MyVeloFit alternative: Pedalyze vs MyVeloFit, compared on facts

The table compares what each product says on its own site, plus what Pedalyze does in its code. The MyVeloFit facts come from its pricing, FAQ, privacy and bike-fitting pages, opened on 30 September 2026. Each price is in the currency the product charges in.

Pedalyze has two input modes. Photo mode takes one side-on photo with the crank at 6 o'clock, and an optional second photo at 3 o'clock for the KOPS offset. Video mode takes a clip of about 10 seconds on a trainer, and accepts up to 30 seconds. The report is ready in under a minute.

MyVeloFit takes video only: about 15 seconds of side-on footage on a trainer or a stationary bike, with your whole body in the frame. There is no photo mode.

The two inputs answer different questions. A photo gives the angles at one crank position, so the crank must be placed exactly. A clip gives the angles all the way round the stroke, plus cadence and ankle movement. A rider rarely reaches full extension while pedalling, so a posed photo and a pedalling clip give different knee angles. Pedalyze judges knee flexion in a clip near full extension, at the 5th percentile of the stroke, so one bad frame cannot move it. The photo guide shows how to set up either capture.

MyVeloFit covers six disciplines, including gravel/CX, MTB and spin. Pedalyze has three fit styles: road/endurance, comfort/upright and TT/triathlon. If you ride gravel or MTB, Pedalyze has no profile made for you. MyVeloFit does.

Pedalyze shows every number it measured. Each angle sits next to its target range and its tolerated band, with a verdict: in range, borderline or out of range. A borderline knee reading comes with a saddle change of 5–8 mm. A reading out of range comes with 10–15 mm, then 5 mm steps, and a re-check after each one. Video reports add a Fit Score, cadence and the range of motion of each joint. The ranges are plain data in an open file in our source code, so you can read the numbers you are judged against.

The road ranges, at 6 o'clock unless stated:

Only knee flexion has one clear published source. Holmes, Pruitt and Whalen recommend 25–30° at the bottom of the stroke, and we use 25–35° because a 2D photo cannot resolve a few degrees of ankle movement. The other ranges are our own reading of published road, comfort and TT guidance. We are not certified fitters, and we are not affiliated with any fitting school.

MyVeloFit's report gives "key joint angles at three positions in your pedal stroke", joint tracking to find excess movement, and a target fit window set from a mobility assessment and your fit goals. Its site does not name the joints or the crank positions. Triathlete's tester (May 2026) called the advice useful but "a little vague", and wanted a range in millimetres. The mobility test, on the other hand, is a real input that Pedalyze does not have: MyVeloFit sets the target window from your mobility and goals, while Pedalyze uses the same ranges for every rider in a fit style.

A Pedalyze clip never leaves your device. The pose detection runs in your browser, only the measurements are sent, and only if you save the report. Pedalyze photos are different: they are uploaded, measured on our server and stored with your report. An anonymous report that you do not claim is deleted after 7 days.

MyVeloFit uploads your video to its servers. Its privacy policy says that fit information is kept "indefinitely… unless you request its deletion", that staff or contractors may view it to help you, and that it may be used in research and development. That is normal for a product that sells a human review: a fitter cannot review a clip that was never uploaded.

Every angle in a 2D photo assumes a level camera, square to the bike. A tilt or an off-axis camera cannot be corrected later, only detected. So Pedalyze checks every photo before it measures anything. If the camera is more than about 4° off level, or the two hips do not overlap, the photo is rejected with the reason. The video mode runs the same checks on every sampled frame and fails the clip only when half the frames have the problem, so half a second of drift does not cost you a good capture. If your photo is rejected, it is protecting your measurements, not being picky.

We have not tested how MyVeloFit handles a tilted or off-axis video, so this page makes no claim about it.

Prices checked September 2026. Pedalyze charges in euros; MyVeloFit in US dollars.

For context, a full studio fit costs €200–400 in France and £230–450 in the UK, and takes 2–3 hours (prices checked September 2026). Both apps cost a fraction of that. Both free plans give you one upload a week, so you can try each one before you pay.

A comparison that finds no weak point in Pedalyze would be an advert. These are the points where MyVeloFit is ahead today.

• A human in the loop. The Expert Review add-on (from $75, on a paid plan) puts a professional fitter's eyes on your video. Pedalyze has nobody in the loop.
• A mobility test. MyVeloFit sets your target window from your mobility. Pedalyze applies the same ranges to every rider in a fit style.
• More disciplines. Six, including gravel/CX, MTB and spin. Pedalyze has three.
• Bike sizing. "Virtual Fit First" sizes a bike before you buy it. Pedalyze only measures the bike you already have.
• A guarantee. Paid plans carry a 30-Day Better Fit Guarantee. Pedalyze has none.
• A track record. MyVeloFit was founded in 2020 by a certified fitter (IBFI Level 2). Cycling Weekly's tester (June 2023) found that it agreed with a human fitter on saddle height and stack. Pedalyze has no published test yet.

And the points where Pedalyze is ahead, each one checkable in the product or in its code.

• Two crank positions from photos. No other product we checked asks for photos at set crank positions. Pedalyze reads the knee at 6 o'clock and KOPS at 3 o'clock, where those conventions were made.
• A KOPS reading. Given at 3 o'clock as a percentage of shin length, because one photo has no scale in millimetres. It uses the ankle as a stand-in for the pedal spindle and the knee centre instead of the kneecap, so it reads a little to the rear of a plumb line. The KOPS guide says so, next to the range.
• Every range published. Target and tolerated bands for all three fit styles are plain data in an open file. You can check the number you are judged against.
• The clip stays on your device. Video is analysed in the browser and never uploaded.
• Bad captures are rejected. More than about 4° of tilt, or an off-axis camera, gives a rejection with the reason instead of a wrong number.
• One payment, 30 days. €29 once, nothing to cancel. The free plan shows every angle and verdict, not a summary.
• Seven languages. English, German, Spanish, French, Italian, Dutch and Portuguese. MyVeloFit is English only.

No app in this category has published a validation against 3D motion capture or a fitter's goniometer. That includes Pedalyze. MyVeloFit's site says its results are "consistent with some of the top professional fitters" and cites no study. Pedalyze's own error against a goniometer is not measured yet.

What is known is the method. A 2024 study of 26 cyclists found that MediaPipe, the pose model Pedalyze uses, had a typical error of 3–9° against digitised markers, with the foot and ankle tracked worst (Bini, Nascimento and Nibali). That is a result for the model in general, not for Pedalyze with its capture checks. The target ranges are 10–30° wide, so a small error rarely changes the verdict for a reading well inside a range. It can for a reading near an edge. Until Pedalyze's error is measured, compare your own photos with each other, not with a fitter's numbers.

Independent tests of MyVeloFit give a mixed picture, and none of them measured its angles. Cycling Weekly (June 2023): the app and a human fitter agreed on saddle and stack, and the human also found badly placed hoods. Gran Fondo (April 2024): MyVeloFit left the saddle where it was, while two professional fitters moved it forward by 4 and 7 mm.

Pick on what you need, not on an accuracy claim.

• You want a fitter to look at your video, or a mobility test to set your targets: MyVeloFit, with the Expert Review.
• You ride gravel, MTB or a spin bike, or you want sizing before you buy a bike: MyVeloFit.
• You want to see each angle next to its range, and to know which bolt to turn and by how many millimetres: Pedalyze.
• You want a photo mode, or a clip that never leaves your phone: Pedalyze.
• You do not want a subscription, or you read German, Spanish, French, Italian, Dutch or Portuguese: Pedalyze.

If you have pain, an injury or a large left–right difference, go to a certified fitter or a physio. A Pedalyze report has a public link that you can send to the fitter before the session.

Is Pedalyze a free MyVeloFit alternative?

Partly. The free plan gives 1 report every 7 days with every angle, its verdict and the Fit Score, and asks for no card. Unlimited analyses cost €29 once for 30 days. MyVeloFit's free plan also gives 1 upload a week, without the detailed report. If you want a fully free app, BikeFittr and MyBikeFitting are free.

Does Pedalyze need a trainer, like MyVeloFit?

Video mode does: you pedal on a trainer for about 10 seconds. Photo mode only needs the bike steady with the crank at 6 o'clock: a trainer, or a friend who holds the bike while another takes the photo.

Does Pedalyze upload my video?

No. The clip is analysed in your browser and never uploaded. Only the measurements are sent, and only if you save the report. Photos are different: they are uploaded and measured on our server.

Is Pedalyze more accurate than MyVeloFit?

Nobody knows, and this page does not claim it. Neither product has published a validation against a fitter or 3D motion capture. A 2024 study puts the pose model Pedalyze uses at a typical 3–9° error against markers. Pedalyze's own error is not measured yet.

Does Pedalyze have a gravel or MTB profile?

No. Pedalyze has three fit styles: road/endurance, comfort/upright and TT/triathlon. MyVeloFit has profiles for gravel/CX, MTB and spin bikes.

Can I send my Pedalyze report to a fitter?

Yes. Turn on the public link on any report and send the URL. The fitter sees the photos and every angle without an account, and you can turn the link off again at any time.

Can I try both on the same clip?

Yes. Both have a free plan with one upload a week. Record a 15-second side-on clip on a trainer: MyVeloFit asks for about 15 seconds, and Pedalyze accepts up to 30. Then compare what each one says about your saddle height.

Every MyVeloFit fact on this page comes from one of its own pages, opened on 30 September 2026, or from a named review with its date. Prices change; check the pricing page before you buy.

1. MyVeloFit, Pricing

   Plan names, prices, upload limits and the 30-Day Better Fit Guarantee. Seen 30 September 2026.

2. MyVeloFit, Bike Fitting

   The 15-second video, what the report contains, the six disciplines and the "World's #1 Online Bike Fit" claim. Seen 30 September 2026.

3. MyVeloFit, FAQ

   Web app only, the trainer video requirements, and its view on online versus in-person fitting. Seen 30 September 2026.

4. MyVeloFit, Privacy Policy

   Retention of fit information, staff and contractor access, research and development use. Seen 30 September 2026.

5. Fellows S. Is AI tech good enough for bike fitting? Cycling Weekly, June 2023.

   One tester, one MyVeloFit fit and one in-person fit. The two agreed on saddle and stack; the human also found badly placed hoods.

6. Staffa M. The perfect fit? Bike fitting put to the test. Gran Fondo, 25 April 2024.

   Apps and studios on one rider. MyVeloFit left the saddle unchanged; two fitters moved it forward 4 and 7 mm.

7. Stover S. We compared 3 top AI bike fit apps to a professional fit. Triathlete, 27 May 2026.

   One fitter scored three apps on ease of use, clarity and value. It did not compare the apps' angles with his own fit.

8. Bini RR, Nascimento VB, Nibali A. Sport Sciences for Health 20(1), 2024.

   26 cyclists. MediaPipe had a typical error of 3–9° against digitised markers, with poor foot and ankle tracking.

9. Holmes JC, Pruitt AL, Whalen NJ. Lower extremity overuse in bicycling. Clinics in Sports Medicine 13(1), 1994.

   The source of the 25–30° knee flexion recommendation at the bottom of the stroke.

• How to take bike fit photos
• KOPS explained
• Knee angle in bike fitting