NHTSA Field Sobriety Test Standards: What Officers Get Wrong

NHTSA's own validation studies found these accuracy rates for field sobriety tests, even when administered correctly: HGN: 77%, Walk-and-Turn: 68%, One-Leg Stand: 65%, all three combined: 82%. That means even under perfect conditions, these tests misidentify sober people as impaired 18% of the time. The one-leg stand alone is wrong more than a third of the time. Officers rarely administer them perfectly.

The officer says you failed the field sobriety tests. You're unsteady. Your eyes were tracking wrong. You couldn't walk a straight line.

But here's a question nobody's asking: did the officer even administer the tests correctly?

NHTSA, the National Highway Traffic Safety Administration, the federal agency that created these tests, has published specific standards for how they must be conducted. Those standards exist because without them, the tests don't mean anything. And officers deviate from those standards constantly.

The 77% Problem

You're standing on the shoulder of a highway. It's 11:47 PM. A semi passes at 65 miles per hour and the wind hits your back. Red and blue lights are strobing in your peripheral vision. Your hands are shaking, not from alcohol, but because an armed stranger is watching you with a clipboard. This is where the "standardized" test happens.

Those accuracy numbers assume perfect administration. Perfect conditions. Perfect scoring. How often do you think that happens on the shoulder of a highway at midnight?

But here's what nobody mentions: the 77% accuracy rate for HGN (NHTSA HS 178 R2/06) is the best-case number from a controlled indoor study. On a highway shoulder with flashing lights, passing traffic, and an officer who's rushing through the protocol, the real accuracy is unknown, because nobody has studied it under those conditions.

The government's own research says these tests are wrong 18% of the time under perfect conditions (NHTSA HS 178 R2/06), and no officer has ever administered them under perfect conditions.

Experienced DUI defense attorneys treat FST administration errors as the foundation of suppression motions. The principle: if the officer didn't follow NHTSA standards, the test results aren't standardized tests at all. They're just an officer's subjective opinion dressed up as science.

Ask your attorney one question: "Have you compared the dashcam footage of my field sobriety tests against the NHTSA manual?" If the answer is no, you now know the single most important thing that hasn't been done in your case.

The Three Standardized Tests

NHTSA only validates three field sobriety tests. Anything else, touching your finger to your nose, reciting the alphabet backward, counting backward from 100, is not standardized, not validated, and not supported by NHTSA research. If an officer used non-standardized tests and cited them as evidence of impairment, that's already a problem.

Test 1: Horizontal Gaze Nystagmus (HGN)

The officer leans close to your face. You can smell coffee on their breath. They hold a pen 12 inches from your eyes and move it slowly to the side. "Follow this with your eyes. Don't move your head." The pen moves. Your eyes follow. You have no idea what they're looking for or what they're writing on the clipboard.

The officer holds a stimulus (pen, finger, penlight) about 12-15 inches from your face and moves it side to side while watching your eyes.

What they're looking for: involuntary jerking of the eye (nystagmus) that becomes more pronounced with alcohol impairment. Six total clues, three per eye.

But here's what nobody mentions: at least 38 medical conditions cause the exact same eye movement the officer is looking for (NHTSA HS 178 R2/06), and officers almost never ask about any of them before starting the test.

At least 38 medical conditions cause the same eye jerking that officers score as "impairment", and they almost never ask about your medical history first.

What officers get wrong:

• Speed of the stimulus. NHTSA says each pass should take approximately 2 seconds. Officers frequently rush through the test, especially when they're standing in traffic at night. Too fast means unreliable scoring.
• Position of the stimulus. It should be held slightly above eye level. Too high forces the subject to tilt their head, which can cause nystagmus from positional factors, not alcohol.
• Distance from the face. 12-15 inches. Closer distorts the tracking. Further makes observation harder. Officers eyeball this distance.
• Medical conditions. At least 38 conditions cause nystagmus unrelated to alcohol: brain injuries, inner ear disorders, certain medications (anti-seizure drugs, antidepressants), eye muscle imbalances, even caffeine and nicotine. NHTSA training mentions this (NHTSA HS 178 R2/06). Officers almost never ask about medical history before the test.
• Environmental factors. Flashing police lights and passing headlights can affect eye tracking. NHTSA's controlled studies were conducted indoors. Your test was conducted on a highway shoulder.

If the officer's dashcam or bodycam captured the HGN administration, an expert can evaluate whether the procedure followed NHTSA protocols. Often, it didn't. For more on using video evidence, read how to read your discovery.

Write down any medications you take, any eye conditions you have, and any head injuries you've experienced. Give this list to your attorney, any one of these could explain the HGN results without alcohol being a factor.

Test 2: Walk-and-Turn

The officer points at the ground. "Walk nine steps, heel to toe, along this line. Turn around. Walk nine steps back." You look down. There is no line. You're standing on gravel at the edge of a highway. You're wearing work boots. The "line" is imaginary.

Nine heel-to-toe steps along a real or imaginary line, turn, nine steps back.

What they're looking for: eight possible clues. Can't keep balance during instructions, starts too soon, stops walking, misses heel-to-toe, steps off line, uses arms for balance, wrong number of steps, improper turn.

But here's what nobody mentions: the test was validated on a flat, dry, hard surface with a visible line. Most roadside administrations happen on gravel, slopes, or wet pavement with no visible line at all. The test is being scored against a standard that doesn't match the conditions.

NHTSA requires a flat, dry, hard surface with a visible line, try walking heel-to-toe on highway gravel in the dark and see how "impaired" you look.

What officers get wrong:

• Surface conditions. NHTSA specifies a "reasonably dry, hard, level, non-slippery surface." The shoulder of a road with gravel, grass, a slope, or uneven pavement is none of those things. You try walking heel-to-toe on a graded highway shoulder in dress shoes.
• The instruction phase. The officer is supposed to demonstrate the walk, demonstrate the turn, and confirm the subject understands before they begin. Abbreviated or unclear instructions mean the subject may "fail" because they didn't understand the task, not because they're impaired.
• Footwear. NHTSA says subjects should be given the option to remove shoes with heels higher than 2 inches. Officers frequently skip this.
• Physical limitations. People over 65, people with back or leg injuries, and people who are significantly overweight can have difficulty with this test completely sober. NHTSA acknowledges this in their training manual but officers routinely administer it anyway.
• Counting clues. The officer scores 2+ clues out of 8 as "failing." But a 1998 NHTSA-funded validation study found that even trained officers frequently miscount clues, score subjective observations as definitive clues, and interpret normal nervousness as impairment indicators.

Think back to the surface where your test was administered. Was it flat? Dry? Level? Was there a visible line to follow? Write down what you remember about the conditions. If any answer is no, the test was not administered per NHTSA standards.

Test 3: One-Leg Stand

The officer says, "Raise one foot six inches off the ground and count: one-thousand-one, one-thousand-two..." You lift your foot. Your leg starts shaking. Not because of alcohol, because you're 47 years old, standing on a slope, wearing dress shoes, and terrified. At thirty-two seconds the officer writes "put foot down" on the clipboard. The form doesn't mention that he let you stand for thirty-two seconds instead of thirty.

Stand on one foot, hold the other foot 6 inches off the ground, count "one-thousand-one, one-thousand-two..." for 30 seconds.

What they're looking for: four clues. Swaying, using arms for balance, hopping, putting the foot down.

But here's what nobody mentions: most officers don't actually time the test. They estimate. If they let you stand for 40 or 50 seconds, even an Olympic athlete might put their foot down. That gets scored as a clue of impairment.

Most officers don't time the one-leg stand with an actual clock, they estimate, and extra seconds turn a sober person's fatigue into "evidence" of impairment.

What officers get wrong:

• The 30-second timing. Many officers don't actually time the test. They estimate. If they let you stand for 40 or 50 seconds, even a sober person may put their foot down. That gets scored as a clue.
• Surface conditions. Same issue as walk-and-turn. NHTSA specs a flat, level surface.
• Physical limitations. People over 65, people with leg or back problems, and people more than 50 pounds overweight are not good candidates for this test regardless of sobriety.
• Footwear again. Try standing on one foot for 30 seconds in heels, boots, or flip-flops. The test assumes flat, stable shoes.
• The stress factor. You've been pulled over, you're scared, there are flashing lights behind you, cars are passing at 60 mph, and an armed officer is staring at you with a clipboard. Stress causes trembling and swaying that have nothing to do with alcohol.

Try standing on one foot for 30 seconds right now, in your current shoes, on whatever surface you're standing on. Notice how your body reacts, the swaying, the arm adjustments, the instinct to put your foot down. Now imagine doing it on gravel at midnight with police lights flashing. That is your defense in one sentence.

The Dashcam Tells the Truth

The officer writes the report three hours after your arrest. From memory. At a desk. The report says you showed 4 of 8 clues on the walk-and-turn. But the dashcam, rolling in real-time from the cruiser's hood, captured something different. The footage shows what actually happened. The report shows what the officer remembers.

Here's where elite DUI defense attorneys build their suppression cases.

Officers write their reports after the arrest. Sometimes hours later. The report says "subject demonstrated 4 of 8 clues on the walk-and-turn test" and lists those clues as definitive findings.

But the dashcam footage, if it captured the tests, shows what actually happened. And frequently, the footage contradicts the report.

But here's what nobody mentions: the dashcam doesn't have an opinion. It doesn't have a quota. It doesn't need to justify the arrest. It just records what happened. And what it records is often different from what the officer remembers writing down three hours later at a desk.

The dashcam doesn't have a quota, doesn't need to justify the arrest, and doesn't write its report from memory three hours later, it just shows what happened.

Common discrepancies:

• Officer reports "subject stepped off the line 3 times" but footage shows the "line" was an imaginary line on gravel with no visible reference point
• Officer reports "subject could not maintain balance during instructions" but footage shows subject standing still with minor weight shifting (which is normal)
• Officer reports "subject swaying while standing" but footage shows the same amount of sway the officer exhibits while standing
• Officer reports administering the HGN test for "approximately 2 seconds per pass" but footage shows the entire HGN test took 45 seconds total for all six passes, meaning each pass was about 1 second. Far too fast.

Effective cross-examination of FST cases starts by extracting favorable admissions before the officer knows where the questioning is going. That means getting the officer to commit to their training on the stand: "You were trained that each HGN pass should take 2 seconds, correct?" Then the defense shows the dashcam footage proving they didn't follow it. The officer has to either admit they violated their training or claim the training doesn't matter. Neither answer helps the prosecution.

Ask your attorney whether they've watched the dashcam and bodycam footage side-by-side with the police report. If the answer is no, that is the next thing that needs to happen in your case, today.

Questions to ask your attorney:

1. Has the dashcam and bodycam footage been obtained and reviewed?
2. Does the footage match the officer's report on the FST administration?
3. Were the tests administered on a flat, dry, level surface?
4. Did the officer follow the NHTSA instruction protocol before each test?
5. Were you offered the option to remove inappropriate footwear?
6. Did the officer ask about medical conditions, injuries, or medications before testing?
7. Was the one-leg stand actually timed, or estimated?

Copy these seven questions into a text message to yourself right now. Bring them to your next attorney meeting. Each one exposes a different way the test administration may have deviated from the standards the prosecution claims were followed.

Why This Matters for Your Case

You're sitting in court. The prosecutor says, "The defendant failed the field sobriety tests." It sounds definitive. But the tests were administered on gravel, the officer rushed the eye test, skipped the medical screening, and let you stand on one foot for 40 seconds instead of 30. "Failed" is doing a lot of work in that sentence.

Field sobriety tests aren't pass/fail exams with objective scoring. They're a trained officer's subjective observations filtered through a scoring framework that the government's own research says is wrong 18-35% of the time (NHTSA).

When officers deviate from NHTSA standards, the tests lose even that limited reliability. Wrong surface, rushed instructions, improper stimulus speed, no medical history inquiry. They become what elite DUI defenders call "an officer's hunch with a scientific label."

But here's what nobody mentions: suppressing the FSTs doesn't just eliminate the test results, it can unravel the entire case. If the FSTs were the basis for probable cause to arrest you and administer a breath test, killing the FSTs kills the arrest. Killing the arrest kills the breath test. One domino topples the chain.

If the FSTs were the basis for your arrest, suppressing them doesn't just remove the test results, it removes the legal basis for everything that happened after.

A suppression motion based on FST administration errors doesn't just challenge the test results. If the FSTs were the primary basis for probable cause to arrest you and administer a breath or blood test, suppressing the FSTs can unravel the entire chain of evidence. No valid FSTs means no probable cause for arrest. No probable cause means no lawful basis for the breath test. The BAC result gets excluded too.

That's the cascade. And it starts with one question: did the officer follow the standards? Combine FST challenges with breathalyzer calibration evidence and you may have grounds for getting your DUI dismissed. For the full picture, read our complete DUI defense guide.

Write down three things you remember about how your field sobriety tests were administered: the surface, the instructions, and the timing. If any of the three deviate from what NHTSA requires, you have a question your attorney needs to answer. Five minutes of recall could be the foundation of your defense.

What We Do

We review DUI arrest reports, dashcam footage descriptions, and case documents to identify exactly where officers deviated from NHTSA field sobriety test standards. We give you the specific questions and evidence points to bring to your attorney so they can build a suppression motion that actually has teeth.

Not sure where your DUI case stands? Take the free Case Progress Score, 5 minutes to see what's been done, what's missing, and what to push on next.

Want the full defense toolkit? The DUI Defense Playbook includes 26 attorney questions, a 12-point evidence checklist, and a case stage roadmap from arrest to resolution.

---

This is legal information, not legal advice. We are not attorneys and do not provide legal representation. Field sobriety test standards and admissibility rules vary by state. Always work with a licensed DUI attorney in your jurisdiction.