In the sport of cycling, humans and machines must combine as seamlessly as possible for optimal performance. It is important to obtain a proper fit between the cyclist and his or her bike in order for the cyclist to perform optimally and to decrease the risk of developing an overuse injury. Cyclists, as with any other group of athletes, are obsessed with finding ways in which to improve their performance. Saddle height is 1 aspect of bike setup that can affect both performance and injury (5,7,13,14,17,19). Obtaining the proper saddle height is also important when conducting tests in a laboratory situation. Tests involving a cycle ergometer are commonly conducted within a laboratory or health and fitness setting by using subjects who are noncyclists. Training books on cycling generally recommend 4 different methods for determining saddle height (3,4,8,9). While there are various methods recommended in common literature, only 2 methods have been examined and published in peer-reviewed journals (5,7,13,17,19).
The Hamley method examined saddle height based on a percentage of inseam (5). Hamley and Thomas examined saddle height and determined that 109% of inseam produced optimal performance. In this method, inseam was measured from the floor to the ischium and multiplied by 109%. The resultant number was then used to measure from the pedal axle to the top of the saddle, with the pedal at its most distal location. Other researchers have validated these findings by using percentage of inseam to set saddle height (13,19).
Holmes et al. (7) recommended using a knee angle set between 25° and 35° for injury prevention. Because cycling is a repetitive motion occurring within a relatively fixed position, overuse injuries are common. A cyclist riding at 90 rpm will pedal 16,200 revolutions within 3 hours. A saddle height that is set incorrectly could place a large amount of strain on the knee and lead to an overuse injury (7,11). A saddle height that is set too low can result in anterior knee pain due to increased compression in the knee joint through the top of the pedal stroke and during the follow through to the bottom. A saddle height that is set too high can lead to posterior knee pain due to overextension of the knee at the bottom of the stroke.
The Hamley method is recommended for optimal performance and the Holmes method is recommended for injury prevention, but previous research has demonstrated that these methods do not produce the same saddle heights (14,17). In 2 previous studies, the author compared methods of setting saddle height and found that using 109% of inseam fell outside the recommended 25° to 35° knee angle 63% of the time in 1 study and 45% of the time in the other (14,17).
Because these 2 methods produce different saddle heights, which method should be used? Should the Hamley method be used for optimal performance, or should the Holmes method be used for injury prevention? In a previous study, the author examined the difference in anaerobic power between these 2 methods (17). In this study, subjects fell outside the recommended 25° to 35° knee angle 45% of the time and experienced a significant decrease in performance. This study recommended the use of a 25° to 35° knee angle, staying closer to a 25° knee angle, for both performance and injury prevention. However, this study examined the effect of saddle height on anaerobic power only. The effect on aerobic power between these 2 methods has yet to be studied. The purpose of this study was to examine the effect of saddle height on aerobic power by using the Holmes and Hamley methods. If it can be determined that a 25° to 35° knee angle is more economical than, or as economical as, using 109% of inseam, the Holmes method can be recommended for both injury prevention and increased performance.
Experimental Approach to the Problem
Cycling is an endurance sport that requires aerobic power, and the best measure of aerobic power is oxygen consumption (2). The purpose of this study was to compare methods of setting saddle height in order to determine which would be most economical. To accomplish this, subjects cycled at a given intensity for 15 minutes at each saddle position. If a position is more economical, it will require less work and therefore less oxygen from the subject at a given workload and result in a lower o2 for that period. This methodology has been utilized in previous research and has been able to detect changes in economy while cycling (15). In a previous study, the author examined the energy cost of riding in an upright or aero position during cycling. A difference was detected between those who regularly train in aero bars and those who do not.
In the current study, 3 saddle heights were chosen for comparison: a 25° knee angle, a 35° knee angle, and 109% of inseam. Because a 25° to 35° knee angle is a very wide range, the extremes were chosen for comparison.
A total of 15 subjects participated in this study. Descriptive data on all subjects are shown in Table 1. The subjects consisted of 5 cyclists (all men) and 10 noncyclists (2 men and 8 women).The cycling group's abilities ranged from recreational to trained. This is evident by their o2max data, which ranged from 49 mL·kg−1·min−1 to 63 mL·kg−1·min−1 on a cycle ergometer. Noncyclists were chosen for this study because a cycle ergometer is a common form of testing in most laboratories and health and fitness settings. All noncyclists were students from the Department of Health and Kinesiology at Mississippi University for Women. While none of these subjects were cyclists, they all participated in some form of physical activity ranging from running to resistance training. Their o2max data are representative of average college-aged students.
Before participating in this study, all subjects filled out an informed consent form in accordance with the policies of the Committee for Use of Human Subjects in Experimentation at Mississippi University for Women. Each subject also filled out a physical activity readiness questionnaire (PAR-Q) and a health status questionnaire (HSQ) in order to assess health status. Any answers on the PAR-Q or HSQ that indicated a possible health risk excluded the individual from participating in this study.
Cyclists were asked to wear their normal cycling attire for testing and to bring in their personal clipless pedals and cycling shoes. Their pedals were attached to the cycle ergometer in order to better simulate normal riding conditions. Noncyclists wore normal workout clothes and used the caged pedals provided with the ergometer. To ensure that there were no changes in sole thickness, noncyclists were asked to wear the same shoes during each trial.
For optimal performance, subjects were asked to report to the laboratory well rested and well hydrated for all testing. Subjects reported to the laboratory on 4 separate occasions, with at least 1 day of rest between sessions. On all 4 occasions, oxygen consumption was measured by using automated indirect calorimetry (TrueOne 2400; ParvoMedics, Sandy, UT). The metabolic system was calibrated before every test to ensure correct measurement.
On the first occasion, subjects performed a graded exercise protocol on a cycle ergometer (Monark 894E; Monark Exercise AB, Vansbro, Sweden) in order to determine o2max. The resistance began at 1 kp and increased by 0.5 kilopond (kp) every 2 minutes until exhaustion. Noncyclists were required to keep their pedaling cadence at 50 rpm, and the cyclists were required to keep their pedaling cadence at 90 rpm. The variation in cadence between the 2 groups is attributable to differences in training. Because cyclists train and race at a pedaling cadence of 90 rpm or higher, it is recommended that they are tested at that cadence (10). Because noncyclists do not train at high cadences, protocols for noncyclists use a pedaling cadence of 50 rpm. Because this study was a within-subject design and the 2 groups were never compared to one another, the differences in cadence did not affect the comparisons in this study. All subjects were encouraged to continue until they could not maintain a cadence above what was recommended and they reached volitional exhaustion. Markers used to determine o2max were a HR equal to or greater than the age-predicted maximum, achievement of a respiratory exchange ratio (RER) greater than 1.15, or the subject's reaching of a plateau (12).
Intensity for the remaining 3 tests was set by determining the resistance at which the subject reached 70% of o2max during the graded exercise protocol. Throughout the remaining tests, the subjects cycled for 15 minutes at a specific saddle height. On 1 occasion, the saddle height was set by using a 25° knee angle, on another by using a 35° knee angle, and on another by using 109% of inseam. The last 3 trials were counterbalanced to protect against the possibility of an order effect. o2, HR, and RPE were recorded every minute and later averaged for comparison between methods of setting saddle height. Subjects warmed up for 5 minutes before the beginning of every test.
A goniometer was used to determine saddle height by using a 25° and 35° knee angle. Subjects were asked to pedal until they were comfortable in the saddle. They were then instructed to stop with the pedal at the bottom of the stroke. The axis of the goniometer was centered on the lateral femoral condyle, with 1 end pointing down toward the lateral malleolus of the ankle and the opposite end pointing upward to the greater trochanter at the hip. All bony landmarks were found by using palpation. Cyclists' feet were locked into the same position each time due to the clipless pedal system. Noncyclists' feet were positioned so that the metatarsophalangeal joint (i.e., ball of the foot) was positioned directly over the pedal spindle. To ensure accuracy, these measurements were taken multiple times.
To set the saddle height by using the Hamley method, the inseam was measured from the floor to the ischium and then multiplied by 109%. The resultant number was used to set the saddle height by measuring from the pedal axle to the top of the saddle. During this process, the pedal was placed at the most distal end, the bottom of the pedal stroke. Once the saddle height was set, the subject's knee angle was then measured by using a goniometer to determine if the saddle height fell within the recommended 25° to 35° knee angle. To ensure accuracy, measurements were taken multiple times.
o2, HR, and RPE were compared between a saddle height set by using a knee angle of 25°, a knee angle of 35°, and 109% of inseam. Comparisons were made within the total group (n = 15) and subgroups [cyclists (n = 5), noncyclists (n = 10), men (n = 7), and women (n = 8)]. In all groups, means were compared by using an analysis of variance and an α priori of 0.05. An interclass correlation coefficient was used to test for reliability.
When comparing the means of the total group (n = 15), a significant difference was found between a 25° and a 35° knee angle (p = 0.000) and a 25° knee angle and 109% of inseam (p = 0.000). o2 at a 25° knee angle was found to be significantly lower when compared to a 35° knee angle and 109% of inseam. There was no significant difference found between a 35° knee angle and 109% of inseam. The reliability correlation was found to be high at an interclass correlation coefficient of 0.977 mean o2 for each saddle height is listed in Table 2.
Data were divided into 4 subgroups and compared for differences. In the cyclist group (n = 5), o2 was found to be significantly lower between a 25° and a 35° knee angle (p = 0.000) and a 25° knee angle and 109% of inseam (p = 0.000). No significant difference was found between a 35° knee angle and 109% of inseam. In the noncyclist group (n = 10), o2 was found to be significantly lower between a 25° and a 35° knee angle (p = 0.000) and between a 25° and 109% of inseam (p = 0.000). No significant difference was found between a 35° knee angle and 109% of inseam. In the group of women (n = 8), o2 was determined to be significantly lower between a 25° and a 35° knee angle (p = 0.000) and between a 25° knee angle and 109% of inseam (p = 0.002). No significant difference was found between a 35° knee angle and 109% of inseam. In the group of men (n = 7), o2 was found to be significantly lower between a 25° and a 35° knee angle (p = 0.000), between a 25° knee angle and 109% of inseam (p = 0.000), and between a 35° knee angle and 109% of inseam (p = 0.023). There were no significant differences found in HR or RPE between any methods used in this study.
For both performance and injury prevention, it is vital that a cyclist's saddle be set correctly. It is also important for a laboratory technician to set an appropriate saddle height when testing cyclists and noncyclists on a cycle ergometer. Previous research has suggested that using 109% of inseam for setting the saddle height produces optimal performance and that a saddle height set by using a 25° to 35° knee angle is optimal for injury prevention (5,7,13,14,19). Recent research has demonstrated that these 2 methods do not produce the same saddle height and that using a 25° to 35° knee angle increases performance when examining anaerobic power (14,17). While previous research has demonstrated increased anaerobic power, economy and aerobic power have yet to be examined. The purpose of this study was to examine the effects of saddle height on cycling economy.
In this study, subjects fell outside the recommended 25° to 35° knee angle 74% of the time when using 109% of inseam. This is similar to the results found in previous studies. In the author's first study, the subjects fell outside the recommended range 63% of the time, and in the author's second study, the subjects fell outside the recommended range 45% of the time (14,17). Of those who fell outside the recommended range in the current study, 45% had a knee angle less than 25° and 55% had a knee angle greater than 35°. A saddle height greater than 35° produces a very low saddle height, and a saddle height that is less than 25° produces a saddle height that is too high. A saddle height that is too high or too low can result in overuse injuries of the knee (3,7,18,19). This study demonstrated that using 109% of inseam produces a wide range of knee angles (i.e., 21-51°). This supports the author's previous research in which the range was 9° to 42° in 1 study and 17° to 54° in the other (14,17).
A possible reason for the wide range of saddle heights produced by using 109% of inseam is that this measure does not take into account individual variations in femur, tibia, and foot lengths. These individual variations cannot be accounted for by inseam alone and result in the wide variations in knee angles recorded in this and previous studies. Nearly all previous studies examined saddle height based on a percentage of inseam only and did not take these variations into account (5,13,19).
This study demonstrated increased performance when comparing a saddle height set by using a 25° knee angle with a saddle height set by using 109% of inseam. This supports the author's previous research, which compared anaerobic power output between varying saddle heights (17). However, unlike the previous research, a difference was also detected when comparing a 25° and a 35° knee angle. It was found that a 25° knee angle produced a significantly lower o2 when compared to a 35° knee angle in all comparison groups. A lower o2 can be directly related to energy savings and increased performance. On average, o2 was 1.50 mL·kg−1·min−1 lower at a 25° knee angle when compared to 109% of inseam and 1.36 mL·kg−1·min−1 lower when compared to a 35° knee angle. Further details on mean differences are shown in Table 3. On the surface, this may not appear to be a dramatic difference, but this small amount will accumulate into increased performance during the course of a race. Economy is extremely important during endurance events, in which races are won and lost by less than a 1% difference in athletes. These findings may also allow for narrowing the recommended range, from a 25° to 35° knee angle, closer to a 25° knee angle. This study demonstrated a much stronger difference in performance than the author's previous study (17).
Previous research has also demonstrated that there is specificity to training position when examining cyclists (6,15,16). In order to determine the saddle height at which the participating cyclists normally trained, knee angle was measured on their personal bikes. All cyclists fell within the recommended 25° to 35° range, with a mean of 26.8° and a range of 25° to 30°. This could explain why there is an increase in performance at a 25° knee angle compared to a 35° knee angle and 109% of inseam when examining cyclists. However, the noncyclists had no previous cycling experience and still demonstrated the same effect.
There was no significant difference in HR found with a 25° knee angle (157 b·min−1), a 35° knee angle (159 b·min−1), or 109% of inseam (157 b·min−1) in this study. While HR and o2 share a linear relationship, daily variations in HR affect this relationship (1). Heart rate can vary daily up to 20% at any given submaximal o2 (1). These tests were conducted in the South, where the heat index is high and individuals have a tendency to become chronically hydrated. Changes in plasma volume, due to hydration levels, could have affected HR. There was also no significant difference in RPE detected in this study. This could be due to the subjective nature of the RPE scale.
There are 3 main findings from this study. First, using 109% of inseam is unreliable and falls outside the recommended 25° to 35° knee angle greater than half the time. Second, using a 25° knee angle is more economical compared to both 109% of inseam and a 35° knee angle. Finally, these first 2 findings are true for both cyclists and noncyclists alike. From these findings, it can be recommended that saddle height be set by using a 25° knee angle for increased performance. The recommended 25° to 35° knee angle provides for a wide range of movement. This study, along with the author's previous study, lends support for moving the recommended range closer to a 25° knee angle for performance. This recommendation also falls within the recommended 25° to 35° range and therefore can also be recommended for injury prevention. While injury prevention was not measured in this study, it has been established by previous research (7).
When working with cyclists or noncyclists in a laboratory setting, a goniometer should be used to establish a 25° knee angle in order to ensure optimal performance for testing. A 25° knee angle is a starting place, and minor adjustments may be necessary due to interindividual variability. Some individuals may pedal heel up, pedal heel down, or “ankle” while pedaling. It may also be necessary to adjust saddle height to accommodate leg length discrepancies or injury.
1. Achten, J and Jeukendrup, AE. Heart rate monitoring: applications and limitations. Sports Med
33: 517-538, 2003.
2. Australian Sports Commission. Physiological Test for Elite Athletes
. Champaign, IL: Human Kinetics, 2000.
3. Baker, A. Smart Cycling: Successful Training and Racing for Riders of All Levels
. New York: Simon & Schuster, 1997.
4. Burke, ER. High-Tech Cycling
. Champaign, IL: Human Kinetics, 2003.
5. Hamley, EJ and Thomas, V. Physiological and postural factors in calibration of the bicycle ergometer. J Physiol
191: 5-56, 1967.
6. Heil, DP, Derrick, TR, and Whitlesey, S. The relationship between preferred and optimal positioning during submaximal cycle ergometry. Eur J Appl Physiol Occup Physiol
75: 160-165, 1997.
7. Holmes, JC, Pruitt, AL, and Whalen, NJ. Lower extremity overuse in bicycling. Clin Sports Med
13: 187-205, 1994.
8. Jeukendrup, AE. High-Performance Cycling
. Champaign, IL: Human Kinetics, 2002.
9. LeMond, G and Gordis, K. Greg LeMond's Complete Book of Bicycling
. New York: Perigee Books, 1987.
10. Lucia, A, Hoyos, J, and Chicharro, JL. Preferred pedaling cadence in professional cycling. Med Sci Sports Exerc
33: 1361-1366, 2001.
11. Mellion, MB. Common cycling injuries management and prevention. Sports Med
11: 52-70, 1991.
12. McArdle, WD, Katch, FI, and Katch, VL. Exercise Physiology
. Baltimore: Williams & Wilkins, 1996.
13. Nordeen-Snyder, KS. The effect of bicycle seat height
variation upon oxygen consumption and lower limb kinematics. Med Sci Sports Exerc
14. Peveler, WW, Bishop, P, Smith, J, Richardson, M, and Whitehorn, E. Comparing methods for setting saddle height in trained cyclists. J Exerc Physiol Online
8: 51-55, 2005.
15. Peveler, WW, Bishop, P, Smith, J, and Richardson, M. Effects of training in an aero position
on metabolic economy. J Exerc Physiol Online
8: 44-50, 2005.
16. Peveler, WW, Bishop, P, Smith, J, and Richardson, M. Effects of training in an aero position
on anaerobic power output. J Exerc Physiol Online
7: 52-56, 2004.
17. Peveler, WW, Pounders, J, and Bishop, P. Effects of saddle height on anaerobic power production in cycling. J Strength Cond Res
21: 1023-1027, 2007.
18. Ross, MJ. Maximum Performance: Sports Medicine for Endurance Athletes
. Boulder, CO: Velo Press, 2003.
19. Shenum, PL and Devries, HA. The effects of saddle height on oxygen consumption during bicycle ergometer work. Med Sci Sports Exerc
8: 119-121, 1976.