practice will be delineated.
-
A short self-paced submaximal test to monitor endurance cycling fitness
Introduction Monitoring the cyclists’ responses to training is important for optimizing training and improving cycling performance (Galán-Rioja, Gonzalez-Ravé, González-Mohíno, & Seiler, 2023; Jeffries et al., 2021). However, this process can be costly and time-consuming. Submaximal fitness tests provide a practical alternative to traditional maximal tests, allowing the monitoring of training effects while minimizing disruptions to the athlete's training and competition schedule (Shushan et al., 2022).Recently, Sangan, Hopker, Davison, & McLaren (2021) proposed the self-paced submaximal run test (SRTRPE) as a more practical form of the running version of the Lamberts and Lambert submaximal cycling test (LSCT) (Lamberts, Swart, Noakes, & Lambert, 2011). The LCST measures cycling power output (w) or running velocity (v) and ratings of perceived effort (RPE) during two 6-minute stages and one 3-minute stage of increasing intensity, which are determined by a fixed percentage of the maximum heart rate (Lamberts et al., 2011; Vesterinen et al., 2016). In contrast, the SRTRPE monitors v and heart rate (HRex) during three 3-minute stages prescribed by RPE 10, 13, and 17 (Sangan et al., 2021).
The self-paced nature of the SRTRPE makes it potentially more specific to the pacing demands of competition and, more importantly, less cumbersome since there is no requirement to collect and to store RPE data for later analysis, nor to do a prior maximal testing to establish the relative intensity of each stage. Furthermore, standardizing intensity by RPE could provide a better insight into the individual's responses to endurance training by reflecting more accurately individual differences in the exercise intensity domains (Iannetta et al., 2020; Seip, Snead, Pierce, Stein, & Weltman, 1991). Sangan et al. (2021) reported promising findings showing that v, at all stages of the SRTRPE, was largely correlated with aerobic fitness parameters such as maximal oxygen uptake (VO2max), v at VO2max (vVO2max), and v at lactate threshold 2 (vLT2). Moreover, test retest analysis showed acceptable reliability with coefficients of variation between 2.5% and 5.6% for v and HRex at all stages (Sangan et al., 2021).
Given the supportive findings and potential benefits of a self-paced design, the aim of our study was to evaluate the feasibility of a cycling-specific adaptation of the SRTRPE protocol. The modified protocol incorporates the latest research on RPE and is tailored towards the specific requirements of road cycling (Ebert, Martin, Stephens, & Withers, 2006; Lopes, Pereira, & Silva, 2022). The self-paced submaximal cycling test (SCTRPE) measures w and HRex responses during three 2-minute stages using Borg's CR100 scale (RPE Moderate, Hard, Very Hard) and evaluates neuromuscular function with one second peak power (wsprint) and post-exercise heart rate recovery (HRR).
Materials and Methods Twenty-three trained to highly trained (29.6 ± 9.2 y, 72.7 ± 9.7 kg, VO2peak 65.6 ± 7.1 ml∙kg-1∙min-1) endurance cyclists (including 3 females) received standardized RPE instructions and anchoring procedures. Then, they underwent the first trial of the SCTRPE. They also performed three 6-second maximal sprints and a graded exercise test (GXT) to determine their maximal neuromuscular one second peak power (wMAXsprint), ventilatory thresholds (VT1, VT2), peak oxygen consumption (VO2peak), and peak power (wpeak). Within a period of 2 to 14 days, the participants completed two additional trials of the SCTRPE to evaluate the test-retest reliability of HRR, wsprint, HRex, and w at each stage of the test. Correlations were used to examine the relationship between the SCTRPE and the parameters measured during the GXT to establish convergent validity. Results Power output measured during all stages of the SCTRPE had moderate to very large correlations with the parameters measured during the GXT, ranging from .48-.85. Notably, the highest correlation (r = .85) was observed between wpeak and w at RPE Very Hard (Figure 2). The intraclass correlation coefficients (ICC3,1) for w and HRex at all stages of the SCTRPE ranged from .86-.92, indicating high to very high reliability, with typical errors (TEs) ranging from 2.7% to 7.9% (Table 1). The reliability of HRR expressed as a percentage of the peak heart rate recorded during the trial (HRR%hrpeak) and wsprint was found to be moderate and very high, respectively, with ICC3,1 value of .74 and .93, and TEs of 4.4% and 6.7% (Table 1). ConclusionsThe study findings suggest that the SCTRPE is a reliable and valid measure of endurance cycling fitness. The high intraclass correlation coefficients and good convergent validity with the parameters of the GXT, particularly wpeak, indicate that the SCTRPE could be an effective and practical test to monitor endurance cycling athletes’ responses to training.
Practical ApplicationsAthletes and coaches using the SCTRPE may be able to observe changes in training status when the power output exceeds the TE measured for each stage. For example, if the power output of a cyclist during the last stage (RPE Very Hard) of the SCTRPE exceeds by 25W his power output of the previous test, we can assume that his cycling fitness is rising because the change is greater than the TE of 22W for this stage.
-
Optimizing performance in cycling through machine learning
In professional sports, optimal performance requires a balance between training and subsequent recovery. To follow-up on this balance, it is important to monitor training load, symptoms of fatigue and predict changes in performance. At present, performance is mostly monitored and predicted based on white-box mathematical models, which were historically based on rigorously configured test protocols and taken under controlled settings with a moderate number of athletes. While these models have clear scientific evidence and provide great value, they are often too coarse grained to assess and predict subtle changes in performance. Moreover, in monitoring performance through e.g., lactate tests, the disruption of the athlete’s training schedule can also not be neglected.
On the other hand, a lot of individual data is available from wearable sensors during training and in everyday life: physiological metrics such as heart rate and heart rate variability, performance metrics such as power and contextual metrics such as altitude. Recent advances in machine learning allow building a more personalized model for an individual athlete: the plethora of data that is available allows building a much more fine-grained and individualized model for monitoring and predicting performance. However, machine learning is facing its own challenges: it is essentially a black-box model and the data that is used is often noisy, as it is collected under non-controlled circumstances.
In this talk, we will explain how we will balance such white-box and black-box approaches. We present advanced machine learning models for approximating the performance measures based on this huge amount of individual data. More specifically, we will present specific machine learning modules for advanced performance monitoring such as estimating individualized fitness-fatigue levels, LT1/2 threshold prediction and automatic training anomaly detection.
-
EFRT: A realistic model of human exercise, fatigue and recovery
We discuss mathematical models for endurance exercise. Such models are needed to accurately assess athletes' fitness (e.g. for guiding training or identifying talent); and to predict performances (e.g. for optimising race strategies and pacing in track-cycling events).
So-called W’-balance models, based around the “critical power” paradigm, are the current state-of-the-art. Unfortunately, W’-balance models cannot account for many essential qualitative features of endurance exercise. For instance, they fail to adequately capture that: (a) low (“heavy” or “moderate”) exercise intensities cannot be sustained indefinitely; (b) pacing impacts exercise tolerance; (c) exercise modality affects subsequent recovery; (d) fatigue from prolonged exercise changes the power–duration relationship (a.k.a. the “durability” concept).
We introduce the exercise, fatigue and recovery tracking (EFRT) model – a novel and rigorous framework for endurance exercise; and demonstrate that the EFRT model is more realistic than W’-balance models in the sense that it can capture all the above-mentioned essential qualitative features of endurance exercise (and many more). The EFRT model is also highly parsimonious: it requires only a small number of parameters and all of these have meaningful interpretations: they represent speed, endurance, durability, and recovery.
-
Will wearable sweat lactate sensors be the next big thing in cycling?
The use of wearable sensors in cycling is rising. Besides heart rate, we can now measure more and more variables continuously and in real-time like glucose (in interstitial fluid), core body temperature (at the heart rate belt), muscle oxygenation (with NIRS at the upper leg), and sweat rate (with wearable sensors). These sensors may provide more insights into what is happening during exercise and possible can be used to optimize performance.
One of the recent additions to the list of variables that can be measured with wearable sensors is the concentration of lactate in sweat. Since blood lactate is a key performance indicator in sport physiology and can be used to monitor exercise intensity, there is a lot of interest in measuring lactate during exercise. Nonetheless, blood lactate has the disadvantages that it is invasive, painful, not continuous, and that the exercise has to be interrupted to measure. If lactate can be measure non-invasively and in real-time in sweat, this will open much more insights and possibilities to monitor lactate concentration during exercise. However, the relationship between sweat lactate and blood lactate still remains unclear. Since sweat lactate concentrations are higher in sweat (1-30 mM) than in blood (1-15 mM), there is debate on the origin of lactate in sweat. Lactate can be produced by the sweat glands itself, derivate from the lactate in blood, or a combination of both. Further research is needed to explore the origin of lactate in sweat.
In the past researchers were using different sweat collection methods to analyze lactate in sweat. With these methods sweat was collected during a certain time period and the lactate concentration was analyzed afterwards. Most of these studies found no or even a negative relationship between sweat lactate concentration and exercise intensity. This can probably be explained by the dilution of the sample due to the increase in sweat rate with increasing intensity. With the recent developments in wearable sweat sensors, sweat lactate can now be measured continuously when there is a constant supply of new sweat. Using these sensors, there is no dilution of the samples which can results in a positive relationship between sweat lactate and exercise intensity (and also blood lactate). This review article will give an overview of the existing literature about the link between sweat lactate (measured with wearable sensors) and exercise, and the usability of wearable sweat lactate measurements in endurance sports like cycling.
In this review we give an overview of the different sweat collection methods and wearable sweat sensors. To explore the relationship between wearable sweat lactate and exercise we searched for relevant articles measuring the sweat lactate concentration with wearable sensors during exercise. We included research articles using sweat lactate during exercise as one of the outcome variables and concept articles where newly developed sweat lactate sensors where tested during exercise. This resulted in 17 articles (Anastasova et al., 2017; Enomoto et al., 2018; Gao et al., 2016; Gillan et al., 2021; Guan et al., 2019; Jia et al., 2013; Karpova et al., 2020; Komkova et al., 2022; Lamas-Ardisana et al., 2014; Mao et al., 2020; Muramoto et al., 2023; Onor et al., 2017; Seki et al., 2021; Vinoth et al., 2021; Xuan et al., 2022, 2023; Zhao et al., 2019). In 10 of these articles they found a positive relationship between sweat lactate concentration and exercise intensity and in 7 articles the found no or a negative relationship. This indicates that sweat lactate can possibly be a non-invasive alternative for blood lactate, but more structured research is needed to explore the origin and mechanisms of lactate in sweat and its relationship with exercise intensity. Sweat lactate can therefore be the next big thing in cycling if these wearable sweat lactate sensors come onto the market.
-
Comparison of mean maximal powers measured in indoor and outdoor tests and recorded during training and competition
The aim of this study was to compare the mean maximal powers (MMP) in road cycling measured during four conditions (training, racing, indoor and outdoor tests). Seven male amateur cyclists completed four test sessions (2 indoors and 2 outdoors) over a 14-day period, consisting of mean maximal power tests for durations of 5, 15, 30, 300 and 1200 s. Indoor testing was done on a home trainer while outdoor testing was on an uphill road. MMPs were also determined from power data recorded for 3-6 months during training sessions and competitions. No significant difference in MMPs was found between road and racing tests, nor between indoor and outdoor tests except for the 5 s duration (-10% for indoor test). The duration MMPs of 5 and 20 min recorded during races, indoor and outdoor tests were significantly higher than during training (+12-14% and +9-11%, respectively). According to these results, cyclists (amateurs) can determine their record power profile for durations of 15 to 1200 s during home trainer tests or uphill road tests or from the race data e, but not at from training data.
