The Core Thesis: Police Expertise Cannot Be Presumed

Maryland DUI litigation often treats police impairment testimony as if admissibility were automatic. The State calls an officer, invokes training, invokes experience, mentions SFST certification or DRE status, and then asks the witness to offer an opinion that the accused was impaired, under the influence, drug-impaired, alcohol-impaired, or incapable of safe driving. That sequence is familiar. It is not automatically reliable.

The admissibility question is not whether the officer has a badge. It is not whether the officer has arrested many drivers. It is not whether a prosecutor has qualified the officer before. The question is whether the opinion offered in this case rests on a reliable foundation that satisfies Maryland Rule 5-702 after Rochkind v. Stevenson. A courtroom conclusion must be supported by more than institutional authority.

The point is narrow and powerful. A police officer may generally testify to ordinary firsthand observations if the testimony is otherwise admissible. The officer can describe driving, odor, speech, eyes, balance, statements, the roadside environment, body-worn camera events, and the instructions given. But when the State transforms those observations into expert impairment science, the court must demand the missing bridge.

Maryland Rule 5-702 and Rochkind Gatekeeping

Maryland Rule 5-702 requires a court to evaluate whether expert testimony will assist the trier of fact, whether the witness is qualified, whether expert testimony is appropriate on the particular subject, and whether there is a sufficient factual basis for the opinion. After Rochkind, Maryland courts apply the Daubert reliability framework to expert testimony. That framework is not limited to laboratory scientists. It applies to technical and specialized testimony, including experience-based opinions.

Experience can be part of the foundation, but experience is not a blank check. An experience-based expert must still explain how the experience leads to the conclusion, why the experience is a sufficient basis for the opinion, and how the experience was reliably applied to the facts. A report that says “based on my training and experience” is not enough if it does not identify the method and reasoning process.

Police DUI testimony is especially vulnerable to over-admission because the same witness often performs multiple roles: investigator, arresting officer, report writer, field-test administrator, prosecution witness, and proposed expert. That creates a risk that a fact witness will be given expert aura without expert reliability. Gatekeeping is the mechanism that separates permissible observation from inadmissible analytical overreach.

What the State Must Prove

• Qualifications for the exact opinion: not merely general DUI experience, but qualification for the specific impairment, HGN, SFST, DRE, or toxicology-linked opinion being offered.
• Fit: the testimony must assist the factfinder on a proper expert subject and must match the legal issue in the case.
• Sufficient factual basis: the opinion must rest on disclosed facts, not hidden assumptions, conclusory reports, or incomplete roadside observations.
• Reliable methodology: the State must identify principles, standards, limits, and known weaknesses.
• Reliable application: the officer must account for case-specific conditions and alternative explanations.

SFST and HGN: Roadside Screens Are Not Automatic Impairment Science

Standardized Field Sobriety Tests are roadside screening tools. They are administered in uncontrolled settings: at night, near traffic, under emergency lights, on pavement that may not be level, after a stressful stop, after a crash, in bad weather, or in front of a uniformed officer with arrest power. Those conditions matter because the tests are not performed in a laboratory and are not insulated from confounding variables.

The walk-and-turn and one-leg stand are particularly vulnerable to instruction quality, demonstration accuracy, footwear, age, injury, fatigue, anxiety, surface conditions, language comprehension, and neurological or orthopedic issues. HGN has its own reliability questions, including stimulus distance, stimulus height, speed of pass, lighting, medical screening, resting nystagmus, equal tracking, equal pupil size, officer positioning, and whether the officer recorded enough detail to permit meaningful review.

The State often uses SFST evidence for a broader conclusion than the evidence can reliably support. A clue is not a diagnosis. A clue is not a statutory element. A clue is not proof that a driver was unable to operate safely. If the officer simply describes the clue, the testimony may be ordinary observation. If the officer says the clue scientifically proves impairment, the court should require a Rule 5-702 foundation.

Common SFST Analytical Gaps

• Using SFST clues to prove impairment without showing validation for that legal endpoint.
• Using HGN as if it proves unsafe driving rather than, at most, possible alcohol influence under specific conditions.
• Failing to document whether the officer asked about injuries, eye conditions, medical history, footwear, fatigue, or balance problems.
• Relying on subjective scoring when the video shows performance that appears substantially compliant or better than the report suggests.
• Giving an incomplete demonstration, unclear instruction, or non-standard instruction and then scoring the defendant against a standardized protocol.

DRE Testimony: A Checklist Is Not a Diagnosis

Drug Recognition Evaluation testimony is often presented as if the officer is diagnosing drug impairment. That is exactly why gatekeeping matters. A DRE label can carry enormous weight with a jury, but the label does not answer the reliability question. The State must show that the DRE opinion is reliable for the specific courtroom claim being made.

Drug presence and drug impairment are different concepts. A person may have a substance or metabolite in the body without being impaired at the time of driving. A DRE opinion may also be confounded by medical conditions, fatigue, anxiety, injury, medication, mental health issues, diabetes, neurological events, eye conditions, or incomplete toxicology. If the officer does not rule in and rule out reasonable alternatives, the conclusion may be prosecution-friendly inference rather than reliable expert application.

The redacted Baltimore City motion preserved in this white-paper archive illustrates the issue. The argument is not that a DRE can never testify. The argument is that a DRE report that discloses a conclusion without the underlying reasoning, validation, error rate, alternative-explanation analysis, and case-specific fit should not be treated as admissible expert science.

Daubert-Rochkind Reliability Audit for Police DUI Expert Opinions

The Proper Evidentiary Line: Admit Facts, Exclude Unsupported Expert Conclusions

The remedy does not need to be all-or-nothing. Courts can draw a disciplined evidentiary line that respects the jury’s role while enforcing Rule 5-702. The officer may testify to admissible facts. The State may play video. The defense may cross-examine. But the officer should not be permitted to use specialized titles or scientific-sounding conclusions unless the State first satisfies the expert reliability standard.

This distinction matters because jurors may give extraordinary weight to the word “expert.” A police officer already carries governmental authority. Adding an expert label can convert ordinary observations into what sounds like scientific certainty. If the State has the science, the State can prove it. If the State has only training, experience, and conclusion, the testimony should be limited.

• Generally admissible if otherwise proper: what the officer saw, smelled, heard, said, did, recorded, and personally observed.
• Subject to limiting instruction or exclusion: SFST scoring opinions that exceed the video or omit non-standard administration problems.
• Requires Rule 5-702 gatekeeping: HGN expert opinions, DRE diagnostic opinions, drug-impairment conclusions, alcohol-impairment conclusions, and opinions that a defendant was unable to drive safely.
• Should be excluded when unsupported: ultimate conclusions connected to the facts only by “training and experience.”

Source Motion Record Used for This White Paper

This page is built from two litigation-record examples. They are not generic blog sources. They are Maryland defense motions preserved as part of the forensic litigation record on DUI Defense MD.

Practical Defense Use: How to Litigate the Police-Expert Myth

A serious DUI defense should start by separating facts from expert conclusions. The defense should preserve the body-worn camera, dash camera, CAD notes, dispatch, SFST worksheet, DRE face sheet, DRE narrative, toxicology reports, chain of custody, lab disclosures, medical information, officer training records, calibration or device records when relevant, and any notice of expert testimony. The litigation goal is to force the State to identify exactly what opinion it wants and exactly what foundation supports it.

The defense should then isolate the endpoint. Is the officer claiming alcohol consumption, drug presence, BAC above a threshold, impairment, substantial impairment, inability to drive safely, drug category recognition, or causation? Each endpoint requires its own foundation. A method that might support one claim does not automatically support another.

Finally, the defense should attack application. Even a generally accepted method can fail in a particular case if the officer administered it incorrectly, omitted material screening questions, ignored obvious alternative explanations, failed to document the conditions, or overstated what the video shows. Reliable methodology plus unreliable application is still inadmissible or subject to sharp limitation.