The dialogue that repeats itself in every cycling chat: “my knee hurt — I raised the saddle — it went away”. Then someone inevitably shows up who had exactly the opposite experience: raised it too far and had to lower it again. And a third person who suggests measuring your inseam and multiplying by 0.883.

All three are partly right, and a study published in September 2025 helps sort it out.

The scale of the problem

The knee is the most frequently affected joint in cycling. According to epidemiological data, between 50 and 90% of cyclists sustain overuse injuries, and anterior knee pain occurs in 40–63% of both amateurs and professionals. Most often it is patellofemoral pain syndrome, patellar tendinopathy, or iliotibial band syndrome.

Saddle height is considered a critical factor precisely because it directly sets the kinematics of the leg and the load on the kneecap.

What the researchers did

The work was published in the Journal of Physical Education and Sport (Chile, University of Viña del Mar). The design was a longitudinal correlational study.

Participants: 38 male amateur cyclists, road and MTB, aged 25–57 (mean 40.3 ± 8.9). Trained people, but not elite: FTP 224.7 ± 38.8 W, that is, 3.3 ± 0.8 W/kg, volume 225 ± 119.9 km per week.

Everyone was split by knee flexion angle at the bottom dead centre of the pedal stroke:

flexion greater than 40° — 21 people;
flexion less than 40° — 17 people.

Kinematics were captured on video with reflective markers and analysed in Kinovea — standard two-dimensional analysis in the sagittal plane. Pain was rated on a visual analogue scale three times: before the fitting, immediately after, and two weeks later.

Results

The difference between the groups before the intervention turned out to be large:

flexion > 40° — pain 4.4 ± 1.9;
flexion < 40° — pain 0.9 ± 1.5;
p < 0.001.

The correlation between flexion angle and pain was r = 0.72 — a strong relationship.

After the saddle was adjusted, angles in the first group came down to the recommended 25–30°, and pain fell:

immediately after the fitting — 2.7 ± 1.4;
two weeks later — 1.4 ± 1.6.

What is more, the correlation between angle and pain disappeared after the fitting — meaning the relationship really was with the angle, and not with something else that depended on it. In the group whose angles were normal to begin with, nothing changed significantly, which makes sense: there was nothing to fix.

Why the angle matters at all

A saddle that is too low means excessive flexion at bottom dead centre and increased compression in the patellofemoral joint. The kneecap is pressed harder against the femur on every revolution — and there are about five thousand revolutions in an hour of riding.

A saddle that is too high means overextension, side-to-side rocking of the pelvis, tension in the hamstrings and discomfort in the saddle.

How sensitive the system is was shown back in a 2011 review in Sports Medicine: changing saddle height by 5% changes knee kinematics by roughly 35%, and joint moments by 16%. That is also where the 25–30° range at bottom dead centre comes from.

What to do with the formulas

The very formulas people break spears over:

0.883 × inseam length — to the centre of the bottom bracket (the LeMond formula);
109% of inseam length — to the pedal surface (Hamley and Thomas).

Do they work? As a starting point — yes. In a 2025 pilot study using 4D scanning, a formula-based fit reduced knee extension angles by 6.3–8.0° compared with the height the cyclists had chosen themselves. In other words, on average intuition is off, and the formula corrects it.

But a formula is an anthropometric average. It does not know your femur length relative to your tibia, your sole thickness, your cleat position, how you sit in the saddle, or your ankle mobility. All of these change the resulting angle at one and the same height on the ruler.

So measure the angle, not your leg. The formula is only there to get you into the right neighbourhood on the first try.

A practical order of operations

Record video. A trainer, filmed strictly from the side at bottom-bracket height, 60 frames per second, your usual cadence and your usual position. The angle is measured at bottom dead centre — between the hip, the knee and the ankle.

Aim for 25–30°. If you come out above 40°, you have a clear reason to work on your fit.

Change no more than 5 mm at a time and ride at the new height for two or three sessions before judging. The body does not adapt to a position instantly.

Read where the pain is. At the front, under the kneecap — typical of a low saddle. At the back, under the knee, plus pelvic rocking — typical of a high one.

Do not adjust height alone. Fore-aft saddle position and cleat position change the effective angle just as much as the seatpost does. Changing everything at once is a guarantee that you will not know what helped.

Being honest about the limitations

There are many limitations here, and they are substantial.

Men only, 38 of them. Women's fit differs in proportions and in pelvic anatomy, and this study says nothing about it.

No control group. Nobody checked what the pain would have looked like two weeks later without any fitting at all. Part of the effect may be the natural course of things.

Pain rating is subjective and unblinded. Participants knew their bike had been adjusted — and rated their own pain. This applies especially to the “immediately after” measurement: expectation is working at full strength there. The two-week result looks more robust.

Two-dimensional analysis. Inward-outward knee movement, which is also linked to pain, is invisible when filming from the side.

In other words, the study is a good confirmation of an already known biomechanical recommendation in real amateur riders, but it is not rigorous proof of causality.

Key points

Overuse injuries affect 50–90% of cyclists; anterior knee pain affects 40–63%.
The 2025 study, 38 amateurs aged 25–57: at knee flexion > 40°, pain was 4.4 ± 1.9 versus 0.9 ± 1.5 at < 40° (p < 0.001), correlation r = 0.72.
After the saddle was adjusted to 25–30°, pain fell to 2.7 immediately and 1.4 two weeks later; the correlation with the angle disappeared.
Changing saddle height by 5% changes knee kinematics by ~35% — the system is very sensitive.
The formulas (0.883 × leg length, 109%) are a starting point, not an answer; what you need to check is the angle on video.
Pain at the front is more likely a low saddle; pain at the back plus pelvic rocking is more likely a high one.
Adjust in steps of no more than 5 mm, and judge after two or three sessions.
Limitations: men only, no control group, subjective and unblinded pain ratings, two-dimensional analysis.

Sources: Martínez M., Pérez M.A., “Biomechanical analysis of knee flexion and saddle height adjustment on knee pain in amateur cyclists”, Journal of Physical Education and Sport, 2025;25(9), Art 217, pp. 1953–1959. https://doi.org/10.7752/jpes.2025.09217. Bini R., Hume P.A., Croft J.L., “Effects of Bicycle Saddle Height on Knee Injury Risk and Cycling Performance”, Sports Medicine, 2011;41(6). https://doi.org/10.2165/11588740-000000000-00000