Is It Illegal to Go 5 Miles Over the Speed Limit?

Three Legal Frameworks for a 5 MPH Overage

Speed limits in the United States are set locally under guidance from the Federal Highway Administration’s Manual on Uniform Traffic Control Devices, which recommends anchoring the posted number to the 85th percentile speed— the speed at or below which 85 percent of vehicles already travel in free-flowing conditions. The logic is that most drivers intuitively select a speed that matches the road’s design, so a driver running 5 over an artificially low posted number may simply be driving the road’s naturally engineered operating speed.^[1]

But how that posted number gets enforced depends on which of three statutory doctrines the state uses. An absolute speed limit makes any excess over the posted number illegal per se — the state only has to prove the number on the device exceeded the number on the sign, with no argument about whether the speed was actually safe. A prima facie speed limit — Latin for “on its face” — presumes a speed above the posted number is unlawful, but lets the driver rebut that presumption in court by showing the speed was safe and reasonable for the conditions at the time. And the Basic Speed Law, present in every state as a backstop, ignores the posted number entirely and asks only whether the speed was reasonable and prudent for the weather, traffic, and road surface actually present.

Texas runs entirely on the prima facie model: its Transportation Code states a speed above the posted limit is “prima facie evidence that the speed is not reasonable and prudent and that the speed is unlawful,” which a driver can contest with evidence of clear weather, light traffic, and a dry, mechanically sound vehicle.^[2] California blends the two: freeway maximums are absolute, but its Basic Speed Law under Vehicle Code 22350 makes the posted limit a ceiling, not a safe harbor — a driver can be cited for 10 mph under the limit in dense fog, and can sometimes argue a few mph over a residential prima facie limit was reasonable if no pedestrians or hazards were present.^[3]

This is the same doctrine explored in our research on driving under the speed limit— the Basic Speed Law cuts in both directions. It can excuse a driver going slightly over the posted number when conditions genuinely support it, and it can penalize a driver going well under the posted number when conditions don’t.

The Speedometer’s Built-In Cushion

Before an officer’s equipment ever gets involved, the instrument a driver relies on to hold the limit is already permitted to be wrong. The Society of Automotive Engineers publishes the calibration standard the auto industry follows: legacy standard SAE J1226 allowed electric speedometers to vary by up to 4 percent across their whole operating range, and its successor, SAE J2976, sets a tolerance of -1.0 percent to +4.0 percent for modern electronic systems that calculate speed from wheel-sensor data.^[4] SAE guidelines also let manufacturers choose a “high bias” calibration philosophy — intentionally setting the instrument to read up to 6 mph faster than true ground speed, never slower — specifically to keep drivers from underestimating their speed.^[4]

That baseline tolerance degrades further as a car ages. A vehicle’s computer estimates road speed from how fast the wheels are spinning, using a fixed assumption about tire circumference. As tread wears down over tens of thousands of miles, the tire’s effective circumference shrinks, the wheel must spin faster to cover the same ground distance, and the computer — unaware the tire has changed — reports a higher speed than the car is actually making. Under-inflation, temperature swings, and non-factory tire or rim sizes distort the same rotational math in the same direction.

For vehicles built to European specifications, the bias is written directly into law. UNECE Regulation No. 39prohibits a speedometer from ever reading lower than a vehicle’s true speed, and permits it to read up to 10 percent plus 4 km/h — roughly 2.5 mph — high.^[5] A European-manufactured car holding a dash reading of 70 mph in a 65 zone can, under that built-in bias, actually be traveling closer to 62-65 mph — inside the limit the entire time.

Passenger vehicles carry no federal accuracy mandate at all: the Federal Motor Vehicle Safety Standards only require that a car have a speedometer that displays in mph. Commercial trucks and buses are different. Under 49 CFR 393.82, the Federal Motor Carrier Safety Administration requires every bus, truck, and truck-tractor under its jurisdiction to carry a speedometer accurate to within ±8 km/h (5 mph) at a cruising speed of 50 mph.^[6] The agency then acted on that same standard in its own enforcement scoring: in 2012, FMCSA removed all 1-to-5-mph speeding violations from the Safety Measurement System used to score motor carriers, explicitly stating the change was needed “to align speeding violations to be consistent with current speedometer regulations (49 CFR 393.82) that require speedometers to be accurate within 5 mph.”^[7] The federal government, in other words, has already put its own name on the position that a 5-mph reading can fall entirely within ordinary mechanical error.

Even post-crash forensic data carries the same limitation. Federal rules governing a vehicle’s Event Data Recorder — the “black box” — require its recorded “Speed, Vehicle Indicated” value to be accurate within roughly ±1 km/h, but only relative to what the speedometer itself displayed, not relative to true ground speed.^[8] Whatever bias already exists in the dashboard reading is preserved faithfully in the black box, which means the “box” investigators pull after a crash inherits the same SAE- and UNECE-permitted variance as the gauge the driver was watching.

How Officers Measure That 5 MPH — And Where the Error Creeps In

The equipment on the other side of the stop carries its own accepted margin of error. Traffic RADAR measures the Doppler shift of radio waves bounced off a moving vehicle and is typically certified with a baseline error of about ±1 to 2 mph.^[9] RADAR also suffers from beam spread: the transmitted beam widens at roughly 12 to 16 degrees, so at 160 feet it can span about 40 feet — wide enough to cover two lanes. If a passenger car at the limit is next to a truck running 10 over, the wider, more reflective truck can pull the reading onto the wrong vehicle. A separate “cosine effect” occurs when the target isn’t moving directly toward or away from the antenna, though that error always reads the target’s speed as lower than true, never higher.^[9]

LIDAR narrows the beam to a laser pulse, letting an officer pinpoint one vehicle in dense traffic, but retains its own accepted variance. In a 2010 Ohio case reviewing the LTI TruCAM laser device, an appellate court found expert testimony establishing a scientifically reliable margin of -2 to +1 mph for that device.^[10] LIDAR also depends on the officer holding the beam perfectly steady; if the narrow beam slips across the angled side of a target vehicle during the reading, the device can interpret that slip as rapid distance change and report an inflated speed.

Older time-and-distance systems carry the largest error. VASCAR calculates average speed by having an officer manually start and stop a timer as a vehicle passes two fixed points, and NHTSA testing found human reaction delays of 0.2 to 0.3 seconds per toggle prevented VASCAR from reaching its manufacturer-claimed ±1 percent accuracy — researchers concluded a tolerance of only ±2 mph was achievable, and only over a measurement window of 4 to 5 seconds.^[11] Pacing — an officer matching a suspected speeder’s speed and reading their own patrol speedometer — adds a further risk: if the patrol car is still closing the gap when the officer checks the gauge, the recorded “catch-up” speed reads higher than the target vehicle’s true speed.

Where 5 Over Is Formally Forgiven — And Where It Isn’t

Several state legislatures translated these mechanical realities directly into statute. Florida Statute § 318.18(3)(b)sets the penalty for exceeding the speed limit by 1 to 5 mph as a “Warning,” carrying a $0 fine — a protection that disappears the moment the infraction happens in a posted school zone, construction zone, toll facility, or enhanced-penalty zone, where a mandatory $50 fine or doubled penalty applies instead.^[12]

Pennsylvania takes a different route to the same outcome. Under 75 Pa.C.S. § 3362, a base fine still applies to any speeding conviction, calculated as $35 plus $2 for every mph over 5 above the limit — but PennDOT’s point schedule assesses zero license points for a violation of 1 to 5 mph over, and only 2 points for 6 to 10 mph over.^{[13] [14]} By neutralizing the points, Pennsylvania avoids the collateral consequences — insurance surcharges, mandatory improvement courses, eventual suspension — that points typically trigger, even though the underlying fine remains payable.

Georgia goes further and restricts local enforcement itself. Under O.C.G.A. § 40-14-8, county, municipal, and campus police may not make a speeding case using a speed-detection device unless the reading exceeds the posted limit by more than 10 mph— meaning a 5-mph overage is entirely immune from local RADAR or LIDAR enforcement. That protection is waived in posted school zones, historic districts, and residential zones under 35 mph, and it doesn’t bind the Georgia State Patrol, which retains full authority to cite any speed over the limit.^[15]

New York represents the opposite extreme. Under New York VTL § 1180, speeding is a strict-liability offense with no numerical grace band: a conviction for 1 to 10 mph over carries a real fine and an immediate 3 license points, and accumulating 6 points triggers a mandatory Driver Responsibility Assessment fee running hundreds of dollars over three years.^[16] Four separate 5-mph tickets in New York, each carrying 3 points, is enough on its own to reach the 11-point threshold that suspends a license.

Georgia’s anti-speed-trap law is the same statute discussed in our research on whether a police officer can hide with no lights on— it’s part of a broader set of restrictions the legislature placed on how, where, and against whom local police may deploy speed-detection equipment.

Automated Cameras Set the Bar Higher, Not Lower

Speed Safety Cameras remove the human officer from the equation entirely, capturing vehicle telemetry and issuing citations automatically from fixed, mobile, or point-to-point installations.^[17] That automation might suggest cameras would be more willing to flag a marginal 5-mph reading than a human officer weighing discretion on the roadside. In practice, the opposite is true: because an automated citation has no officer testimony behind it, most jurisdictions deliberately set the camera’s trigger threshold well above 5 mph — commonly at 10 or 11 mph over the posted limit — specifically so the captured violation is unambiguous enough to survive a legal challenge without a human witness.^[17]

Fighting a 5-MPH Citation: What Actually Works in Court

When a 5-mph citation is contested, the underlying science of RADAR and LIDAR is no longer up for debate. Courts routinely take “judicial notice” of the scientific reliability of these device types — accepting the technology as valid without requiring expert testimony to re-prove it in every case — which is exactly what the Ohio appellate court did for the LIDAR device at issue in Columbus v. Kambi.^[10] That forecloses arguing the technology itself is unreliable; the fight moves entirely to whether that specific unit was accurate on that specific day.

To convict, the prosecution has to establish three things: that the device type is scientifically reliable (already covered by judicial notice), that the particular unit used was properly calibrated at the time of the stop, and that the officer operated it correctly. The second point is where citations most often fail — the state must produce a calibration record, typically showing the officer tested the RADAR unit with certified tuning forks both before and after the shift. A missing certificate, an expired tuning-fork certification, or reliance on the device’s internal self-test alone can make the reading legally inadmissible.^[9]

The “stack-up” defense combines both layers of error covered above. If a vehicle’s own SAE-permitted speedometer variance runs a few mph in one direction, and the citing RADAR or LIDAR unit carries its own certified variance of 1 to 2 mph, a citation for precisely 5 mph over can fall entirely inside the combined range of mechanical uncertainty between the two instruments — making it difficult for a prosecutor to prove, beyond a reasonable doubt, that the car was genuinely exceeding the limit rather than sitting inside overlapping equipment tolerances.