Editors Mailbox

Letters to Dr. Edmund R. Burke Ph D., Science Editor

Reproduced from: Cycling Science Spring '96

Response to "Re-examining Optimum Cycle Cadence" by J.R. Coast

Dear Dr. Burke:
I read with interest the recent article entitled "Re-examining optimum cycle cadence" by J. R. Coast. The purpose of this letter is to add to the important points made in the article and to provide challenges for further research. To keep this letter concise, I am oversimplifying the problem description and following with a hypothesis for testing.

SUBJECT SELECTION

It is of paramount importance that experiments with human subjects be carried out as a double blind. Further, cycle cadence should be affected by the percentage of very fast twitch, fast twitch and slow twitch muscle in subjects. It would probably be best to group subjects by the percent fast twitch muscle (e.g., 10, 50, 90 percent fast twitch).

Hypothesis: Cycle cadence varies with percent fast twitch muscle.

MODEL SYSTEM

Model systems fall short of reality. Such is the case when stationary bicycles are used in the laboratory or when bicycles are used on the treadmill. Limiting factors that must be kept in mind include the gyroscopic effect of cycle cadence and that a bicycle is hinged in the front. For instance, if a bicycle were fixed without a hinge, it would not be ridable. Proficient riders keep the bike on a near straight line by constant adjustment of the line through use of legs and upper body, all of which require energy. The faster the cycle cadence, the easier it is to maintain a near straight line. None of that energy to maintain the gyroscopic effect is required on a stationery bicycle. There are two additional factors that occur while riding the bicycle outside, which are not sufficiently simulated by a stationery bike in the laboratory. These include wind and road surface. Both demand higher cycle cadence to maintain efficient power output. The wind, because it's gusting, is partially offset by the gyroscopic effect of cycle cadence. Rough road surfaces result in a dampening effect. Energy is lost through the legs. The latter is partially offset by higher cycle cadence.

Hypothesis 1: As the wind increases, efficient power output increases with higher cycle cadence.
Hypothesis 2: As road surface becomes rougher, with higher cycle cadence, efficient power output increases.

POWER OUTPUT

Power output is easily measured on the stationary bike model system, however, the larger the rider and the greater the percent fast twitch muscle, the greater the power output potential. Thus, riders must be selected carefully for similar power output characteristics.

Hypothesis 1: Power output varies with height and weight.
Hypothesis 2: Road speed can be the same even though power output is different. Hypothesis 3: Power output varies with percent fast twitch muscle.

I look forward to future articles on technical aspects of cycling.

Sincerely,
George C. Martin

Response letter from: J. Richard Coast

Dear Dr. Burke:
Thank you for allowing me the chance to respond to some of the comments of Mr. George Martin. Based on some of the comments, it is obvious that he and I have thought about the problem of optimal cycling cadence from different perspectives. Discussion of these questions from people of varied backgrounds is precisely what I had hoped to accomplish with the article: "Re-examining Optimal Cycling Cadence" (Coast, 1994). I will order my responses as he has ordered his comments.

SUBJECT SELECTION

While blinding experiments is highly advisable; when assessing a variable such as cadence, having a double-blind study is neither feasible nor possible. It is easily apparent to both the subject and the investigator that the object being investigated is cadence, since pedal rate is varied between tests. Further, the dependent variables being measured (e.g. heart rate VO2, blood lactate, etc) tend to respond to the power output and less to stimuli that could be altered by blinding the experiment, if blinding was possible.

The idea of testing subjects based on their muscle fiber composition is an interesting one, and one which some investigators have attempted, with varying results. Recently, Coyle (1995) discussed the effect of percent Type I (slow twitch) fiber on cycling efficiency. Their studies found that cycling efficiency at 80 RPM was correlated to the percent Type I fibers in top flight cyclist with 3776% Type I fibers. They also argued that cyclists with high percent Type I fibers had the highest performance. Therefore, in terms of cycling where high efficiency is important (aerobic events), the slow twitch people seemed to have an advantage. For this reason, even if some cyclists are actually more higher percent of fast twitch fibers, they will not be likely to be successful in road cycling, where prolonged events requiring a high VO2max and high lactose threshold are the rule. Track cycling, which was not addressed in this article (and has not been studied to nearly the same extent), may pose a whole different set of criteria in optimization and a higher optimal cadence.

MODEL SYSTEM

It is agreed completely that a cycle ergometer, bike on rollers, or bicycle on a treadmill are not perfect simulations of cycling on the road. However, it is not agreed that a higher cadence drastically improves the gyroscopic action of the bicycle. Most of the gyroscopic action on a bicycle is due to the rotation of the wheels (at 25 mph, a 27" wheel will rotate at about 300 rpm). While changing from a cadence of 60 or 80 rpm to 100 rpm may improve this a little, the increase in the cost associated with pedalling faster should outweigh the decrease in the cost of holding a straight line.

Wind and rough surfaces increase the cost of cycling. Work by Kyle (1996), Davies (1980), and many others have addressed this topic, with always the same results. The assumption that these conditions necessitate an increase in cycling cadence is untested, however. Based on our work (Coast and Welch, 1985), it would be assumed that the increase in power output needed to overcome these variables would result in an increase in the optimal cadence of few rpm. However, to my knowledge, these studies have not been performed specifically to deal with wind and rough surfaces and may need to be done.

POWER OUTPUT

It is not agreed that subjects need to be selected based on power output characteristics. If one selects subjects of comparable skills (e.g. USCF Cat III), then one is better simulating field conditions, where people of different heights, together. Additionally, prolonged power output is related to much more than maximal power output. Maximal sustained power output is related to the subjects VO2max, lactate threshold, and training state, where maximal short term power output is much more related to body size. In the field, many things such as terrain, wind, and pack vs single rider events will aid larger or smaller people, who will have much different maximal power output values. Therefore, selecting subjects based on power output characteristics may not add much to the knowledge of cycling optimization.

I hope this note has helped clarify some of the differences and agreements we have on this topic. Exchanges such as this are enlightening and help generate ideas for new studies, which is really what this article was meant to do.

Sincerely,
J. Richard Coast

REFERENCES
Coast, J.R. Re-Examining optimal cycling cadence. Cycling Science, 6(1): lh18, 1994.

Coast, J.R. and H.G. Welch, Linear increase in optimal pedal rate with increased power output in cycle ergometry. European Journal of Applied Physiology, 53:339-342. 1985.

Coyle, E.F. Integration of physiological factors determining endurance performance ability Exercise and Sport Science Reviews, 23:25-63, 1995.

Davies, C.T.M. Effect of air resistance on the metabolic cost and performance of cycling. European Journal of Applied Physiology, 45:245-254, 1980.

Kyle, C.R. Selecting cycling equipment. pp 1-44, In: Burke, E.R. (ed) High Tech Cycling, Human Kinetics Publ. Champaign IL. 1996