The Argument Window: Where Radar and Camera Stop Agreeing
Bay One runs a Doppler unit and a photometric camera on every full swing. Most of the time they tell the same story. When they do not, that disagreement is usually worth more than either reading on its own.
People assume we run two launch monitors so we can average them into one better number. That is not really it. What two instruments actually buy you is a way of noticing when one of them is guessing, because a single unit cannot tell you it is having a bad afternoon. It only ever reports what it thinks it saw, with the same confidence either way.
Two Different Ways of Watching a Golf Shot
The radar unit sits behind you and emits a signal that reflects off the ball. It reads the shift in that returning signal as the ball moves away, which makes it very good at anything that reveals itself across a stretch of flight — ball speed, launch angle, and how spin holds or decays once the ball is actually travelling.
The camera fires a rapid sequence of exposures around the moment of impact and compares the frames. It never watches the ball fly at all. It reads what the club and ball did in the first fraction of a second, then hands the rest of the job to a flight model. That makes it excellent at strike location, face angle and path — things the radar has to infer from a much shorter, weaker signal off a small, fast-moving clubhead.
Indoors, this split gets sharper rather than softer. The ball travels perhaps six or eight feet before it meets the screen, which is a fraction of the distance a radar unit wants to see. The camera does not care, because its whole measurement happens before the ball has gone anywhere.
What Each One Owns on Our Sheet
Every card that leaves this building is built from a specific division of labour between the two units, and it does not change from session to session:
- Camera leads: strike location on the face, club path, face angle, dynamic loft, attack angle.
- Radar leads: descent angle and peak height, once there is enough flight for the unit to have something to track.
- Both, and we expect agreement: ball speed and launch angle.
- Neither, full stop: carry and total distance indoors. Both units are handing that number to a flight model, not observing it.
What a Disagreement Usually Turns Out to Be
We hold a tolerance on ball speed and on spin. When a strike falls outside it, the strike gets discarded and the reason gets written down rather than quietly dropped. After nine years of doing this in Bay One, the causes cluster into a short list.
An off-centre strike, by a wide margin. When contact happens well toward the heel or toe, gear effect changes the ball's behaviour in a way that stresses both units' inference chains at once. The disagreement is not a fault in either instrument. It is a symptom of the strike itself.
A ball that has lost its markings. The camera reads spin partly from features on the ball. A scuffed reference ball that has taken a full afternoon of strikes gives the camera a weaker signal than a fresh one, and the radar reading starts to pull away from it. This is the single most common cause on a busy Saturday, and it is entirely avoidable, which is why the reference ball gets swapped on a schedule rather than when someone remembers to look at it.
A genuinely unusual delivery. Very steep attack angles, a face that is wide open or shut at contact, or an unusually slow ball speed all push both instruments toward the edge of what they were built to read cleanly. This shows up most in junior sessions and in the slowest swing speeds we see.
Something wrong with a unit. Rarest of the four, and exactly what a paired setup exists to catch. If the same one-sided gap keeps appearing across different golfers and different clubs on the same morning, that is not a strike problem, that is a calibration problem, and it gets looked at before the next session starts.
Why We Show You Both Screens
A shop running a single unit, of either kind, is not doing anything wrong. It is simply reporting inferred numbers as though they carried the same confidence as observed ones, because there is nothing else in the room to check them against. We would rather you watch the two argue occasionally than take either screen's word for a figure it is quietly calculating rather than seeing.
Most sessions, the two units never disagree by enough to matter and this whole conversation stays in the background. When they do split, it is usually the single most informative thing that happened in that block, and we would rather point at it than smooth it over.
Nate Brogan runs Bay One and the driver and iron benches at Dialed In Lab in Eugene, Oregon.
Come Watch Them Disagree
Every full-swing block in Bay One runs both instruments at once, and the build slip notes which figure came from which. Ask about a specific column during your session and we will show the raw readings rather than just the summary.