Jorge Castillo v. the State of Texas

• Court: Court of Appeals of Texas
• Decided: March 13, 2025
• Docket: 01-23-00334-CR
• Status: Published

Opinion

Opinion issued March 13, 2025

In The

Court of Appeals For The

First District of Texas ———————————— NO. 01-23-00334-CR ——————————— JORGE CASTILLO, Appellant V. THE STATE OF TEXAS, Appellee

On Appeal from the County Criminal Court at Law No. 4 Harris County, Texas Trial Court Case No. 2350292

MEMORANDUM OPINION

A jury convicted Jorge Castillo of driving while intoxicated, 1 and the trial

court sentenced him to 30 days in jail. In two issues on appeal, Castillo contends he

received ineffective assistance of counsel at trial because counsel failed to

1 See TEX. PENAL CODE § 49.04(a)–(b). (1) meaningfully cross-examine, impeach, or rebut the testimony of the patrol officer

who administered the field-sobriety test and (2) make a legally supportable argument

for suppression of the field-sobriety-test results. Because Castillo has not met at

least one element of an ineffective-assistance claim for each contention, we affirm.

I. Background

A little after 2 a.m. on March 20, 2021, a patrol officer with the La Porte

Police Department saw a car traveling in front of him cross over several lanes

without signaling. The patrol officer made a traffic stop and noticed that the driver,

Castillo, had “slurred speech,” “red glossy eyes,” “sluggish” movement, and an

“odor of alcohol.” Castillo admitted to having two to three drinks and consented to

field-sobriety testing.

The standard field-sobriety testing developed by the National Highway

Traffic Safety Administration (“NHTSA”) involves three tests: the horizontal gaze

nystagmus (“HGN”), the walk and turn, and the one-leg stand.2 The patrol officer

2 Castillo asks us to take judicial notice of the (1) 2018 NHTSA Participant Manual for DWI Detection and Standardized Field Sobriety Testing and (2) 2018 NHTSA Instructor Guide for DWI Detection and Standardized Field Sobriety Testing, which were in effect at the time of Castillo’s arrest but not included in the record. See NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., Participant Manual: DWI Detection & Standardized Field Sobriety Testing (SFST) (2018), https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/sfst_full_participant_manu al_2018.pdf; NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., Instructor Guide: DWI Detection & Standardized Field Sobriety Testing (SFST) (2018), https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/sfst_full_instructor_manual _2018.pdf. This Court and others have taken judicial notice of NHTSA manuals before. See, e.g., Emerson v. State, 880 S.W.2d 759, 765–69 (Tex. Crim. App. 2 determined that Castillo could not perform either the walk and turn or the one-leg

stand because of a preexisting back injury, so only the HGN test was administered.

Castillo told the patrol officer that he had a “dropped eye socket,” which the patrol

officer interpreted as meaning a “lazy eye.” But the patrol officer did not notice a

lazy eye when he checked Castillo’s eyes for tracking ability and pupil size, which

were both equal. According to the patrol officer, Castillo demonstrated all six clues

indicating intoxication—lack of smooth pursuit, nystagmus or an involuntary

jerking movement in both eyes at maximum deviation, and onset of nystagmus

before 45 degrees in both eyes.

Castillo was arrested for driving while intoxicated and verbally consented to

giving a blood sample, which was drawn about an hour and a half after the traffic

stop and two and a half hours after Castillo’s last drink. Testing of the blood sample

showed that Castillo’s blood-alcohol content exceeded the legal limit for driving at

the time of the blood draw. A retrograde extrapolation to determine Castillo’s

blood-alcohol content at the time of the stop was not performed.

After a trial at which the patrol officer testified about Castillo’s traffic stop

and arrest and a forensic toxicologist testified about the blood-sample testing and

1994) (taking judicial notice of HGN technique as set out in NHTSA manual); Kirby v. State, No. 01-07-00444-CR, 2008 WL 2930181, at *3–6, n.5 (Tex. App.— Houston [1st Dist.] July 31, 2008, no pet.) (mem. op., not designated for publication) (same). And we do so here. 3 results, the jury convicted Castillo. He was sentenced to 30 days in county jail. This

appeal followed.

II. Ineffective Assistance of Counsel

Both of Castillo’s issues on appeal challenge whether counsel provided

ineffective assistance at trial because he failed to meaningfully challenge the

admissibility and credibility the HGN evidence, which was critical to the State’s

case. In his first issue, Castillo argues that no competent attorney would have failed

to cross-examine, impeach, or rebut the patrol officer’s “scientifically incorrect”

testimony about HGN-testing procedure, statistics, and data. And in his second

issue, Castillo argues that counsel unreasonably failed to present two “legally

supportable” theories for suppressing the HGN-test results: (1) the patrol officer did

not follow the NHTSA manual when he administered the smooth-pursuit phase of

the test; and (2) Castillo had an eye injury that invalidated the test results. But for

counsel’s failures, Castillo says, he would not have been convicted because the

State’s only other evidence of intoxication was “unreliable blood test results” and

“some video footage of Castillo exhibiting behavior that could be attribut[ed] to his

disabilities.”

A. Standard of review

The United States Constitution and the Texas Constitution guarantee

individuals the right to assistance of counsel in a criminal prosecution. U.S. CONST.

4 amend. VI; TEX. CONST. art. 1, § 10. This right to counsel is more than the mere

presence of a lawyer; it is a right to effective assistance. Lopez v. State, 343 S.W.3d

137, 142 (Tex. Crim. App. 2011); see Ex parte Flores, 387 S.W.3d 626, 633 (Tex.

Crim. App. 2012) (observing that “the right to counsel preserves the fairness,

consistency, and reliability of criminal proceedings by ensuring that the process is

an adversarial one.”). Effective assistance is not “errorless counsel,” but rather

“objectively reasonable representation.” Lopez, 343 S.W.3d at 142.

We evaluate claims that counsel was constitutionally ineffective under the

standard set out in Strickland v. Washington, 466 U.S. 668, 687 (1984). Under

Strickland, the appellant must show that (1) counsel’s performance fell below an

objective standard of reasonableness and (2) there is a reasonable probability that,

but for counsel’s unprofessional errors, the result of the proceedings would have

been different. Id. at 687–94; Lopez, 343 S.W.3d at 142. “A reasonable probability

is a probability sufficient to undermine confidence in the outcome.” Strickland, 466

U.S. at 694. The appellant must establish both Strickland prongs by a preponderance

of the evidence. Jackson v. State, 973 S.W.2d 954, 956 (Tex. Crim. App. 1998).

Failure to make a showing under either prong defeats an ineffective-assistance claim.

See Rylander v. State, 101 S.W.3d 107, 110 (Tex. Crim. App. 2003).

In reviewing counsel’s performance, we look to the totality of the

representation to determine counsel’s effectiveness, indulging a strong presumption

5 that his performance fell within the wide range of reasonable professional assistance

and trial strategy. See Robertson v. State, 187 S.W.3d 475, 482–83 (Tex. Crim. App.

2006). Generally, a silent record that provides no explanation for counsel’s actions

will not overcome the strong presumption of reasonable assistance. See Goodspeed

v. State, 187 S.W.3d 390, 392 (Tex. Crim. App. 2005). “[T]rial counsel should

ordinarily be afforded an opportunity to explain his action before being denounced

as ineffective.” Id. (quotation omitted). “Absent such an opportunity, an appellate

court should not find deficient performance unless the challenged conduct was ‘so

outrageous that no competent attorney would have engaged in it.’” Id. (quoting

Garcia v. State, 57 S.W.3d 436, 440 (Tex. Crim. App. 2001)). “In the majority of

cases, the record on direct appeal is undeveloped and cannot adequately reflect the

motives behind trial counsel’s actions.” Mallett v. State, 65 S.W.3d 59, 63 (Tex.

Crim. App. 2001).

B. Failure to cross-examine or impeach the patrol officer’s testimony

Castillo argues that counsel provided ineffective assistance by failing to

meaningfully cross-examine or impeach the patrol officer’s testimony about

HGN-testing procedure, statistics, and data. Specifically, Castillo argues that

counsel should have used the NHTSA manual to challenge the patrol officer’s

“scientifically incorrect” testimony on:

6 • the reliability of HGN testing when conducted alone;

• whether lighting or weather conditions can affect the test results;

• whether the HGN test involves a neurological function; and

• the proper way to conduct the smooth-pursuit phase of the test. Castillo also argues that counsel should have “followed through on

cross-examination about the effect of an eye disorder or injury on [the] officer’s

decision to conduct the HGN test.” According to Castillo, counsel’s failure to

challenge the patrol officer’s testimony on these subjects can only be attributed to a

“lack of preparedness,” and not a “strategic choice.”

In considering Castillo’s complaint, we are mindful that “[c]ross-examination

is inherently risky, and a decision not to cross-examine a witness is often the result

of wisdom acquired by experience in the combat of trial.” Ex parte McFarland, 163

S.W.3d 743, 756 (Tex. Crim. App. 2005). “Furthermore, cross-examination is an

art, not a science, and it cannot be adequately judged in hindsight.” Id.

The record here contains no explanation for why counsel decided not to

cross-examine or impeach the patrol officer about the complained-of testimony.

This silence makes the cases Castillo relies on to show ineffective assistance

distinguishable.

For instance, Castillo relies on Ex parte Ard to support his argument that the

failure to present expert testimony constitutes ineffective assistance. See No. AP-

75704, 2009 WL 618982, at *2 (Tex. Crim. App. 2009) (mem. op., not designated 7 for publication). But Ard was decided on a record developed through post-

conviction writ proceedings. Id. at *1. The applicant, who was convicted of

aggravated sexual assault of a child, argued that his trial counsel rendered ineffective

assistance by failing to adequately present expert testimony on memory

implantation. Id. at *2–3. Memory implantation was a critical issue for the defense,

which rested on a theory that the complainant’s accusations resulted from suggestion

and coaching that tainted his memory. Id. at *2. The applicant argued that expert

testimony on the scientific basis of memory implantation might have convinced the

jury that the complainant’s version of events was unreliable. Id. And the Court of

Criminal Appeals agreed after considering a record that included statements from

trial counsel about the defensive strategy and testimony showing the substantive and

comprehensive differences in the expert’s trial and writ-hearing testimony. Id. at

*3–5.

Likewise, the Court of Criminal Appeals’ decision to grant post-conviction

habeas relief in Ex parte Saenz rested on a developed record. See 491 S.W.3d 819,

828–29 (Tex. Crim. App. 2016). There, the applicant sought relief from murder and

aggravated-assault convictions based on trial counsel’s failure to impeach the

complainant with a prior inconsistent statement to police. Id. at 821. The

complainant, who had been shot in the back in a drive-by shooting, told police that

he could not identify the shooter after the shooting. Id. at 822. But he purported to

8 identify the applicant as the shooter at trial. Id. at 823. The record included

deposition testimony from trial counsel that the defensive strategy was to challenge

the complainant’s credibility, that he knew the complainant told police that he would

not recognize the shooter, that it would have been a mistake not to impeach the

complainant with the prior inconsistent statement, and that such an omission likely

would have an oversight on his part. Id. at 827. In criticizing the habeas court’s

finding that trial counsel’s failure to impeach the complainant could have been a

strategic choice, the Court noted the posture of the case and the state of the record:

If this case were before us on a petition for discretionary review with an undeveloped record, [the habeas court’s] speculation might well be founded. . . . [I]n the absence of evidence of counsel’s reasons for the challenged conduct, an appellate court commonly will assume a strategic motivation if any can possibly be imagined, . . . and will not conclude the challenged conduct constituted deficient performance unless [it] was so outrageous that no competent attorney would have engaged in it. Id. at 828 (quotation omitted). But because the record was developed and trial

counsel had responded to the allegations of ineffectiveness, the Court could conclude

that counsel’s decision was not strategic. Id.

Although Castillo has cited other cases involving direct appeals, rather than

post-conviction writs, they too benefited from developed records. See, e.g., Fulton

v. State, 576 S.W.3d 905, 917 (Tex. App.—Tyler 2019, pet. ref’d) (record supporting

claim that counsel was ineffective for failing to impeach an eyewitness was

developed as part of motion for new trial). In contrast here, Castillo did not move

9 for a new trial or take any other steps to develop the record on appeal. His suggestion

that cross-examination should have been conducted in a different way “does not

rebut the presumption that counsel’s conduct fell within the wide range of reasonable

professional assistance.” Resendiz v. State, 112 S.W.3d 541, 548 (Tex. Crim. App.

2003). For example, even if, as Castillo suggests, the patrol officer incorrectly

testified that the accuracy rate of HGN testing is 88 percent when conducted alone,

rather than 77 percent, we cannot say that a strategic decision not to emphasize the

77 percent accuracy rate would be objectively unreasonable. Nor can we say that it

would be objectively unreasonable to decide not to cross-examine the patrol officer

about eye disorders when the patrol officer testified that Castillo qualified for the

HGN test because he had equal tracking and pupil size.

In short, we cannot say on this record that counsel’s failure to cross-examine

or impeach Officer Ellis fell below an objective standard of reasonableness. See

Strickland, 466 U.S. at 687–88. We overrule the part of Castillo’s first issue

claiming counsel was ineffective for failing to meaningfully impeach or

cross-examine the patrol officer. See id.; see also Williams v. State, 301 S.W.3d

675, 687 (Tex. Crim. App. 2009) (“An appellant’s failure to satisfy one prong of the

Strickland test negates a court’s need to consider the other prong.”).

10 C. Failure to call a rebuttal expert

Like cross-examination, the decision whether to present witnesses is largely a

matter of trial strategy. Shanklin v. State, 190 S.W.3d 154, 164 (Tex. App.—

Houston [1st Dist.] 2005, pet. dism’d). To prevail on his claim that counsel was

ineffective for failing to call an intoxication expert witness to rebut the patrol

officer’s testimony about the HGN testing, Castillo must show that an intoxication

expert was available and would have testified in a manner beneficial to him. See

Noriega v. State, No. 01-16-00404-CR, 2017 WL 3597732, at *5 (Tex. App.—

Houston [1st Dist.] Aug. 22, 2017, no pet.) (mem. op., not designated for

publication); Jones v. State, 500 S.W.3d 106, 116 (Tex. App.—Houston [1st Dist.]

2016, no pet.). But Castillo has not done so, and thus the record does not contain

the evidence necessary to overcome the presumption that counsel’s actions were

reasonable, professional, and motivated by sound trial strategy. See, e.g., Rivera v.

State, No. 01-18-00078-CR, 2019 WL 2041058, at *3 (Tex. App.—Houston [1st

Dist.] May 9, 2019, pet. ref’d) (mem. op., not designated for publication) (rejecting

claim that trial counsel was ineffective for failing to obtain an expert in cell-site

location information because record did not show counsel’s reasons for not obtaining

such an expert, how such an expert could have assisted trial counsel, or “the

substance of the testimony that such an expert could have provided”); Jones, 500

S.W.3d at 116 (“Though appellant suggests two categories of expert testimony that

11 would have assisted in his defense, he has not adduced evidence showing that such

experts were available or that either could have offered beneficial testimony.”).

We hold that Castillo has not met Strickland’s first prong requiring proof that

counsel’s performance fell below an objective standard of reasonableness, and we

overrule the part of Castillo’s first issue claiming counsel was ineffective for failing

to call a rebuttal expert on intoxication. See Strickland, 466 U.S. at 687–88; see also

Williams, 301 S.W.3d at 687.

D. Failure to suppress HGN evidence

Castillo next complains that counsel should have filed a motion to suppress

the HGN evidence because (1) the patrol officer incorrectly administered the

smooth-pursuit phase of the test and (2) the test is not reliable when administered to

people, like him, who have an eye injury.3

1. The HGN test

To prevail on his claim that counsel was ineffective for failing to file a motion

to suppress, Castillo must show that the result of the proceeding would have been

different—that the motion to suppress would have been granted and that the

remaining evidence was not enough to support his conviction. See Jackson, 973

3 Castillo acknowledges that counsel filed a pretrial motion to suppress the HGN evidence and objected to its admission at trial. His complaint is that counsel “inexplicably focused on multiple unsupported legal theories of inadmissibility” in challenging the HGN evidence, instead of the potentially winning theories he asserts on appeal. 12 S.W.2d at 957; Ex parte Jones, 473 S.W.3d 850, 854 (Tex. App.—Houston [14th

Dist.] 2015, pet. ref’d) (explaining failure to file motion to suppress is not per se

ineffective because counsel does not have to file futile motions).

Testimony on HGN testing is considered scientific testimony. See Emerson

v. State, 880 S.W.2d 759, 763 (Tex. Crim. App. 1994); McCarthy v. State, No.

01-12-00240-CR, 2013 WL 5521926, at *3 (Tex. App.—Houston [1st Dist.] Oct. 3,

2013, no pet.) (mem. op., not designated for publication). Under Texas Rule of

Evidence 702, the proponent of scientific evidence must show by clear and

convincing proof that the proffered evidence is “sufficiently relevant and reliable to

assist the jury in accurately understanding other evidence or in determining a fact in

issue.” Weatherred v. State, 15 S.W.3d 540, 542 (Tex. Crim. App. 2000); see Kelly

v. State, 824 S.W.2d 568, 572 (Tex. Crim. App. 1992). In a criminal case, the

reliability inquiry has three criteria: (1) the underlying scientific theory is valid,

(2) the technique applying the theory is valid, and (3) the technique was properly

applied on the occasion in question. Kelly, 824 S.W.2d at 572. The Court of

Criminal Appeals has taken judicial notice of the validity of the scientific theory and

technique applying the theory for HGN testing. See Emerson, 880 S.W.2d at 769.

But the proponent of the evidence must still show that the officer properly applied

the HGN test. See id.; see also McRae v. State, 152 S.W.3d 739, 743 (Tex. App.—

Houston [1st Dist.] 2004, pet. ref’d) (“HGN evidence is reliable, admissible

13 scientific evidence under rule 702 when performed by a police officer who is

certified by the [NHTSA] and who applies the technique properly.” (footnote

omitted)). Whether the HGN test is properly applied is decided on a case-by-case

basis. Somers v. State, 368 S.W.3d 528, 536 n.27 (Tex. Crim. App. 2012).

An officer must follow the standardized procedures set out in the NHTSA

manual when administering the HGN test. Emerson, 880 S.W.2d at 768; McRae,

152 S.W.3d at 743. The officer looks for three criteria: (1) the lack of smooth

pursuit—an inability to pursue an objection, or stimulus, moving horizontally across

the suspect’s field of vision smoothly; (2) distinct and sustained nystagmus at the

eye’s maximum deviation; and (3) the onset of nystagmus before 45 degrees. See

Emerson, 880 S.W.2d at 766 (citing 1992 NHTSA manual). An officer must look

for these criteria in each eye, for a total of six “clues.” Id.; see Compton v. State,

120 S.W.3d 375, 377–78 (Tex. App.—Texarkana 2003, pet. ref’d) (“In other words,

if the person’s eyes fail to smoothly follow a stimulus across a field of vision, exhibit

nystagmus when held as far to the side as possible, or display nystagmus at an angle

prior to forty-five degrees from looking straight ahead, each will be considered a

clue indicating possible intoxication.”). To ascertain these clues, an officer conducts

“passes”—asking an individual to follow an object or stimulus, such as a pen, with

their eyes. See Kamen v. State, 305 S.W.3d 192, 195 n.1 (Tex. App.—Houston [1st

Dist.] 2009, pet. ref’d). The procedures also require an officer to assess possible

14 medical impairment that could lead to an incorrect determination of intoxication by

looking for equal pupil size, equal tracking, and resting nystagmus. See Emerson,

880 S.W.2d at 766.

2. The trial court could have reasonably denied a motion to suppress

We turn first to Castillo’s argument that counsel should have moved to

suppress the HGN evidence because the patrol officer administered the smooth-

pursuit test too quickly. Specifically, Castillo points to the NHTSA’s instruction to

move the stimulus “steadily at a speed that takes approximately 2 seconds to bring

the eye from center to side.” NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., Participant

Manual: DWI Detection & Standardized Field Sobriety Testing (2018), at VIII-41.

He also points to the video of his arrest, which he argues shows the patrol officer

taking less than one second to move the stimulus, instead of two. The State responds

that Castillo cannot satisfy Strickland’s prejudice prong because even assuming he

is correct that the patrol officer administered the smooth-pursuit test too quickly, the

trial court would not have been compelled to suppress the HGN evidence. In support

of their respective positions, Castillo and the State both cite authority from this

Court.

To support his position that counsel was ineffective in failing to move to

suppress the HGN evidence based on the patrol officer’s administration of the test,

Castillo relies on this Court’s holding in McRae. 152 S.W.3d at 743–44. The

15 appellant in McRae argued that the arresting officer’s testimony that the appellant

had six clues on the HGN test should have been excluded because the officer

admitted he incorrectly administered the test. Id. at 743. Although the officer

initially testified that he administered all three parts of the HGN test to the appellant,

he later admitted that he had administered only two of the three parts required by the

NHTSA standards. Id. He also admitted making several other misstatements about

the HGN test. Id. When confronted about his errors and asked whether there was a

“valid HGN test,” the officer replied that there was not. Id. at 744. In agreeing with

the appellant that the HGN evidence should have been excluded on that basis, this

Court observed that, although “[s]light deviations in the administration of the HGN

test do not render the evidence inadmissible or unreliable,” the officer’s own

admissions precluded any finding that the HGN technique was properly applied or

that any variation from the testing procedure was slight. Id.

For its contrary position that counsel was not ineffective, the State relies on

McCarthy v. State, another DWI case. No. 01-12-00240-CR, 2013 WL 5521926

(Tex. App.—Houston [1st Dist.] Oct. 3, 2013, no pet.) (mem. op., not designated for

publication). As in McRae, the McCarthy appellant argued that the HGN evidence

was inadmissible because the officer did not administer the test according to the

NHTSA standards. Id. at *2. Specifically, the appellant asserted the officer (1) did

not make the required number of passes for a complete HGN test, (2) performed the

16 smooth pursuit portion of the test in half the time the NHTSA manual required, and

(3) completed the distinct and sustained nystagmus portion too quickly by holding

the stimulus at maximum deviation for two or three seconds instead of four. Id. at

*5. The State responded that the NHTSA manual provided only recommendations

for the time and number of passes and that the officer administered the test within

acceptable limits. Id. The trial court admitted the HGN evidence, concluding that

any deviation went to the “weight and not the admissibility” of the evidence. Id.

This Court agreed. Id. at *4–6.

In holding the HGN evidence was admissible in McCarthy, the Court

distinguished McRae. Id. at *5. Unlike in McRae, the officer in McCarthy did not

testify that the HGN test was invalid. Id. at *5. Instead, the officer testified that,

even if he had not performed the test “a hundred percent,” he performed all three

parts of the test and made an appropriate number of passes. Id. The officer also

testified that although the NHTSA manual recommends repeating passes, it does not

prescribe the number of passes. Id. Considering these differences and other

evidence that the officer was trained and certified in standard field-sobriety testing,

had administered the HGN test many times, asked the appropriate screening

questions, checked the appellant’s eyes for tracking ability and pupil size to

determine whether he could follow the stimulus and whether there were signs of

medical impairment, administered all three parts of the test, and explained how he

17 did so, the Court held that the trial court’s decision to admit the HGN evidence was

not an abuse of discretion. Id. at *7. The Court reasoned that, based on an

assessment of arrest video and the officer’s testimony, the trial court could have

reasonably concluded that the officer properly administered the HGN test or that any

variations in the administration of the test were slight and did not affect

admissibility. Id.

Castillo’s case is more like McCarthy than McRae. Unlike McRae, the patrol

officer did not admit that Castillo’s HGN test was invalid; rather, he testified that he

administered the HGN test in accordance with his training and the NHTSA manual’s

procedures.4 As instructed in the NHTSA manual, the patrol officer screened

Castillo for medical impairment that would affect his performance on the test by

confirming that Castillo’s eyes tracked equally and had equal pupil size.

We agree with the State that, even if Castillo is correct that the patrol officer

took less time than procedure recommends, the trial court reasonably could have

found that any variation was slight and did not affect the admissibility of the HGN

evidence. See, e.g., Plouff v. State, 192 S.W.3d 213, 221 (Tex. App.—Houston [14th

Dist.] 2006, no pet.) (rejecting appellant’s claim that HGN test results were

4 The only deviation from procedure the patrol officer acknowledged was making an accommodation for Castillo’s back pain by allowing Castillo to sit on the car during the test instead of stand with his feet together. This accommodation is not the subject of any complaint on appeal. 18 inadmissible because officer performed portions of test too quickly, did not repeat

passes, and held stimulus for only one second at maximum deviation); Compton,

120 S.W.3d at 378–79 (holding officer’s slight deviation in time to conduct lack-of-

smooth pursuit and maximum-deviation portions from time recommended in

NHTSA manual did not invalidate test results). The NHTSA manual provides only

an approximation of the time required for properly conducting the smooth-pursuit

test. See NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., Participant Manual: DWI

Detection & Standardized Field Sobriety Testing (2018), at VIII-41. And as the

Court of Criminal Appeals has noted, the HGN test will not be administered in strict

conformity with an officer’s training every time it is performed. See Emerson, 880

S.W.2d at 767 (noting “[t]he accuracy of the HGN test has been estimated at various

levels, depending on such factors as testing conditions and the ability and experience

of those conducting the test”).

Likewise, the trial court likely would not have been compelled to exclude the

HGN evidence based on Castillo’s claimed eye injury, considering the patrol

officer’s testimony that he asked the appropriate screening questions before

administering the HGN test, checked Castillo’s eyes for tracking ability and pupil

size, and did not see any symptoms of an eye injury. See McCarthy, 2013 WL

5521926, at *4 (“Before administering the HGN test, Officer Hardt asked McCarthy

the appropriate screening questions. Hardt checked McCarthy’s eyes for tracking

19 ability and pupil size to determine whether he could follow the stimulus and whether

there were signs of brain injury or illness.”); Liles v. State, No. 01-08-00927-CR,

2009 WL 3152174, at *6 (Tex. App.—Houston [1st Dist.] Oct. 1, 2009, no pet.)

(mem. op., not designated for publication) (“Secondly, Officer Godden checked to

see whether appellant’s eyes tracked equally and made sure appellant did not have a

‘lazy eye.’ Officer Godden testified that appellant had equal tracking. Because

Officer Godden testified that he checked appellant for the indicators as he performed

the HGN test, Gooden met the requirements of the NHTSA Manual.”); see also

Webster v. State, 26 S.W.3d 717, 722–23 (Tex. App.—Waco 2000, pet. ref’d)

(stating that screening for other causes of nystagmus can be performed while

conducting HGN test and is consistent with NHTSA manual).

In short, Castillo has not shown that a motion to suppress would have been

granted based on either the patrol officer’s variation from HGN-testing procedure or

an eye injury. Castillo has not met Strickland’s second prong requiring proof that

the result of the proceeding would have been different. See Jackson, 973 S.W.2d at

957. Accordingly, we overrule Castillo’s second issue.

20 III. Conclusion

We affirm the trial court’s judgment.

Andrew Johnson Justice

Panel consists of Justices Rivas-Molloy, Johnson, and Dokupil.

Do not publish. TEX. R. APP. P. 47.2(b).