Effects of crank length on cycling efficiency, sprint performance, and perceived fatigue in high-level amateur road cyclists

Abstract

Purpose

This study aimed to evaluate the effects of three different crank lengths (165 mm, 170 mm, and 175 mm) on cycling efficiency, sprint performance, and perceived fatigue in high-level amateur road cyclists.

Methods

A single-blind, randomized crossover design was employed, involving 28 trained male cyclists who completed three trials, each using a different crank length. Performance metrics, cycling economy (CE) including 20-min cycling under 60 % VO_2MAX, peak and average power during a 6s sprint, cadence, heart rate, and perceived exertion, were measured using validated ergometers and metabolic analyzers.

Results

The results indicated significant differences in 60 %VO_2MAXRPE were observed between the 165 mm and 175 mm cranks (p < 0.001, δ = −0.72) and between the 175 mm and 170 mm cranks (p < 0.001, δ = −0.67). In contrast, no significant difference was observed between the 165 mm and 170 mm cranks (p = 0.25, δ = −0.17). For 60 %VO_2MAX, 6sMaxPower, 6sAVPower, 6sAVRpm, CE, 60 %VO_2MAXAVHeart, 6sMaxRpm no significant differences were observed between any of the crank length groups (p > 0.34).

Conclusion

For high-level amateur road cyclists, using a 165 mm or 170 mm crank reduces subjective fatigue compared to a 175 mm crank, influencing affecting cycling efficiency or sprint performance. These findings help refine crank length recommendations to optimize performance, reduce injury risk, and improve comfort for high-level amateur road cyclists.

1. Introduction

Road cycling refers to cycling on paved roads using lightweight bicycles designed for speed and endurance.¹ This activity is both a highly competitive sport, exemplified by professional road racing, and a recreational pursuit, attracting participants ranging from elite athletes to amateur cyclists. Compared to other types of bicycles, such as mountain bikes and leisure bikes, the road cycling bike requires athletes to maintain cycling efficiency at high speeds, optimizing energy conservation to sustain long-endurance cycling (e.g., 150–250 km of riding in a single-day).² As a sport that combines human performance with mechanical devices, cycling efficiency and energy expenditure can be significantly influenced by subtle adjustments in cycling posture, handlebar height, pedal position, crank length, and saddle alignment.^3,4 (see Fig. 1, Fig. 2).

Fig. 1.

The crank of GARMIN NEO BIKE PLUS bicycle ergometer and GARMIN Edge 1040 bicycle computer.

Fig. 2.

The GARMIN NEO BIKE PLUS bicycle ergometer and COSMED metabolic cart.

The crank of a road bike, a key lever connecting the chainring and pedals, has gained increasing attention as part of the bike's mechanical structure. Studies have shown that the crank length contributes significantly to rider's posture, alters the pedal radius, which influences the pedaling torque and cadence technique, and impacts cycling efficiency,^5,6 potentially pertaining to the injury prevention.⁷ Recently, studies have examined the effects of different crank lengths on road cycling performance with the primary focus on the public and professional road cyclists. Specifically, studies have shown that for the general population, shorter cranks, typically ranging from 145 to 165 mm, can reduce muscle fatigue in the knee extensors and joint fatigue, maintain a higher cadence, and improve cycling economy at lower intensities.⁵ However, in high-performance scenarios, where maximum power output is essential, shorter cranks do not provide adequate torque needed for sprints or climbs, limiting their application in competitive cycling.^8,9 Studies have shown that longer cranks, typically 170 mm or more, are particularly beneficial in scenarios demanding maximum power output, such as sprints and climbs, where they optimize cadence and alter the use of different muscle groups, helping to increase instantaneous power and reduce fatigue accumulation during high-intensity sprints.^6,10 As a result, they are preferred by professional road cyclists.¹¹ Research has demonstrated that longer cranks not only boost power output for professional cyclists but, under certain conditions, also improve VO_2MAX and ventilatory thresholds, contributing to sustained high-intensity performance. But, small adjustments in crank length have a minimal impact on cycling efficiency at submaximal intensities.^5,7 On the other hand, there is currently no consensus on the optimal crank length for different populations, and therefore, the optimal crank length for various groups still requires further investigation.⁴ Current, still, the relationship between crank length and performance has not been examined in high-level amateur road cyclists, who share similar performance goals with professional athletes but also value the comfort and lower intensity demands prioritized by recreational cyclists. While high-level amateurs road cyclists engage in higher training volumes compared to the general population, they still have considerable gaps in aerobic and anaerobic capacities when compared to professional cyclists.¹² This intermediate status makes the selection of the optimal crank length a challenging decision. Therefore, it is essential to explore the optimal balance of crank length adjustments for high-level amateur road cyclists, considering both power output and comfort to ensure both performance enhancement and injury prevention.

This pilot study examined the effects of different crank lengths (i.e.,165 mm, 170 mm, 175 mm) on cycling efficiency, short-distance sprint performance, and perceived fatigue in high-level amateur road cyclists. We specifically hypothesized that, compared to the 175 mm crank, the 165 mm and 170 mm cranks may reduce perceived fatigue during cycling without affecting performance.

2. Methods

2.1. Participants

A total of 28 male cyclists were recruited for this study. The inclusion criteria were as follows: participants must have previously competed in amateur road cycling races and had no experience with professional road cycling teams; all cyclists had at least four years of continuous road cycling training, with a minimum of three training sessions per week and regular weekly training volumes of 300–450 km; VO_2MAX ≥ 50 ml/kg/min, and VO_2MAXPower ≥ 4 W/kg.¹³ Exclusion criteria included any known cardiovascular disease, serious musculoskeletal injuries, or other health issues that could impair performance or those who had not participated in regular physical activity in the past six months. No participants reported any injuries or cardiovascular issues, and all signed informed consent prior to participation. The study protocol was approved by the university institutional review board (BSU2024404H) and adhered to the ethical principles of the Declaration of Helsinki.

2.2. Experimental protocol

A single-blinded, randomized crossover design was employed to evaluate the influence of three crank arm lengths (165 mm, 170 mm, 175 mm) on cycling economy, sprint power, and perceived fatigue. A researcher, not involved in the performance testing, was responsible for adjusting and recording the crank lengths. However, given the participants' experience, they may have detected differences in crank length through proprioception. This inherent limitation will be discussed in the limitations section.

The testing was conducted using the GARMIN NEO BIKE PLUS bicycle ergometer, which is specifically designed to replicate the riding posture of road cycling. This design allowed participants to maintain a standard cycling position throughout the trial. In conjunction with this, the GARMIN Edge 1040 bicycle computer was used to synchronously record key performance metrics, including cadence, power output, and heart rate. Together, these devices ensured accurate measurement and consistent cycling posture, closely replicating outdoor cycling conditions. The equipment was fully calibrated, and all participants were given sufficient practice time with the devices before the formal experiment.¹⁴ The testing order for each participant was determined based upon computer-generated random sequences. A rest period of 72 h was provided between trials to mitigate the effects of fatigue and minimize learning, or adaptation influences on the performance in following visits. Prior to each trial, seat height, handlebar height, and fore-aft adjustments were made based on individual anthropometrics and rechecked to ensure consistency throughout the experiment.15, 16, 17

2.3. Experimental procedure

2.3.1. Baseline testing

Participants underwent a VO_2MAX test using a standard 170 mm crank to determine maximal aerobic capacity.¹⁸ Following a 10-min warm-up (100–150 W) and 5-min rest, an incremental ramp protocol began at 150 W, increasing by 25 W every 2 min until volitional exhaustion.¹⁹ The key performance metrics recorded during the test included VO_2MAX (ml/kg/min), which represents the participant's maximal oxygen uptake and their ability to utilize oxygen during intense exercise. Additionally, we recorded VO_2MAXPower (W/kg), which refers to the maximal power output achieved at the point of reaching VO2max, normalized to body weight. Exhaustion was defined by the following criteria, which ensure the measurement of VO_2MAX: Respiratory Exchange Ratio (RER) > 1.1; heart rate (HRmax) ≥ 90 % of the predicted value; Rating of Perceived Exertion (RPE) ≥ 9; a plateau in oxygen consumption (VO₂ plateau), where oxygen consumption remains stable despite an increase in workload; cadence <80 RPM.²⁰

Maximal oxygen uptake (VO_2MAX) was measured using a COSMED metabolic cart (COSMED, Rome, Italy), a reliable and validated tool.^21,22

2.3.2. Cycling economy test

Cycling economy was tested across three sessions with different crank lengths (165 mm, 170 mm, or 175 mm). Before each session, adjustments were checked for consistency.^15,16,23 Each session began with a 10-min warm-up (100–150 W), followed by 20 min of cycling at 60 % VO_2MAX, maintaining a consistent cadence of 85 ± 5 RPM across all crank lengths.²¹ The key performance metrics recorded included 60 %VO_2MAX (ml/kg/min) to assess aerobic efficiency, 60 %VO_2MAXAVHeart (bpm) as an indicator of cardiovascular response, and 60 %VO_2MAXRPE (scale 1–10) for subjective fatigue, Cycling Economy (CE) is typically expressed in units of ml/kg/min, which represents the amount of oxygen (ml/kg/min) consumed per kilogram of body weight per minute of exercise. In this study, the unit is further refined to ml/kg/min/W, which represents the amount of oxygen consumed per watt of power output. This unit provides a more accurate reflection of energy efficiency at specific power outputs, calculated using the following formula,²⁴: calculated using the following formula:

$$CE=\frac{60\%{\text{VO}}_{2\mathrm{MAX}}}{60\%{\text{VO}}_{2\mathrm{MAX}}\text{Power}}$$

Data on oxygen consumption were measured using the COSMED metabolic analyzer.²² Average power, heart rate, and cadence were recorded using the GARMIN Edge 1040 cycle computer (Garmin Ltd., Olathe, Kansas, USA). Subjective fatigue was assessed via RPE.

2.3.3. Maximal sprint test

After a 72-h rest period, participants completed a 6s maximal sprint power test for each crank length in a randomized order. Following a 10-min warm-up at 100–150 W and a 5-min rest, participants performed a seated 6s maximal sprint. Key performance metrics recorded included 6sMaxPower (W/kg), reflecting peak anaerobic power output; 6sAVPower (W/kg), indicating sustained maximal effort; 6sAVRpm (RPM), showing average cadence and pedaling efficiency; and 6sMaxRpm (RPM), representing peak pedaling speed.

Data collected included peak power, average power, and cadence (GARMIN Edge 1040 cycle computer).

2.4. Statistical analysis

Descriptive statistics (mean ± standard deviation) were first calculated to summarize the overall distribution of each outcome. To assess the normality of data distributions, the Shapiro-Wilk test was employed, and Levene’ s test was used to evaluate the homogeneity of variances across groups. To examine the effects of crank length on cycling performance, when the data were normally distributed with homogeneous variances, one-way ANOVA was used. The factor was group of crank length.

If significant differences were detected, Games-Howell post-hoc tests were applied to determine where the significance was. Effect sizes were calculated using Cohen's d, with small (0.2 ≤ d < 0.5), medium (0.5 ≤ d < 0.8), and large (d ≥ 0.8) classifications.²⁵ When the data were not normally distributed, the Kruskal-Wallis H test was used, and Mann-Whitney U post-hoc tests were used to examine where the significance was. The Cliff's Delta was used to measure the effect size, with small (0.147 ≤ δ < 0.33), medium (0.33 ≤ δ < 0.474), and large (δ ≥ 0.474) effect size categories.¹⁷

For repeated measures data (i.e., cycling economy, perceived fatigue, and sprint power), Friedman tests were used to assess differences across the different crank lengths. Pairwise comparisons were conducted for every two crank lengths, and percentage differences were shown to further investigate the influence of crank length on cycling performance. The significance level for all statistical tests was set at p < 0.05.

3. Result

All participants completed the study, and all data were included in the analysis. Table 1 shows the descriptive statistics of participants' basic information, baseline tests, and cycling performance for each crank length (165 mm, 170 mm, 175 mm). Levene's test indicated that all variables met the assumption of homogeneity of variances (F = 0.005 to 1.84, p > 0.166). The Shapiro-Wilk test showed that 60 %VO_2MAX, 6sMaxPower, 6sAVPower, and 6sAVRpm were normally distributed (p > 0.061), while CE, 60 %VO_2MAXAVHeart, 60 %VO_2MAXRPE, and 6sMaxRpm were not normally distributed (p < 0.014).

Table 1.

Descriptive statistical analysis results of cycling test AV(SD).

Variable	165	170	175
Age (Years)	28.00(5.49)
Height (cm)	177.48(5.82)
Weight (kg)	69.64(8.38)
VO2Max (ml/kg/min)	59.39(5.75)
VO2MAXPower (W/kg)	5.15(0.52)
60 %VO2MAX (ml/kg/min)	42.64(4.72)	42.54(4.77)	43.29(6.19)
CE (ml/kg/min/W)	0.20(0.03)	0.20(0.03)	0.20(0.04)
60 %VO2MAXAVHeart (bpm)	146.70(10.69)	147.11(10.52)	148.68(10.84)
60 %VO2MAXRPE (scale 1–10)	3.21(0.88)	3.46(0.69)	4.50(0.69)
6sMaxPower (W/kg)	13.23(1.90)	13.17(1.95)	13.18(2.13)
6sAVPower (W/kg)	9.46(1.60)	9.54(1.58)	9.22(1.35)
6sAVRpm	117.96(10.60)	116.00(8.84)	115.29(12.45)
6sMaxRpm	149.25(7.08)	148.75(7.99)	147.25(11.89)

Note: All power-related metrics are divided by the participant's body weight for standardization.

3.1. Short-distance sprint

One-way ANOVA showed no significant differences in 6sMaxPower, 6sAVPower, and 6sAVRpm between different crank lengths (F = 0.06 to 1.1, p > 0.34). Kruskal-Wallis H test showed no significant effect of crank length on 6sMaxRpm (p = 0.987, δ = 0).

3.2. Cycling economy

One-way ANOVA showed no significant effect of crank length on 60 %VO_2MAX (F = 1.03, p = 0.36). Kruskal-Wallis H test showed no significant effect of crank length on 60 %VO_2MAXAVHeart and cycling economy (CE) (p > 0.67, δ = 0 to 0.01).

3.3. Perceived fatigue

Kruskal-Wallis H test showed that crank length significantly affected 60 %VO_2MAXRPE (p < 0.01, δ = 0.37). Post hoc Mann-Whitney U tests revealed significant differences between 165 mm and 175 mm (p < 0.01, δ = −0.72) and 175 mm and 170 mm (p < 0.001, δ = −6.67), but no significant difference between 165 mm and 170 mm (p = 0.25, δ = −0.17). Furthermore, Friedman test for all variables showed no significant differences between crank lengths (p > 0.05).

4. Discussion

This study examined the effects of three crank arm lengths (165 mm, 170 mm, and 175 mm) on cycling performance in high-level amateur cyclists. The results indicate that crank length affects the subjective fatigue level of high-level amateur cyclists during submaximal intensity cycling but has a minimal impact on performance and efficiency. Based on the findings of this study, high-level amateur male road cyclists should use shorter cranks (165 mm or 170 mm) to reduce perceived fatigue compared to 175 mm cranks.

The use of shorter cranks (165 mm; 170 mm) significantly reduced perceived fatigue compared to longer cranks (175 mm), likely due to reduced joint movement and load during high cadence cycling.^26,27 In contrast, while longer cranks provide better leverage for power output at high intensities, they increase muscular load, particularly on the knee and hip joints, leading to higher fatigue.^11,26 Thus, shorter cranks may offer a practical solution for reducing fatigue without compromising performance in high-level cyclists. Despite the relatively extensive experience and training of high-level amateur road cyclists, they may still struggle to fully adapt to the additional muscular demands imposed by longer cranks, resulting in higher subjective fatigue. Future research could explore the differences in the effects of crank length on RPE across various cycling disciplines, such as mountain biking, time trials, and triathlons, providing valuable insights into optimal crank length selection for different cycling contexts.

In this study, no significant effects of crank length were observed on 6s maximal sprint performance. This may be because only crank length was varied, while other components remained constant. Although crank length influences pedaling mechanics and efficiency,^27,28 its impact on overall performance is limited when considered alone. Longer cranks provide greater leverage but also require larger pedal circles and greater propulsion force,²⁹ and this increased torque may offset any power gains.²⁶ Some studies suggest that shorter cranks can reduce fatigue at high cadences by allowing for more precise pedal control, thereby improving cycling efficiency.³⁰ In this study, a decrease in cadence was observed with longer crank lengths, and fatigue also decreased to some extent. However, fatigue can also lead to a decline in pedaling technique.³¹ Thus, the relationship between pedaling technique, mechanical structure, and fatigue levels may interact with each other, and this warrants further investigation in future studies.³² Furthermore, biomechanical studies have shown that changes in crank length greater than 35 mm significantly affect joint angles and cycling performance,¹¹ while the crank length changes in this study (165 mm–175 mm) are within a smaller range. This may explain why the performance in short-distance maximal sprints with different crank lengths varied between high-level amateur road cyclists and professional cyclists.^12,33

Additionally, different crank lengths did not have a significant impact on cycling economy (CE) during submaximal cycling efforts in high-level amateur road cyclists. This finding aligns with previous research, as although extending the crank length alters torque, the minor changes in knee and hip joint range of motion (ROM) were insufficient to influence overall efficiency.⁴ Cycling economy may be closely related to cadence, rider experience, pedaling technique and the range of crank length variation.¹¹ In novice cyclists, using shorter cranks during submaximal cycling has been shown to improve power output and cycling economy.⁵ However, well-trained professional cyclists can quickly adapt to changes in cadence and power output associated with different crank lengths, maintaining stable heart rate performance.³⁴ Since the participants in this study were high-level amateur cyclists, whose weekly aerobic training volume is like that of professional cyclists, changes in crank length within this range may not cause significant additional variations in cardiovascular system responses and cycling economy.

Several limitations should be noted in this pilot study. First, the small sample size limits the generalizability of the findings. Secondly, the lack of strength assessments, detailed anthropometric measurements, and biochemical monitoring restricts the understanding of individual biomechanical factors and physiological responses to different crank lengths. Accurate biochemical and biomechanical monitoring (e.g., electromyography and metabolic analysis) could offer a more detailed understanding of muscle activation patterns and metabolic responses. Despite efforts to blind participants, experienced cyclists may have sensed differences in crank length, which could bias fatigue measures. Future studies should assess participants' subjective perceptions of crank length and explore how these perceptions impact performance outcomes. Furthermore, future research could investigate the effects of longer-term adaptation to different crank lengths and observe the multi-dimensional impacts of sudden crank length changes after prolonged use. It could also examine the differences in the effects of crank length on athletes with different muscle fiber types. This study was conducted with male athletes only. Given the biomechanical and fatigue-related differences between males and females, including variations in femur length,¹⁰ muscle mass, and strength,³⁵ the selection of crank length may differ significantly between genders. Therefore, the findings of this study may have limited applicability to female athletes. Future research should address these limitations by incorporating a broader range of perspectives, more comprehensive strength and anthropometric assessments, a larger and more representative sample, as well as additional observation and intervention protocols. Given the interaction between pedaling technique, mechanical structure, and fatigue levels, future studies could further examine how different populations respond to varying crank lengths under different fatigue states and their effect on pedaling technique.

5. Conclusion

For high-level amateur road male cyclists, using a 165 mm or 170 mm crank reduces subjective fatigue compared to a 175 mm crank, without affecting cycling efficiency or sprint performance.

Author contributions

Jinsong Li and Yunqing Zhang were responsible for the experimental design and the actual execution of the experiments. Liang Zhang contributed to the theoretical preparation in the early stages. Guangtong Ma was responsible for the recruitment and organization of the participants. Shuting Peng, Binbin Wang, and Jiuyuan Wang participated in the implementation of the experiments and the collection of data. Qiushi Wang was responsible for statistical analysis of the results and writing of the manuscript. Gengxin Dong performed the initial proofreading and review of the draft. Junhong Zhou and Dapeng Bao handled the final grammar corrections, content editing, and finalization of the manuscript. Dapeng Bao was also involved in the entire process of the experiment and manuscript preparation, providing continuous supervision and guidance.All authors have read and agreed to the published version of the manuscript.

Statements

All authors collectively make the following declarations.

• (i)All authors agree with the content of the article and approve its submission to the journal;
• (ii)The material contained in the manuscript has not been previously published and is not being concurrently submitted elsewhere;
• (iii)The experiments reported in the article were undertaken in compliance with the current laws of the country in which the experiments were performed. Authors will be held responsible for any false statements.

Data availability statement

The raw data supporting the conclusion of this article will be made available by the authors without undue reservation.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.