When a police officer pulls someone over, for any reason, they are trained to look for signs of drinking and driving. If the officer suspects impairment, they will usually ask the driver to perform field sobriety tests, which are meant to gauge the level of impairment and whether further chemical testing is warranted. The catch is that these tests are only as reliable as the way they are administered, and they are frequently given wrong, or under conditions that would trip up a completely sober person.
The standardized tests, and their accuracy claims
Horizontal Gaze Nystagmus (HGN). You follow a pen, finger, or small light with your eyes without moving your head while the officer watches for an involuntary jerking of the eye, which can appear at a BAC above the .08 legal limit. Proponents put its accuracy near 80%, but only when administered exactly by protocol, and nystagmus has many innocent causes unrelated to alcohol.
Walk-and-Turn. You walk a straight line for about nine to ten heel-to-toe steps, pivot on one foot, and return. Falling, losing balance, or using your arms to steady yourself are scored as signs of impairment. Its accuracy is put near 70%, and the roadside (uneven pavement, traffic, flashing lights, the wrong shoes, cold, nerves) is a poor place to take it.
One-Leg Stand. You stand on one leg for about 30 seconds. Swaying, hopping, putting the foot down, or using your arms are scored as clues, again at roughly 70% accuracy. Age, weight, injuries, inner-ear conditions, and simple anxiety all affect balance and have nothing to do with alcohol.
Non-standardized tests carry even less weight
Officers sometimes add tests with no standardized protocol at all, closing your eyes and touching your nose with a fingertip, or reciting the alphabet backward. These are not validated the way the three standardized tests are, and their results are correspondingly easier to challenge.
What made the officer stop you
Regardless of the stated reason, officers watch for driving they associate with impairment: weaving in and out of a lane, driving too fast or too slow, stopping in the wrong place, or drifting to the wrong side of the road. But sober drivers do all of these too, distracted, tired, unfamiliar with the area, and that gap between what an officer assumes and what actually happened is often where a defense begins.
Then comes the chemical test
If the officer still suspects impairment, they can request a chemical test, usually a breath test, sometimes blood or urine. You do not choose which test; the officer does. You can ask to consult a lawyer about whether to take it, but refusing carries its own, generally stiffer, penalties. A reading of .08 or higher leads to a DWI arrest. Everyone metabolizes alcohol differently, and body size matters, which is one more reason a single roadside performance is not the whole story. See the stop, breath test, and refusal.
Where the defense comes in
These tests are not pass-fail science. The accuracy figures assume the officer followed the protocol exactly, in acceptable conditions, and scored honestly, and in case after case, that is not what happened. We examine how the tests were given, the conditions on the roadside, your physical condition, and whether any video matches the officer's written account. Weaknesses there can undermine the probable cause for the arrest and the strength of the whole case. Our DWI FAQ lays out many more of the specific challenges we look for.