Cleat position calculator

The starting point is not your shoe size: it is the metatarsal line, the one joining the first metatarsal head to the fifth. The first metatarsal head falls at around 72 % of the foot length measured from the heel and the fifth at around 63 %, so the line is not perpendicular to the foot but oblique, and its midpoint sits between 10 and 12 mm behind the first metatarsal, depending on how long the foot is. That midpoint is the reference everything else is built on. Until you type in the two distances, they are estimated with those percentages and the result declares it: that is the difference between an anatomical reference and a shoe size.

A setback window is applied to that reference, depending on discipline, on what you prioritise and on the symptom you declare, and it scales with the length of your foot: a 245 mm foot and a 295 mm one cannot share the same figure in millimetres. The windows run from “between the metatarsal line and 8 mm behind it” on the road to “between 6 and 18 mm behind” in triathlon and time trial, and the declared symptom shifts them backwards when the pattern is consistent with forefoot pressure.

Those windows do not come from a study: they are fitting practice, sorted by discipline. What is published is the direction — moving the cleat back relieves the forefoot and unloads the calf — and the fact that the margin is wide. That is why the result is a band, and why fore-aft confidence does not rise above medium however well you measure.

Why the result is a band and not a figure

The systematic review by Husband and colleagues (2024) analysed 47 studies on cycling position and concluded that, unlike saddle height, there are no clear recommendations for cleat position, crank length or Q-factor. That is the real state of the evidence, and a calculator returning an exact figure would be inventing a precision nobody has published.

There is one more reason, which is that the margin appears to be wide. Chartogne and colleagues (2023) moved the cleat 15 mm forward and 15 mm back relative to the first metatarsal head in twelve competitive cyclists: they found no differences in physiological variables, pedalling kinetics, activity of six muscles, perceived exertion or performance in a 10 s sprint. What did change was hip, knee and ankle kinematics, with less knee extension in the forward position. Put precisely: in that protocol — twelve cyclists, submaximal efforts and a short sprint — no significant differences in performance were detected, and joint angles did change. Finding no difference in twelve people does not prove there is none in yours, nor does it say anything about the effect of months of pedalling; and changing angles does not demonstrate fewer injuries either. The study does not establish the optimum individual position.

Where rotation and stance width come from

Rotation is not calculated, it is respected: the tool takes your natural foot angle and places it at the centre of your system’s float travel, so the foot has margin to both sides. That this margin matters is not a workshop opinion: Ruby and Hull (1993) compared four foot-pedal interfaces and found that those allowing rotation significantly reduced axial and varus-valgus moments at the knee, without increasing any other load. Pure mediolateral translation, by contrast, changed nothing.

Stance width is the variable worst estimated from a sheet of paper, and that is why its confidence never rises above “needs checking”. All that is done here is to cross your forefoot width with your foot angle to warn about two specific things: when the heel may rub the crank and when the cleat track runs out and you have to move to pedal axle extenders. Disley and Li (2014) measured four Q-factors between 90 and 180 mm in 24 cyclists and found slightly greater mechanical efficiency with the two narrowest, with a difference of about three tenths of a percentage point and without individualising by hip width. That is far too little to move anyone towards a width their hips are not asking for.

Why the cleat can be moved by hand

The drawing is not an illustration: it is to the scale of your measurements, with the pedal axle at the origin of the piece that moves. Dragging it shifts fore-aft and stance width at once, and the three sliders do the same from the keyboard, one degree and half a millimetre at a time. The cleat rotates with the foot angle, and the faint fan spreading towards the toe is your system’s float — the margin the foot has left on each side before releasing.

Being able to move it is not decoration. The recommended range is wide and within it you have to choose, and a position is understood far better by seeing it land on the foot than by reading “22 mm behind M1”. When what you choose falls outside the range, the axle is drawn dashed and below it says which way you went and what that implies. The tool does not stop you: it tells you where you are.

Where it stops working

• It does not know what your knee does. The measurement that decides stance width and much of the rotation is knee tracking seen from the front, under load and with fatigue. That cannot be deduced from the shape of the foot.
• Home measurements carry error. Palpating a metatarsal head and transferring it to the shoe easily drifts three or four millimetres, and that error goes straight into the result. That is why the band returned is as wide as it is.
• A difference between feet is not a diagnosis. One foot being 8 mm longer does not mean the leg is, or that anything needs compensating. It means it should be looked at while pedalling before any correction is fitted.
• It does not replace wedges or insoles. A forefoot tilted relative to the rearfoot is assessed in examination, not on a sheet of paper, and a badly placed wedge creates the problem it was meant to solve.
• Dragging the cleat on screen is not measuring it on the shoe.The drawing works with the length, the width and the two metatarsal heads you declared, not with your shoe’s actual last or your sole’s actual slot. It is for understanding the position and deciding it; transferring it to the sole is done with the M1 and M5 marks and a ruler.
• The data for each pedal system is approximate. The float declared by colour is reliable; longitudinal and lateral travel depend on the shoe, the plate and the cleat generation. Check yours before taking a “does not reach” warning at face value.