The cosine effect in speed detection is a built-in geometric error: because an officer with a radar or lidar unit is almost never standing directly in your path, the device measures only the portion of your speed pointed along its beam, and the displayed number comes out lower than your true velocity. For stationary radar and all lidar, that error works in your favor. For moving radar, a related version of the same geometry can inflate the reading against you. The size of the error depends on the angle between the beam and your direction of travel, and below about 10 degrees it barely matters.
The Geometry Behind the Error
Radar and lidar only read the component of your speed along the line between the device and your vehicle. If that line sits perfectly parallel to your path, the device sees your full speed. As the angle widens, it sees less. The formula is straightforward: displayed speed equals actual speed multiplied by the cosine of the angle.1National Highway Traffic Safety Administration. Speed-Measuring Devices Specifications – Down-the-Road Radar Module
Federal radar specifications describe the effect as always lowering the Doppler shift frequency “in direct proportion to the cosine of the angle between the direction of travel and a line from the radar device to the target.”1National Highway Traffic Safety Administration. Speed-Measuring Devices Specifications – Down-the-Road Radar Module For a vehicle actually traveling 70 mph, the geometry plays out like this:
- At 5 degrees, the displayed speed is about 69.7 mph.
- At 10 degrees, about 68.9 mph.
- At 15 degrees, about 67.6 mph.
- At 20 degrees, about 65.8 mph.
At 10 degrees you lose roughly 1.5 percent of your true speed. NHTSA training treats anything under 10 degrees as insignificant for enforcement.2National Highway Traffic Safety Administration. Speed-Measuring Device Operator Training At 20 degrees, the 6 percent loss can be enough to shift a reading into a lower penalty bracket. At 90 degrees the cosine is zero, which is why a radar aimed straight across traffic reads nothing at all.
Stationary Radar and Lidar: The Error Favors You
When an officer parks on the shoulder or in a median, the patrol car sits some distance to the side of the travel lane, creating a permanent angle between the beam and your path. You can’t drive through the patrol car, so the angle is never truly zero. The practical question is how big it gets.
NHTSA’s operator training suggests a working rule: set up no more than 10 feet from the edge of the road for every 100 feet of distance to the target.2National Highway Traffic Safety Administration. Speed-Measuring Device Operator Training A 10-foot offset against a vehicle 1,000 feet out produces a trivial angle. The same offset against a vehicle 50 feet away pushes past 10 degrees, and the reading starts dropping noticeably.
The key point for stationary radar is direction: the cosine effect always pulls the reading down. An officer sitting on the shoulder cannot get a reading inflated by cosine geometry alone. Whatever number the device shows, your actual speed was equal to or higher than that.
Lidar is subject to the same geometry but encounters it differently because the beam is extremely narrow. A radar beam on a Ka-band unit can spread up to 15 degrees wide, while a lidar beam at 1,000 feet is only about three feet across.1National Highway Traffic Safety Administration. Speed-Measuring Devices Specifications – Down-the-Road Radar Module3National Highway Traffic Safety Administration. L.I.D.A.R. Participant Manual NHTSA’s lidar performance specifications offer a concrete guideline: the distance to the target should be at least 20 times the unit’s lateral offset from the center of the road. An officer standing 13 feet off the road should be measuring targets at least 260 feet away, which keeps cosine error below 0.1 percent.4National Highway Traffic Safety Administration. LIDAR Speed-Measuring Device Performance Specifications As with radar, NHTSA’s lidar manual sets the practical threshold at roughly 10 degrees. Lidar is only used in stationary mode, so the error always favors the driver.
Moving Radar: When the Error Hurts You
Moving radar is more complicated, and this is where the cosine effect can push a reading higher than your real speed. A patrol car in motion runs the radar on two signals at once: a low-Doppler signal reflected off the road and stationary objects to calculate patrol speed, and a high-Doppler signal from your vehicle to calculate closing speed. The device subtracts patrol speed from closing speed and displays the result as your target speed.
If the radar antenna is slightly off from the patrol car’s direction of travel, the patrol-speed signal bounces off ground objects at an angle. The cosine effect understates the patrol speed. Since target speed equals closing speed minus patrol speed, understating patrol speed inflates target speed. NHTSA’s training manual states this plainly: “The target speed will be high because the R.A.D.A.R.’s reading of the patrol vehicle’s speed is low.” The same manual warns that “misalignment of the antenna, no matter how slight, may increase the possibility of a high target speed reading.”2National Highway Traffic Safety Administration. Speed-Measuring Device Operator Training
Two related errors show up in moving mode and tend to travel with cosine problems. Shadowing happens when a large, slow-moving vehicle ahead of the patrol car dominates the radar’s ground-speed signal. The radar then reads patrol speed from the closing speed between the two vehicles instead of from the roadway, which understates patrol speed and inflates the target reading. Verifying patrol speed against the speedometer during tracking will catch it. Batching happens when the radar updates patrol-speed and target-speed calculations at different intervals during rapid acceleration or deceleration, briefly displaying an inflated target speed. Modern digital units have largely eliminated this, though older analog equipment can still show it.2National Highway Traffic Safety Administration. Speed-Measuring Device Operator Training
What Makes the Angle Bigger in a Real Stop
The angle changes constantly as a vehicle approaches. When a car is far down the road, the angle between the beam and the travel path is tiny and the reading is nearly perfect. As the vehicle closes, the angle grows and the displayed speed drops. Officers are trained to capture readings at longer distances for exactly that reason.
Lateral offset matters more than drivers tend to assume. An officer five feet off the road measuring a vehicle at 500 feet has an angle of about 0.6 degrees, which is meaningless. The same officer measuring a vehicle at 50 feet has an angle approaching 6 degrees. Officers who set up at wide pull-offs or behind guardrails that push them farther from the travel lane build in larger errors from the start.
Beam width is a radar-specific factor. NHTSA specifications cap Ka-band and K-band radar beams at 15 degrees wide at the half-power point.1National Highway Traffic Safety Administration. Speed-Measuring Devices Specifications – Down-the-Road Radar Module A wider beam increases the chance of catching a return from a vehicle in an adjacent lane or at a different angle than the officer intended, and in heavy traffic that ambiguity compounds the cosine problem.
Using the Cosine Effect to Fight a Ticket
Theory meets the courtroom awkwardly here. Simply telling a judge about the cosine effect without supporting evidence rarely works. Courts expect either expert testimony or concrete evidence that the specific geometry of your stop produced a meaningful error.
For a stationary-radar ticket, the math usually cuts against the argument. If the officer was in a standard roadside position, the cosine effect likely trimmed 1 to 2 percent off your real speed, meaning you were actually going faster than the ticket says. The defense has real teeth only when the officer was positioned at an unusually wide angle, or when the recorded speed sits close enough to the limit that even a small correction matters.
Moving radar is where the argument gets stronger. If you can show the patrol car’s antenna was misaligned, or that the officer was taking the reading on a curve, you have a legitimate basis to argue the number was inflated. Requesting the device’s patrol-speed history and comparing it against the patrol car’s speedometer is one way to probe for shadowing or cosine-driven patrol-speed errors. These arguments carry more weight when an expert can walk the court through the geometry of your specific stop.
The practical reality is that most traffic courts treat cosine-effect arguments as background noise unless you tie the physics to the facts of your stop: the officer’s position, the measurement distance, the device type, and whether it was in stationary or moving mode. If your recorded speed was well above the limit, cosine math won’t rescue you, because even a generous correction still leaves you speeding. Where the argument carries real weight is in borderline cases and in moving-radar stops where the potential for an inflated reading is documented in the training materials the officer was taught from.