ALPHA VISION · FOOTBALL
Concept and fusion proposal · not a product

Twenty two players are tracked.
The ball is inferred.

Football has built an extraordinary optical apparatus for knowing where people are. Every player, every second, every match. The ball rides along inside that system as one more tracked point, and its own physics, how fast it actually left the boot and how it was rotating while it flew, is derived rather than measured. That gap is the only thing on this page. We are not proposing to replace anything that already works.

0
Side spin, rev/s
0.0
Deflection at the goal, m
0
Separation after 200 ms, cm
−28 dBsm
Ball reflectivity measured
111 m
Detection range design
01 · The principle

The one ball in sport that has actually been measured

Almost every reflectivity figure in this industry, including our own for a golf ball, is modelled rather than measured. The football is an exception. It is one of only three sports balls with a published measurement at 24 GHz in the peer reviewed literature, which means the range figure on this page rests on somebody else's laboratory rather than on our arithmetic. That is a small thing, and it is the kind of small thing we would rather point out than let you discover.

What is unusually solid

The ball's reflectivity at our frequency is measured and published. Our detection range for it is therefore better grounded than the one on our own golf page.

What radar would add

Strike speed from the Doppler shift rather than from differencing two camera frames, and a flight path measured in any light and any weather with nothing inside the ball.

What video keeps

Who touched it, where everybody stood, what the event was. Optical systems are extraordinary at this and we would fuse with them rather than compete.

The commercial truth

Strike power is not currently a budget line in football the way shot speed is in some other sports. We would be partly creating the demand, not answering an existing tender. You should weigh that.

02 · The strike

Two shots that look identical, until they do not

Take a shot from the edge of the box. Give it side spin and it bends. Take the spin away and it does not. The interesting part is not the ending, it is the beginning: for the first fraction of a second the two balls are in almost exactly the same place, well inside the error of what a camera can resolve at match frame rates. The difference between them is not in their positions yet. It is in their rotation, and rotation is a frequency, which is the kind of thing a Doppler sensor is built to look at. Move the slider.

ONE STRIKE · PLAN VIEW · 25 M FROM GOALsimulated physics
Side spin 6.0 rev/s
the struck ball the same strike with no spin what a camera resolves early on

03 · What we would have to prove

The spin claim is not ours to make yet

Everything above about spin is physics, and physics is not evidence. We have never measured spin on any object, in any sport. The signature exists in theory as a pair of sidebands around the main return, roughly 20 dB below it, and turning that into a number in a stadium is an engineering problem we have not solved and have not tried. This section exists so that nobody has to ask.

Open question 01

Spin, at all

The sideband is far weaker than the ball itself, and it has to be extracted while the ball is decelerating and the sensor is looking through a stadium. Until we have measured it once, on any object, it stays on this page as a roadmap item and nowhere else.

Open question 02

The stadium

Stands, roofs, hoardings and a crowd. The same multipath problem every venue presents, in a bigger box. Calculable in principle, settled only by measurement.

Open question 03

Which ball

The published measurement is for a ball, not for every ball. Panel construction and surface differ between makes and between generations, and how much that moves the number is unknown to us.

What we would claim after a successful trial

Strike speed

Measured directly, in any light and any weather, with nothing inside the ball. That alone is a real thing and it is the part we are confident in. Spin would join the list only after it had been measured, and not before.

What we would not claim, ever

Player tracking

Twenty two players, their positions, their identities and the events between them. Video already does this to a standard we have no intention of competing with, and any proposal we make sits underneath it.

04 · How it would actually work

Underneath the video, not against it

The realistic route into this sport is not a competing tracking system. It is a measurement layer that an existing optical platform can call, at the moment of a strike, to get a number it cannot produce itself.

01
The camera keeps the context

Who struck it, from where, in what phase of play, with which foot. None of that is a radar problem and we would not want it to be.

02
The sensor supplies the physics

Strike speed at the moment of contact, measured rather than differenced. One number, produced well, at the instant the optical system already knows something happened.

03
One sensor, one end

A shot at goal travels towards a sensor placed behind that goal, which is close to ideal Doppler geometry. Coverage of the whole pitch is a different and much larger conversation, and not the one we would start with.

04
Where the value shows up first

Content rather than coaching. A measured hardest strike travels further as a clip than as a coaching metric, and it is the honest first use of a number this sport does not currently buy.

05
What a trial looks like

One sensor behind one goal, at a ground that already has camera tracking, for a handful of matches. Scored against the optical estimate of the same strikes. Small, cheap and conclusive either way.

Why this order and not the other one

We could try to sell a full pitch measurement system. It would be a large installation, a long sale and a direct argument with something that already works. A single sensor behind a goal producing one measured number is a smaller promise, it is testable in an afternoon, and if it fails we will have found out cheaply.

05 · Technology

The numbers, each with its class

Ball reflectivity

−28 dBsm measured

Published in the peer reviewed literature at 24 GHz. One of only three sports balls with a direct measurement at this frequency, and the reason the next figure is better grounded than most on this site.

Detection range

111 m design

To a football at 24 GHz under fixed power, from our link budget with the measured reflectivity above. A pitch is 105 m long, so one sensor behind a goal reaches the far end.

Range resolution

1.5 m design

Set by the chirp bandwidth. How finely the sensor separates two objects in depth, and the reason nobody honest quotes centimetres at this bandwidth.

Clutter gate

2 m/s

Anything moving slower than this along the sensor line disappears into the returns from the ground and the stands. A rolling ball in midfield is invisible. A strike is not, which is why the strike is the product.

The heritage

±0.05 % measured

Certified velocity measurement on objects from 30 to 10,000 m/s. The affiliate instrumentation heritage this team comes out of, not a specification of a football product that does not exist.

What is missing

Everything measured

No accuracy figure from a stadium. No spin measurement on any object, ever. No multipath characterisation. No product. This page describes physics we are confident in and experiments we have not run.

06 · Where this sits

Behind golf, and behind ice hockey

Now

Physics and signal chain

The measurement chain runs on development hardware for our first sport. The same chain would measure a struck football without modification, which is why this page exists at all.

Next

Spin, measured once, on anything

Before spin appears in any proposal we make, it has to be extracted from a real return on a real object. That experiment is small and it gates everything on this page beyond strike speed.

When a ground says yes

One sensor, one goal, a few matches

Scored against the optical estimate of the same strikes, with the raw returns kept so the result can be checked rather than believed.

Only if that works

A fusion conversation

With a platform that already owns the video. Not before, and not around them.

Talk to us

One sensor, one goal, a few matches.

If you run an optical tracking platform, a club with cameras already installed, or a competition that has wondered how fast that strike really was, that is the conversation. Bring the scepticism about spin, because it is ours too.

Talk about a trial

Ole Kjærgaard Schrøter · Alpha Vision · oks@weibelventures.com · +45 20 97 56 20