How to Challenge Field Sobriety Test Results in Texas

The eye test counts as scientific evidence. That means the State only gets to use it if it proves three things, and the third one is whether the officer did the test correctly in your case.

Two of the three are already settled against you. Texas courts have accepted the science and the method. The third has to be proved every single time, and it is the one the State most often cannot prove, because the test has a required rhythm measured in seconds and the whole thing is on video.

Sixteen seconds minimum for one clue. Sixteen more for the next. Four seconds held at each corner of your vision. You can count it yourself.

But there is a catch: somebody has to ask for the hearing. It does not happen on its own.

Three tests, two different arguments

The three roadside tests are not treated the same way in law, and treating them the same is the most common mistake made here.

The eye test counts as scientific evidence. It rests on a claim about how alcohol affects involuntary eye movement, and that is a scientific claim. So the State has to clear a formal bar before it can be used, which means it can be kept out of your case entirely.

The walk-and-turn and one-leg-stand are not scientific in that same way. They are divided-attention exercises, and an officer describing how you did is closer to describing what they saw. The attack there is on how the test was given and how it was scored, and it usually affects how much the evidence is worth rather than whether the jury hears it at all.

So the eye test is a fight about whether it comes in, and the other two are usually a fight about what they mean. Both matter, and they are run differently.

For what the tests are and how they fail as science, see field sobriety tests.

The hearing somebody has to ask for

Texas has a rule governing all scientific evidence. If specialized knowledge will help the jury, a qualified expert can give an opinion about it.

Because the eye test involves the science of how alcohol affects your eyes, it always falls under that rule.

But somebody has to ask for the hearing.

That sentence is the practical heart of this page. The hearing does not happen automatically. If nobody requests it, the officer just takes the stand, testifies that you showed six clues, and the whole framework never gets used. The evidence comes in by default.

The hearing happens with the jury out of the room, and it has to be requested before the officer gives that opinion. Some judges handle it at the start of trial. Others send the jury out in the middle of testimony. Either way, once a jury has heard that you showed six clues, the argument has already been lost in the way that counts.

The three things the State has to show

Texas dropped the older test in 1992. The standard now asks whether the evidence is reliable, and it has three parts.

For scientific evidence to count as reliable:

1. The science behind it has to be valid
2. The method for applying that science has to be valid
3. The method has to have been applied correctly in your case

The burden is on the State, and it is a higher one than usual. Because juries have a hard time judging scientific testimony, the State has to convince the judge by clear and convincing evidence, not just by tipping the scale.

In deciding whether it is reliable, the judge can look at things including:

• How widely the science and the method are accepted by scientists
• How qualified the witnesses are
• Whether published research supports or rejects it
• How often the method gets the wrong answer
• Whether other experts are available to test and evaluate it
• How clearly it can be explained to the court
• The experience and skill of the person who actually did it that night

Notice the fourth and the last. The error rate, and the skill of the specific officer who tested you, are both officially part of the analysis.

The one that is still open in every case

Here is where the fight actually is, and it is worth being honest about what is not available to you.

In 1994 the highest criminal court in Texas applied this standard to the eye test. It did its own research into the science, looked at the published literature, and concluded that both the science is reliable enough and that the method, as designed by the federal highway safety agency, is reliable.

So the first two are settled. Arguing to a Texas judge that the eye test is junk science means arguing against binding authority, and it will not work.

The third was left open, and it is open in every single case.

Whether the method was applied correctly on your night has to be decided case by case. And the court signaled that a badly administered test should be thrown out when the officer did not follow the official protocol.

That is your argument. Not that the science is wrong. That this officer, on this night, did not do it the way the protocol requires, which means the reliability the appeals courts accepted does not attach to what happened to you. How that hearing runs, and what the State has to establish in it, is at the reliability hearing.

It is also, unusually for a legal argument, something anyone can check with the video and a stopwatch.

Is the officer even an expert?

Before any of this, there is a bar the State has to clear, and it clears it more easily than you would expect.

The State can only prove those three things through an expert witness. So it first has to establish that the officer who gave you the eye test counts as an expert.

Texas courts have made that easy. An officer with a practitioner’s certificate qualifies as an eye test expert, and nothing more is needed.

Here is what those certificates actually involve, because the difference is revealing:

Texas requires only the first one. Three days in a classroom is enough to qualify an officer as an expert witness on the scientific evidence being used against you. The certificate that involves actually proving they get it right in the field, checked against real test results, is not required.

And there is a practical point worth raising every time. Most courts do not ask for proof of certification. The officer’s word that they have it is usually accepted. Asking for the actual certificate, and checking its date against your arrest, costs nothing and sometimes turns something up.

The line the officer is not allowed to cross

This limit is clear, it gets crossed often, and it is easy to miss when it happens.

An officer qualified on how to give the eye test can testify about how you did on it. They cannot connect your performance to a specific alcohol level.

So an officer can say you showed six clues. They cannot say that six clues means your alcohol level was 0.10, or above 0.08, or any particular number.

Watch for it in softer forms, because it rarely shows up as a bare number. “Six clues indicates a person is over the legal limit” is the same testimony wearing a coat.

This connects to something covered on the evidence pages: more clues does not mean more drunk. The clues are not a scale, and treating the count as a measurement is exactly what the rule forbids.

Counting the seconds on your video

Roadside tests are normally recorded, and that is what makes this winnable. You can watch the test and work out whether it was given properly.

The eye test has a required rhythm, and the minimum times are countable:

Two clues alone need at least thirty-two seconds of moving the stimulus, before the pre-test checks and before the third clue.

Officers routinely finish the whole eye test in less time than that. When the video shows it, the argument is not a matter of opinion. It is arithmetic.

Add the pre-test screening questions, which are also part of the protocol and also on video: equal pupil size, jerking while looking straight ahead, whether your eyes track together, and whether you have a head injury. Skipping those is a departure too. The full protocol is at the HGN eye test.

The walk-and-turn and one-leg-stand

Different tests, different attack, and usually a fight about what they are worth rather than whether they come in.

These two get attacked on how they were given and scored, and the productive line is comparing what the officer marked against what each clue actually requires. Each clue has a specific definition, several are narrower than officers treat them, and each counts once no matter how many times it happens. The clue-by-clue detail is in the evidence cluster: the walk-and-turn and the one-leg-stand.

What matters procedurally is this. Because these are usually arguments about weight, they get made to the jury, through questioning and the video, rather than to the judge before trial. That changes the preparation completely, and it means the video is not supporting material. It is the argument.

There is also an angle available when an officer presents these as science, as measuring your alcohol level, or offers an opinion that you “failed.” That is the same framework used against non-standardized tests.

What if I felt pressured into doing them?

People ask this constantly, and the answer is not the one you want. Texas courts have admitted test results where someone first refused, then agreed after the officer threatened to take them to jail. So the fact that you felt you had no real choice does not, by itself, keep the tests out.

What it can still do is shape how the jury sees the whole encounter, and it is worth establishing from the video: what you were told, what you were asked, whether anyone told you the tests were optional, and what happened when you hesitated.

Whether you had to take them at all, and what declining actually costs you, is at your rights during a stop.