Fitting the road bicycle: a proportional approach

BY CHARLES HOWE

No single position definitively maximizes power generation for any given individual, which is why there has
never been a controlled, peer-review study demonstrating as much; rather, cycling economy the ratio of
mechanical power output to oxygen uptake is largely determined by the percentage of Type I (slow-twitch)
fibers present within the working muscles, although muscles are adaptive, and will, in a relatively short time,
become marginally more efficient for any position within a given range of movement. Even so, bicycle fit can
clearly be optimized for overall riding comfort and to minimize aerodynamic drag.
The sort of proportional method described here does have its detractors; see, for instance, the excellent site
http://www.cranklength.info (formerly http://www.thankstomycranks.com). In all likelihood, factors such as
flexibility (which can increase over the course of a season, or be lost due to aging, injury, or inactivity), riding
style (e.g., touring, recreation, competition), aerodynamics, and comfort will all modify the guidelines that result
from the following procedure, so keep in mind that they are just that guidelines which should not be taken
rigidly, as any sort of final word, but as a starting point that gets you in the ballpark, subject to further
adjustment, such as from an old-fashioned, low-tech visual inspection by an experienced eye. Whatever you do,
dont make any sudden or drastic changes to your present set-up, rather, evolve your position gradually,
changing each parameter (such as saddle height) one at a time, by no more than 3 mm every two weeks or so.
More dramatic alterations may be made in the off-season.
The baseline: your present bicycle
Begin by carefully measuring, in millimeters, the following frame dimensions. Make sure the top tube is level,
or, for frames with sloping top tubes, that the bicycle is on a level surface and the tires are of identical make,
width, pressure, and condition:

tt
hs
ss
ds

ts

dh

1. ts seat tube length (or frame size center-to-center). From the radial center of the bottom bracket to the
junction of the centers of the seat and top tubes, in a line parallel to the seat tube.
2. tt top tube length. The distance between the center of its junctions with seat and head tubes.
3. ts seat tube angle. As specified by the manufacturer, or drop a plumbline from the top tube at the point
exactly over the center of the bottom bracket. Measure the distance from this point to the junction of the
centers of the seat and top tubes, then divide this by ts, take the inverse sine, and subtract from 90.
4. e stem extension. From the center of the steering tube to the center of the handlebar clamp, in a line
parallel to the ground.
5. c crankarm length. As specified by the manufacturer.
6. hs saddle height. Same starting point as #1, but continued up to the well of the saddle.
7. ss saddle set-back. The perpendicular distance from a plumbline hung over the nose of the saddle to the
center of the bottom bracket.
8. d saddle-handlebar difference. Measure perpendicularly from the ground to the well of the saddle (ds)
and from the ground to the top edge of the handlebars (dh), then subtract the two values.
Measuring the body
The following dimensions should be taken at the same time of day, such as just after rising, since the pull of
gravity will compress the spine and reduce torso length as the day wears on. In socks and cycling shorts, stand on
a hard, even surface, feet about 6-8 inches apart from instep-to-instep (or so that legs are perpendicular to the
ground), knees completely extended, in front of a smooth wall. Tape a sheet of blank paper to the wall at the point
opposite your crotch. Have a partner (the measurer, and someone you trust!) sit on floor to either side of you and
push a 2-foot long 2 4 up between your legs, with the 2 top edge facing up and kept level, and the front edge
butted against the paper on the wall (attach a torpedo level to the top edge and be sure to monitor the levelness of
the board, giving feedback to the measurer). When the board is level and the measurer is exerting enough pressure,
mark the position of the end of the top surface of the board on the paper. The distance from the mark to the floor is
inseam length i. What is sufficient pressure? Enough to compress the chamois liner of the shorts, but not so much
that you are lifted off the floor. Take 3-4 measurements, always being careful to keep the board level, and average
the closest 3 values. Usually, the first one is a bit short and should be discarded.
Next, measure torso length, standing in front of the wall in the same position. Tape a sheet of blank paper to
wall across from your sternal notch (the depression at the top of the sternum, just below the Adams Apple).
Gently insert a straight dowel (about 12 long) in this depression and move forward until the face of the
dowel is flush against the wall. Once again, you need to strap a small level to the dowel, but this time, the
measurer checks it. When its level, mark the lowest point at which the dowel contacts the wall, and label this
measurement n, for strernal notch height. Repeat several times, and average the 3 closest measurements.
Subtract inseam length i from n to obtain torso length t.
Now, take a seat, and with the lower leg exactly vertical and the upper leg horizontal, obtain lower leg length by
measuring from the floor to the knee joint separation, which is located just above the proximal end of the fibula.
Subtract this from inseam length to obtain upper leg length lu.
One last anthropometric data point . . . arm length a. With elbow joint fully extended and arm raised to 45
below horizontal, measure from the acromial (lateral) end of the clavicle (bump on top of shoulder) to the distal
end of the ulna (bone directly below knuckle of pinky finger). Finally, measure lower arm length al by having
the person being measured sit at a table as though they were about to arm-wrestle you. With the arm
perpendicular to the table (i.e., vertical), measure from the table surface to the distal end of the ulna.

a
n

Anthropometric information for male competitive road cyclists.

RIDER

DATA

RATIOS

(mm)

r1

r2

833

1470

637

610

0.433

0.415

870

1440

570

597

0.396

0.415

794

1393

599

556

0.430

0.399

932

1556

624

610

0.401

0.392

861

1456

595

585

0.409

0.402

851

1475

624

572

0.423

0.388

770

1395

625

550

0.448

0.394

896

1480

584

640

0.395

0.432

899

1540

641

635

0.416

0.412

10

850

1452

602

545

0.415

0.375

11

806

1378

572

555

0.415

0.403

12

830

1395

565

533

0.405

0.382

13

895

1507

612

625

0.406

0.415

MEANS
SD

853
46

1457
57

604
26

586
37

0.415
0.016

0.402
0.016

i = inseam; n = height of sternal notch; t = torso (n - i); a = arm;

r1 = t/n; r2 = a/n

Calculating frame dimensions

Record all these values, and keep a careful record of each change in bicycle set-up (date, parameter, amount of
change), then, calculate bike dimensions in mm as follows:
1. seat tube length = i 0.65 or 0.66
2. top tube length = (t + a) 0.485 or 0.49
3. seat tube angle = 1.135{90 - [sin-1(lu/i)]}
4. stem extension = (t + a) 0.105 or 0.11
5. crankarm length = i 0.203; the conventional wisdom used to be to add 2.5 mm for time trials, 5.0 mm
for pure hillclimbs, but from recent research, there appears to be no advantage to this, and longer cranks
may actually decrease comfort in the aero position
6. saddle height = i 0.87-0.88, taking foot length, pedal/cleat stack height, shorts, and saddle firmness
into account; generally, the saddle is moved up and forward for events of shorter duration requiring
higher power outputs, down and back for road races and hilly terrain
7. saddle set-back = lu 0.190; this is, admittedly, the most questionable formula, since saddle length and
shape can significantly affect the fore-aft position. Better, perhaps, is the KOPS (Knee Over Pedal
Spindle) referencing guideline, where the position of the tibial tuberosity (bump just below the kneecap)
is measured in relation to the pedal spindle, with the bike on a level surface. As a variant of this, some
recommend aligning the plumb line with the front edge of the knee cap and the end of the crank arm.
Dr. Andrew Pruitt recommends that with the pedal at bottom dead center (6 oclock), the tibia should be
vertical, and a line drawn from the femoral head in the hip to the center of the knee joint, when extended,
should form a 25-30 angle with the tibia (to obtain this measurement, you need a goniometer, such as is
available from http://chponline.com). He observes that for the vast number of elite cyclists he has
measured, the mode range for this measurement is 30 with a very small standard deviation 1-2.
8. saddle-handlebar difference = a 0.125
Remaining specifications, such as wheelbase, chainstay length, head tube angle/fork rake, and bottom bracket
height should be determined by desired handling characteristics and intended use of the bicycle. For instance, a
touring bicycle will have a relatively longer wheelbase and chainstays, with a shallower head tube angle and/or
less fork rake, all of which make for more stable but slower handling, whereas a racing bike will be just the
opposite, with a shorter wheelbase and steeper head tube angle and/or more fork rake, for more responsiveness.
How does it all look and feel?
Once again it must be emphasized that these formulae should not be taken hard and fast, or substituted for a
visual inspection and perceptual feedback:
What is the overall impression of the how the bike fits?
Is the saddle too high? Is your knee joint too extended at the bottom of the pedal stroke? Do your hips

rock noticeably?
Is the saddle too low? Is your knee joint angle markedly acute?
Do you appear to be bunched up? Is your torso angle markedly upright, even when you are on the

drops? Do you have trouble sustaining your position on the handlebar drops? How near to horizontal is
your back when you are on the drops? If a plumb line were dropped from your nose, does it fall between
the rear edge and the centerline of the handlebars?
4

 Are you too extended, or stretched out? Do you strain at all to reach the brake lever hoods or

handlebar drops, such that you seldom use these positions, and are most always on the tops of the
handlebars? Are your arms locked out when you are riding in the drops? Do you find yourself
frequently riding up on the nose of the saddle?
Is there any pain or discomfort?
Anterior knee pain may indicate a saddle that is too low or too forward, while discomfort posterior to the

knee (e.g., the distal end of the hamstrings) is often caused by a saddle that is too high or too far back.
Forefoot cant affects knee alignment, and therefore knee comfort. Most people tend to ride knocked-kneed,

and require some amount of forefoot varus (canting of the forefoot upward on the inside), while others will
require forefoot valgus (downward canting of the forefoot on inside). Specially-made shims called The
Wedge (http://www.bikefit.com/products.php) can correct for either condition, while the Specialized BG
line of shoes has a very minor amount of forefoot varus, equivalent to one Wedge. Some individuals will
need more correction than this, or different degrees of correction for each foot, and these shoes may be
inappropriate in such cases, as well as those with a neutral forefoot or who are in need of valgus correction.
Lower back pain/discomfort may be caused by too much extension (a position that is too stretched

out), while too little extension can result in neck pain.

Pain in between the shoulder blades may indicate handlebars that are too wide.

On-line resources
http://billbostoncycles.com/bicycle_fit.htm
http://billbostoncycles.com/crank_length.htm
http://web.archive.org/web/20021204220945/www.bicyclesports.com/technical/index.html
http://www.bikefit.com
http://www.bikefitting.com
http://www.bsn.com/Cycling/ergobike.html
http://byrn.org/gtips/cobb_fit.htm
http://cptips.com/crnklth.htm
http://nettally.com/palmk/crankset.html
http://peterwhitecycles.com/fitting.htm
http://www.rivbike.com/article/bike_fit/fit_sizing_position
http://roble.net/marquis/crank.length
http://sheldonbrown.com/frame-sizing.html
http://www.slowtwitch.com/mainheadings/techctr/bikefit.html (time trial-specific)
http://www.machinehead-software.co.uk/cyclist_crank_length_calculator.html
http://cyclemetrics.com/pages/Fitlinks/bike_fit_links.htm
References
Burke, E.R. Proper fit of the bicycle. Clinical Sports Medicine 13(1):1-14, January 1994.
Coyle, E.F., L.S. Sidossis, J.F. Horowitz, and J.D. Beltz. Cycling efficiency is related to percentage of Type I
muscle fibers. Medicine and Science in Sports and Exercise 24(7):782-88, July 1992.
Horowitz, J.F., L.S. Sidossis, and E.F. Coyle. High efficiency of type I muscle fibers improves performance.
International Journal of Sports Medicine 15:152-57, April 1994.
Martin, J.C., and W.W. Spirduso. Determinants of maximal cycling power: crank length, pedaling rate and
pedal speed. European Journal of Applied Physiology 84(5):413-8, May 2001.
McDaniel, J., J.L. Durstine, G.A. Hand, and J.C. Martin. Determinants of metabolic cost during submaximal
cycling. Journal of Applied Physiology 93(3):823-8, September 2002.
Peveler, W.W. Effects of saddle height on economy in cycling. Journal of Strength and Conditioning Research
22(4):1355-9, July2008.
Fit session checklist
1. 2 4 with torpedo level and rubber bands
2. 12 (20 mm) diameter dowel with level attached
3. spool, line, and lead sinker or plumb bob
4. metric tape measure
5. masking tape
6. Sharpie marking pen
7. 2 sheets blank paper, pencil, and cellophane tape
8. fitting sheet and fit article
9. The cyclist should be in riding shorts, socks, and tank-style top.
10. For the visual assessment, the bicycle should be mounted in a
stationary trainer on a level surface.
6