STATE v. HRADEK (2024)

OPINION

Appellee was convicted of recklessly causing serious bodily injury to her infant son and sentenced to thirteen years in prison. The trial court granted her motion for new trial based on ineffective assistance of counsel (“IAC”). A majority of the court of appeals reversed the trial court's ruling and reinstated the conviction. State v. Hradek, 2023 WL 174878, 2023 Tex. App. LEXIS 230 (Tex. App.—El Paso, January 12, 2023) (not designated for publication). The question here is whether the court of appeals applied the correct standard of review. We conclude that the court of appeals failed to view the record in the light most favorable to the trial court's ruling. The trial court's ruling was within the zone of reasonable disagreement, so we reverse the court of appeals's judgment and remand the case to the trial court.

I. Charge and Trial

After her two-and-a-half-month-old son died, Appellee was indicted for intentionally or knowingly causing injury to a child by omission. The State later moved to amend the indictment to allege serious bodily injury. The jury charge authorized three possible verdicts: not guilty, guilty of intentionally or knowingly causing serious bodily injury by omission, or guilty of recklessly causing serious bodily injury by omission. The jury convicted her of the reckless charge. The trial's central issue was cause of death; the State's medical examiner said it was positional asphyxia, and the defense expert said that the cause could not be determined.

I.A. The State's Evidence

Appellee's son, Colton, was born six weeks early and spent a week and a half in the neonatal intensive care unit. He had a medical history of breathing problems, including decreased oxygen and sleep apnea. He was hospitalized at three weeks of age for an upper respiratory infection and had multiple apnea episodes while in the hospital. After he was released from the hospital, he was prescribed an apnea monitor. His pediatrician's records noted a week after his release from the hospital that he was on the apnea monitor and had had “no spells of apnea.”

The medical equipment supplier and the pediatrician testified that the monitor should have been used for at least six weeks, but Appellee returned it after three weeks and signed a form acknowledging that she did so against medical advice. The pediatrician testified that an order from his office to discontinue the monitor would have been documented, but there was no record of such an order. He testified that apnea monitors are the best method for monitoring a child who has apnea or a high risk for SIDS, but parents sometimes find the monitors to be annoying because up to nineteen of twenty alerts are false alarms.

On October 9, 2012, less than three weeks after she returned the apnea monitor, Appellee found Colton non-responsive and called 911.¹ First responders arrived minutes later at 10:53 a.m. and briefly administered cardio-pulmonary resuscitation but only because their protocol required it—Colton was cold and had fixed lividity and rigor mortis. According to their records, the “mother” or “family” said he had been “down” for ten minutes, and “Dad” said they last saw him moving at 3:00 a.m. when he was fed.

Because lividity was fixed in the upper, right part of his body and head, and there was blanching on his upper thighs, responding officers surmised that Colton had been strapped into an overturned car seat that was found next to Appellee's bed. The car seat was lying on its back with its base facing the wall; the lower left corner of the car seat appeared to be resting on a pile of pillows.

El Paso County deputy medical examiner Juan Contin determined that Colton died of positional asphyxiation; he could not breath because he was upside down, and the lividity indicated that he had been in that position for some time. Dr. Contin testified that Colton had deep lividity on the right side of his face, indicating that he was upside down and tilted to the right. There was little lividity in Colton's lower legs or on his left side. Colton had a marked “indentation” on his right thigh from either the car-seat strap or the diaper, but Dr. Contin said diapers do not usually cause such indentations.

Dr. Contin explained that lividity patterns are not exact because blood does not flow freely through the tissues, and its flow may be stopped by pressure. One such example was lividity in the back of Colton's legs because the diaper “was very tight” and prevented the blood from moving. Dr. Contin did not personally examine the scene or the car seat, and he could not say conclusively that Colton was in the car seat when he died, but he did not believe that Colton died while lying flat in a bed or sleeping in his crib. He believed Colton died while he was “upside down” and tilted to the right, and he remained in that position for many hours.

Dr. Contin's meaning of “upside down” is unclear from the record. He testified that it was like “sitting in an easy chair” and “towards the right”; Colton could have been positioned that way in the car seat or “on a pile of dirty clothing,” but he was not lying in a bed. But parts of Dr. Contin's testimony suggested that Colton was face down. He testified that Colton was “suspended by the straps at the thighs,” and that his “head was floating in the air. Wasn't touching anything.” Dr. Contin dismissed the absence of shoulder strap marks, saying that the “pressure was mostly on the thigh and not the shoulder; he was suspended[,]” and the shoulder straps would not leave marks if they were not tight.

In two video-recorded statements to police, Appellee denied that Colton had died in his car seat. She said her husband, Bobby Soto, grabbed the car seat in a panic when they found Colton was not breathing and then dropped the car seat by the bed when they called 911. In her first statement she said she usually put Colton in his crib but must have fallen asleep while feeding him because she woke up with him in the bed between her and Bobby. In her second statement, she said Colton would not sleep in his crib, so he routinely slept in bed with her and Bobby. She said she initially lied about Colton sleeping in the crib because she knew it was not safe for him to sleep in bed with them, and she was afraid that they would be blamed for his death. She thought that the slope of the mattress from Bobby's weight accounted for the lividity on Colton's face, head, and right side and that the blanching on his thighs was from the diaper. She said she always put Colton's diaper on too tight and it would leave marks on his legs, but she did so to keep it from leaking. She defended Bobby and did not believe that he would have done anything to hurt Colton. As for the discontinued apnea monitor, she claimed that it had not gone off in the three weeks that she used it and that someone from her pediatrician's office had told her Colton no longer needed it.

The State sought to introduce evidence that Appellee used cocaine around the time of Colton's death to show her state of mind. In an excerpt from a jail-recorded phone call between Appellee and her mother, Whitney, Appellee admitted that she used cocaine while her son was alive.² The trial judge ruled that the excerpt referencing cocaine use was admissible and asked the defense for its “thoughts on how much” of the recording's remainder to admit. The second-chair defense attorney replied, “All of it, Your Honor.” It was admitted as State's Exhibit 51 and played for the jury.³

In the 43-minute recording, Appellee referenced her job as an exotic dancer, cursed out the medical examiner and his findings, complained about the unfairness of her incarceration, and claimed that she was in jail for cocaine use rather than her baby's death. She said everyone was making her regret her son, and she wished she had never gotten pregnant and that her son had never been born. She complained about being in jail; she did not believe she had done anything wrong and was focused on getting out on bond. She relayed her plan to dye her hair and get a spray tan when she got out of jail so no one would recognize her from news coverage of her case. She promised that she was “not going to go out and start dancing again,” but she did not believe she could get a job because everyone had seen the case on the news. She dismissed her family's hardship in dealing with the situation and was concerned about being in jail, losing her son, being called a murderer, and having her picture on the news. She expressed fear that the State could increase the charges against her and said she was considering pleading guilty.

Appellee said she was “125% sure” that Colton was next to her the whole night and that she would have felt it if someone had moved him because she was a light sleeper. She asked her mother to send her photos of Colton and Bobby and to put money in Bobby's commissary account. Toward the end of the call, she expressed feelings for Colton and tearfully said she just wanted him back, but her sadness was short lived. The next minute she was no longer crying and was again asking for more money for herself and for Bobby.

Whitney was frustrated with Appellee throughout the call. She expressed her love and support for Appellee but was sometimes dismissive, harsh, or angry. She told Appellee to stop listening to “jailhouse lawyers” and to follow her attorney's advice. She repeatedly admonished Appellee to tell the truth and take this seriously. She told Appellee to stop being so demanding. When Appellee mentioned that her jail friend charged with a more serious crime had been released on bond, Whitney pointed out that the friend's victims did not die and reminded Appellee that her baby died. Whitney pointed out Appellee's selfishness in complaining about unfairness and asked Appellee, “was it fair to Colton what happened?” She referred to the medical examiner's conclusion that Colton was upside down in the car seat when he died. Whitney suggested that Appellee undergo a “psych eval” and said everyone thinks she is a murderer.

I.B. Appellee's Evidence

The defense presented Mark Shuman, a forensic pathologist who ruled out the car seat scenario as the cause of death. Dr. Shuman testified that the blanching on Colton's thighs did not match the width of the car seat's straps or their probable location on Colton's body, nor did their weave pattern appear in the autopsy photos. He testified that if Colton had been suspended in the car seat, there would have been blanching and lividity on his shoulders, but there was neither. He noted, too, a discrepancy between the blanching on the thighs and the way the car seat's straps would have fit on his legs; that is, the blanching encircled Colton's thighs, which the car seat's straps would not do. He associated the blanching's appearance with folded fabric or elastic and hypothesized that it was caused by the diaper. He noted lividity on the back of the legs and found that inconsistent with the upside-down hypothesis. He thought Colton died while on his right side or face down but not upside down.

Dr. Shuman criticized the State's failure to document the scene more thoroughly and the medical examiner's failure to test his car-seat hypothesis by placing Colton's body into the car seat. He surmised that Colton was in the bed but did not believe enough investigation had been conducted at the scene to determine the cause of death. He agreed that it was unsafe for a young baby to sleep in an adult bed and said he regularly sees accidental infant deaths due to co-sleeping, but the parents are usually not prosecuted for such accidental deaths. He said Colton had risk factors for unexplained death including prematurity, history of apnea episodes, age under six months, and low weight, but without more investigation he could not say whether Colton died of suffocation in an adult bed or of sudden unexplained death. In his opinion, cause of death could not be determined, and the charges should not have been filed.

I.C. Closing Arguments

The defense argued for acquittal because Colton's death was an unexplained accident, and the State failed in its burden of proof; the detective came up with the car seat theory, and the medical examiner adopted it without investigating it. The defense urged the jury to listen to the 911 call because it captured Appellee's “panic and horror” as she tried “to breathe life into” Colton and contradicted the detective's testimony about his observations of Appellee and the apartment.

The State argued that it did not have to prove that Colton was in the car seat when he died, only that Appellee knowingly left him in an unsafe sleep position, knowing that he had sleep apnea, and that the bed and the crib were “both unsafe.” It pointed out her inconsistent statements given during the investigation. The State maintained that Appellee used cocaine while Colton was asleep and vulnerable and knowingly put him in an unsafe sleep position, leaving him “to die for hours” and that her conduct was at least reckless.

II. Motion for New Trial

At the motion-for-new-trial hearing, the second-chair defense attorney, Nicole Maesse, testified that she had not listened to the recording of the phone call before she requested that it be played for the jury; the lead attorney told her to request its admission, and she assumed he had listened to it. If she had listened to it beforehand, she would not have admitted it. She agreed that much of its content was prejudicial to Appellee, presented her in a negative light, showed her to be a bad person generally, and was hurtful to her case.

Maesse testified that she would encourage her client to present herself as taking the situation seriously and would advise against cussing in court. She would object to evidence that a client was in jail, was considering a guilty plea, or believed she might spend the rest of her life in prison. She would object to a witness's opinion testimony that the client was lying or untruthful or might have anger issues. She would not want evidence admitted showing that the client's mother thought the client had mental health issues. In defending against a charge of injury to a child, she would object to lay opinion testimony that the client might have hurt the decedent, the situation was unfair to the decedent, and the client should think about the decedent. She would not want a family member or other witness to say the prosecution had a lot of evidence against the client. She would not want evidence presented that the client regretted having a baby or getting pregnant with the child she allegedly caused to die; it would “annihilate” the defense case. She would not knowingly offer that kind of evidence, and she would object if the State offered it. But the phone call included that kind of evidence.

The lead defense attorney, Dave Contreras, testified similarly but inconsistently. The trial court found that he was not credible when he testified that he had listened to the entire recording before it was introduced but credible when he testified that he had listened only to the snippet offered by the State. He said his trial strategy was to show that the State did not prove its case because there was no omission; the cocaine use, unsafe sleep arrangement, and returning the apnea monitor were all acts.

The trial court found that absent the defense request to admit the entire recording, only the cocaine-use excerpt would have been admitted, and its remainder included prejudicial material like Appellee wished her son had never been born, wished she had never gotten pregnant, was not taking this seriously, hurt the baby, was considering pleading guilty, was a stripper, and had anger and mental health issues. Referring to the recording, the trial court found that defense counsel realized that “the whole thing” was bad. The trial court found no credible, reasonable trial strategy for defense counsel to request that the entire phone call be admitted. It concluded that portions of the call were highly inflammatory and inadmissible, the probative value of the call was substantially outweighed by the danger of unfair prejudice, its admission was unjustified by any objectively reasonable trial strategy, and without the recording's admission and other errors by the defense, there was a reasonable probability of a different, better outcome. See Strickland v. Washington, 466 U.S. 668, 687, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984).

III. Appeal

A majority of the court of appeals overruled the trial court. Hradek, 2023 WL 174878, at *––––, 2023 Tex. App. LEXIS 230, at *17. It acknowledged the deferential standard of review applicable to new-trial rulings, but it applied the de novo standard applicable to IAC prejudice claims. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *18-21. It broke the call down into two categories: comments on Appellee's credibility and comments related to her guilt or innocence. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *55-56.

The court of appeals considered the following aspects of the call as diminishing Appellee's credibility: her topless-dancing job, Whitney's belief that Appellee underappreciated the seriousness of her case, Whitney's repeated demands that Appellee tell the truth, Whitneys’ suggestion that Appellee get a “psych eval” while in jail, Appellee's expressions of regret over Colton's birth, and Appellee's use of profanity. The court of appeals determined that no evidence in the new trial hearing showed whether or how these statements prejudiced Appellee at trial, and without such evidence, Strickland’s prejudice standard was not met. Hradek, 2023 WL 174878, at *––––, 2023 Tex. App. LEXIS 230, at *61. The court concluded that without such a showing, the record did not support a finding that the evidence undermined the reliability of the verdict. Id.

The court next addressed comments related to Appellee's guilt or innocence, including Whitney's implying that Appellee was involved in Colton's death and that Appellee was considering pleading guilty. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *62-63. The court determined that there was no evidence that Whitney's comments about Appellee's role in Colton's death affected the jury. Id. at *––––, 2023 Tex. App. LEXIS 230, at *63. The court acknowledged that plea offers are generally inadmissible and that a defendant's consideration of a guilty plea could prejudice the jury and undermine the verdict. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *63-64. But the court discounted the likelihood of prejudice from this evidence because Appellee maintained her innocence throughout the call, and the jury found her guilty of the lesser offense of reckless injury to a child. Id.

The court of appeals reasoned that the phone call was only 43 minutes in about 20 hours of testimony, and it allowed Appellee to protest her innocence without cross-examination and to support her defense that she was guilty only of reckless omission. Id. at *––––, 2023 Tex. App. LEXIS 230, at *65. The State did not emphasize the evidence from the call. Id. at *65-66. And Appellee's theory of the case supported the jury's verdict; she knew it was unsafe for Colton to sleep in bed with her, but she allowed it anyway. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *66-68.

The court of appeals said there was no tangible evidence of prejudice under Strickland. Hradek, 2023 WL 174878, at *––––, 2023 Tex. App. LEXIS 230, at *68. It said that properly admitted evidence, including the defense theory of accidental death, supported the jury's verdict of reckless injury to a child and therefore disagreed with the trial court's determination that, but for defense counsel's erroneous admission of the jail call, there is a reasonable probability that the outcome of the case would have been different. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *68-69. The court concluded that without evidence of prejudice, the trial court improperly granted a new trial based on ineffective assistance of counsel. Id. at *––––, 2023 Tex. App. LEXIS 230, at *69. The court of appeals reversed the trial court's ruling and affirmed Appellee's conviction. Id.

The dissent emphasized the deferential standard applicable to rulings on new-trial motions. Id. at –––– – ––––, 2023 Tex. App. LEXIS 230, at 73-75 (Palafox, J., dissenting). It disagreed with the State's contention that the jail call was at most minimally prejudicial or was cumulative of other admissible evidence. Id. at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *100-01. It noted that the applicable standard requires viewing the evidence in the light most favorable to the trial court's ruling. Id. at *––––, 2023 Tex. App. LEXIS 230, at *114. It said that under this standard, the record supported the trial court's conclusion that the highly inflammatory and inadmissible comments in the jail call contributed to the prejudicial harm resulting from counsel's deficient representation. Id. It maintained that the prejudicial evidence portrayed Appellee in a negative light, damaged her credibility, and led to a breakdown in the adversarial process. Id. at *––––, 2023 Tex. App. LEXIS 230, at *116 (citing Miller v. State, 548 S.W.3d 497, 499 (Tex. Crim. App. 2018)). The dissent concluded that there was a reasonable probability that, but for defense counsel's erroneous admission of otherwise inadmissible evidence, the jury would have had a reasonable doubt about Appellee's guilt. Hradek, 2023 WL 174878, at *––––, 2023 Tex. App. LEXIS 230, at *117.

We granted Appellee's ground for review:

When four judges have considered whether to properly grant a motion for new trial and two of them have decided that such a motion was properly granted, then that decision cannot be outside the zone of reasonable disagreement. The two-justice majority of the Court of Appeals never explicitly found that the trial court's decision was either arbitrary or unreasonable, and their Opinion failed to give proper deference to the trial court's ruling. When reviewing the Court of Appeals’ decision, it is clear the State was unable to show that trial counsel's decision to order his subordinate attorney to play the entirety of one of, if not, the most damning piece of evidence in the entire trial and admit it into evidence did not undermine confidence in the outcome.

For reasons explained below, we reject Appellee's assertion that an even split among judges might be the measure for an abuse of discretion, but we agree that the court of appeals failed to give proper deference to the trial court's fact findings.

IV. IAC Prejudice Standard of Review

IAC claims have two elements: deficient performance and prejudice. Strickland, 466 U.S. at 687, 104 S.Ct. 2052. The prejudice element is fulfilled by a reasonable probability of a better outcome but for the deficient performance. Id. at 694, 104 S.Ct. 2052. Prejudice requires a reviewing court to examine “the totality of the evidence” heard by the factfinder and determine whether the result of the proceeding is “unreliable because of a breakdown in the adversarial process that our system counts on to produce just results.” Miller, 548 S.W.3d at 499 (quoting Strickland, 466 U.S. at 695-96, 104 S.Ct. 2052).

IAC prejudice is a mixed question of law and fact, and some questions of fact may turn on the credibility and demeanor of the witnesses who testify in the new-trial hearing. Kober v. State, 988 S.W.2d 230, 233 (Tex. Crim. App. 1999). Appellate courts should afford almost total deference to a trial court's determinations of those questions. Id. (citing Guzman v. State, 955 S.W.2d 85, 89 (Tex. Crim. App. 1997)). Mixed questions of law and fact that do not depend on an evaluation of credibility and demeanor may be reviewed de novo. Guzman, 955 S.W.2d at 89. Unless the trial court is in “ ‘an appreciably better position’ than the appellate court to decide the issue, the appellate court may independently determine the issue while affording deference to the trial court's findings on subsidiary factual questions.” Id. (quoting Villarreal v. State, 935 S.W.2d 134, 139 (Tex. Crim. App. 1996) (McCormick, P.J., concurring)).

In reviewing the trial court's grant of Appellee's motion for new trial, the court of appeals cited Johnson v. State, 169 S.W.3d 223, 239 (Tex. Crim. App. 2005) for the proposition that the “ultimate question of whether prejudice existed is reviewed de novo.” Hradek, 2023 WL 174878, at *8, 2023 Tex. App. LEXIS 230, at *21. But the court of appeals failed to include the rest of the standard articulated in Johnson: “While the ultimate question of prejudice under Strickland is to be reviewed de novo, the trial court should be afforded deference on any underlying historical fact determinations.” Johnson, 169 S.W.3d at 239.

V. Analysis

The trial court's prejudice conclusion depended on an evaluation of the credibility and demeanor of the new-trial witnesses and the probable impact of State's Exhibit 51 on Appellee's chances for acquittal. It was in an appreciably better position to make those assessments than an appellate court because it saw and heard the evidence, significant parts of which are ambiguous on the cold record or, as with the 911 call, missing in action.

The trial court's prejudice conclusion is supported by Maesse's opinion testimony that State's Exhibit 51 annihilated the defense, and that opinion, in turn is supported by the exhibit's contents. The unguarded, unfiltered conversation suggested that Appellee's own mother thought she was unserious about her case, untruthful about the cause of Colton's death, and self-centered. It exposed Whitney's suspicions that Appellee hurt Colton and might have mental health issues and revealed Appellee's regret about Colton's birth.

How impactful was State's Exhibit 51? The trial court was best positioned to assess its effect on the chance of acquittal because it heard and saw the trial testimony and its attendant gestures and demonstrations that flesh out the cold record. The most significant and ambiguous passages were about the cause of death. For example, as it appears in the record, Dr. Contin's in-court demonstration of “upside down” is meaningless; it consists of his request to put his feet on the furniture and his testimony that Colton was “upside down like this, with the legs like this (indicating).” Similarly, the grainy, black-and-white copies of the autopsy and scene photos whose significance was highlighted by witnesses’ “like this” commentary hinder our efforts to evaluate the record, but the trial court could see the original photos and the “like this” gestures. In addition, the trial court heard Appellee's 911 call, but we cannot; it was not part of the appellate record and seemingly has been lost during the pendency of the appeal.

Even from the cold record, however, Dr. Contin's testimony was problematic for the prosecution because it was internally inconsistent. He testified to mutually exclusive meanings for “upside down”—face down (head floating free) or face up (reclining). And he undermined his primary, car-seat hypothesis by offering an alternative, “pile of dirty clothing” scenario that would not have accounted for the blanching on the thighs after he had insisted on the car seat's straps as the cause of those marks. He was also inconsistent by dismissing the possibility that the diaper caused the thigh marks but attributing the lividity on the back of Colton's legs to the “very tight” diaper. Moreover, his opinion about cause of death was challenged by Dr. Shuman who pointed out the discrepancies between the lividity pattern and the car seat's configuration and offered different explanations for the observed patterns: the slope of the bed due to Bobby's weight on it and the diaper's leg openings causing circumferential blanching of Colton's thighs.

The court of appeals redeemed the phone call in part on grounds that it allowed Appellee to proclaim her innocence without subjecting her to cross-examination, an argument that the State echoes. But the video recordings of her police interrogations served the same function, and they did so more appealingly. In the videos, Appellee was articulate, polite, and rational, but in the phone call she was profane, angry, whining, grasping.

The court of appeals reasoned that the jury could have believed the State's theory and convicted Appellee of reckless injury even if it disbelieved the car-seat hypothesis because she used cocaine and let Colton sleep in her bed knowing that it was an unsafe sleep position. But even if Appellee's cocaine use was evidence of her recklessness, it was not compelling because there was no evidence that she used it on the night Colton died. The evidence that she used cocaine near the time of his death was admitted only because the defense opened the door to it with its inept cross-examination of one of the detectives. And as a persuasive matter, Appellee's claim that she let Colton sleep in her bed after being advised against it was far less damning than the idea that she left him for hours upside down in his car seat. The former seems common, natural, and nurturing, but the latter is aberrant, cruel, and callous. In short, Appellee's defensive theory was more viable before State's Exhibit 51 was played for the jury.

The court of appeals reasoned that the conviction for the lesser offense of reckless injury instead of intentional or knowing injury defeats any claim of prejudice from the admission of State's Exhibit 51. Hradek, 2023 WL 174878, at *–––– – ––––, 2023 Tex. App. LEXIS 230, at *64-65. But the acquittal for the greater charge testifies more to the State's overreaching with its initial charging decision than it undermines the trial court's prejudice conclusion. There was no evidence of intent to kill, and the case that Appellee knowingly killed Colton was weak, at best, as suggested by the State's fallback request for the recklessness option. The court of appeals also mischaracterized Appellee's defense as a claim “that she was only guilty of reckless omission.” Id. at *23, 2023 Tex. App. LEXIS 230, at *65. That was not her defense. Her defense aimed for acquittal, and the jury was instructed on recklessness over her objection.

The court of appeals did not explicitly address the phone call's prejudicial effect. Instead, it re-evaluated the evidence and concluded that it was sufficient to support the conviction. It reasoned that even without the phone call's inadmissible evidence, the jury could have determined that Appellee was reckless in using cocaine, returning the sleep apnea monitor, and placing Colton in an unsafe sleep position. But sufficiency of the evidence is not the standard for evaluating IAC prejudice or a trial court's grant of a new trial. Rather, prejudice depends on a reasonable probability of a different outcome, and a ruling granting a new trial will be upheld unless it is arbitrary or unreasonable. Strickland, 466 U.S. at 694, 104 S.Ct. 2052 (referencing IAC prejudice standard); Becerra v. State, 685 S.W.3d 120, 127 (Tex. Crim. App. 2024) (referencing standard for reviewing new trial ruling).

The State proposes alternate interpretations of the evidence and says the inadmissible parts of the call were cumulative of other properly admitted evidence. For example, it says that Whitney's comments about the evidence against Appellee recounted evidence from the investigation that was properly admitted at trial, and Whitney telling Appellee to take the matter seriously and tell the truth was cumulative of the detectives asking her about the night Colton died. But consider the source; it is one thing to have a law enforcement officer recount the evidence or interrogate a suspect, but it is another for the suspect's mother to do so. According to the State, Appellee's consideration of pleading guilty was an expression of frustration at facing prosecution for a crime she did not think she committed, and her cussing demonstrated the passion with which she believed in her innocence. But she protested her innocence in the video interviews without cussing and without expressing an interest in pleading guilty. State's Exhibit 51 undermined those more pristine claims of innocence. The State argues that Appellee's expression of regret about Colton's birth would be seen as insincere because she also said she regretted losing him, but it still jeopardized any sympathy the jury might have had for her.

The dissenting opinion has a different take. It argues that much of State's Exhibit 51 was admissible, and removing the admissible parts from consideration changes the prejudice result. Dissenting Op. at –––– (Keller, P.J., dissenting) (citing Lockhart v. Fretwell, 506 U.S. 364, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993)). The trial court found, however, that no part of the recording other than the cocaine excerpt would have been admitted but for the defense request to admit it in its entirety, and the dissent does not show that such a ruling would have been an abuse of discretion. See State v. Mechler, 153 S.W.3d 435, 439 (Tex. Crim. App. 2005) (holding evidentiary rulings rest in the trial court's discretion). Without such a showing, the rest of the recording is pertinent to the prejudice analysis.

The defense's erroneous admission of highly inflammatory, prejudicial evidence had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture. See, e.g., Strickland, 466 U.S. at 695-96, 104 S.Ct. 2052. The recording destroyed Appellee's defense that Colton's death was a SIDS tragedy that could happen to any mother; it portrayed her as a self-centered, possibly mentally-ill, exotic dancer who wished her son had never been born. Whitney's comments on the evidence against Appellee, her admonishments that Appellee should tell the truth and take this seriously, and her angry questions asking Appellee if this was fair to Colton were unlike the medical examiner's testimony or the police questioning of Appellee as part of the investigation. And there was no other evidence showing Appellee's flippant attitude, self-centeredness, or regret that her son was born. Thus, the trial court's assessment of the recording's prejudicial effect was not arbitrary or unreasonable.

VI. Conclusion

There are two plausible interpretations of State's Exhibit 51: Either it was so prejudicial that there is a reasonable likelihood that it extinguished Appellee's chances of acquittal, or it was not so bad. The trial court was better positioned to make that call, and its ruling was within the zone of reasonable disagreement. See Najar v. State, 618 S.W.3d 366, 372 (Tex. Crim. App. 2021) (“[I]f there are ‘at least two’ plausible interpretations of the evidence, it is within the trial court's exclusive purview to decide which interpretation to believe.”); Burch v. State, 541 S.W.3d 816, 820 (Tex. Crim. App. 2017) (“The trial court's ruling is within the ‘zone of reasonable disagreement’ when there are two reasonable views of the evidence.”). We reverse the judgment of the court of appeals and remand to the trial court for further proceedings.

The proper standard for an appellate court to review a trial court's grant of a motion for new trial is an abuse of discretion standard. The bare fact that a trial court may decide a matter differently from an appellate court does not demonstrate an abuse of discretion.¹ Appellate courts must view the evidence in the light most favorable to the ruling, defer to the court's credibility choices, and assume that all reasonable fact findings in support of the ruling have been made.² This abuse of discretion standard applies even when one of the grounds raised in the motion for new trial is ineffective assistance of counsel.³

Reviewing courts must defer to the trial court's role as fact-finder due to its role as the judge of witness credibility, but they apply a de novo review in those situations in which we can isolate a pure question of law.⁴ In Johnson v. State, we explained as much in the context of a motion for new trial based upon ineffective assistance of counsel.⁵ And, in that case, we noted that “while the ultimate question of prejudice under Strickland is to be reviewed de novo, the trial court should be afforded deference on any underlying historical fact determinations.”⁶

But the bottom line in Johnson was far more callipygian than the court of appeals perceived. Indeed, we did not apply a categorically de novo standard when evaluating prejudice.⁷ We deferred to the trial court's credibility determinations regarding how the trial court could have viewed the impact of the alleged deficient performance.⁸ But we also noted that there would not be a reasonable probability that the outcome would change even if we assumed the trial court believed the defendant's version of events.⁹

I do not read Johnson as suggesting that reviewing courts categorically apply a strictly de novo standard or a purely deferential one when evaluating prejudice as part of the trial court's “ruling.” Rather, I believe the proper standard is the one articulated by Presiding Judge Keller in Kober v. State, which she relied upon when she authored Johnson.¹⁰ In Kober, we recognized that “the prejudice prong of Strickland is a mixed question of law and fact, [and] that question often contains subsidiary questions of historical fact, some of which may turn upon the credibility and demeanor of witnesses.”¹¹ Given this understanding, appellate courts must afford “almost total deference” to the trial court's determination of the historical facts and to mixed questions of law and fact that turn on an evaluation of credibility and demeanor, but purely legal questions can be reviewed de novo.¹²

In short, I do not believe the standard is an either-or proposition. To the extent that the court of appeals suggested that the prejudice prong of Strickland should always be reviewed de novo, it erred by failing to defer to the trial court's factual determinations.

By applying a de novo standard on the issue of prejudice, the court of appeals appears to have evaluated both legal and factual determinations by the trial court de novo. This ignored the more nuanced aspect of the standard of review regarding the distinction between factual findings and legal conclusions. Reviewing courts are not expected to defer to a trial court's ruling on a matter of law because the trial court is in no better position to make that determination than the reviewing court.¹³ Failing to correct the court of appeals’ analysis in this regard would likely have resulted in confusion amongst both trial courts and reviewing courts given how challenging it can be to differentiate between findings of fact and conclusions of law. I join the Court's opinion with the understanding that reviewing courts are not required to defer to a trial court's legal determination on the issue of prejudice after viewing the historical facts in the record in a light most favorable to the trial court's ruling.

DISSENTING OPINION

The trial court granted a motion for new trial on the basis of ineffective assistance of counsel. It found that numerous errors prejudiced Appellee. The court of appeals reversed, rejecting all of the ineffective assistance allegations. This Court now says that the court of appeals erred in failing to view the record in the light most favorable to the trial court's ruling. Specifically, this Court concludes that the issue of prejudice depended on an evaluation of the credibility and demeanor of the new-trial witnesses and that the trial court was better positioned than an appellate court to evaluate the issue. The Court then says that, under a proper review, prejudice was shown.

But the trial court made a crucial error of law that makes deferring to it problematic—it concluded that, except for the small portion the State sought to admit, the entire phone call was inadmissible. But in fact, much of the remainder of the phone call was admissible. Under Lockhart v. Fretwell,¹ any admissible evidence should weigh in the State's favor for the purpose of assessing prejudice. Once the admissible evidence is taken into account, it becomes clear that Appellee has not shown prejudice.

In Conclusions 1 and 2, the trial court suggested that, except for the part the State specifically offered (Appellee admitting to using cocaine), the rest of the phone call was inadmissible:

1. The additional portions of the recorded phone call in State's Exhibit 51 contained statements by Defendant and her mother Whitney Hradek that were inadmissible pursuant to Texas Rules of Evidence.

2. The probative value of the additional portions of State's Exhibit 51 admitted by defense counsel were substantially outweighed by the unfair prejudice to the defense and were highly inflammatory.

But the trial court's assessment is incorrect because much of the remainder of the phone call was admissible. It is true that the mother's statements were inadmissible hearsay.² But Appellee's statements constituted party-opponent admissions that would not be barred by the hearsay rule.³ Appellee's references to being a dancer and a stripper were inadmissible character evidence.⁴ The admissibility of her statement about considering a plea of guilty is murkier, because it does not seem to fall squarely within the rule against admitting evidence of plea negotiations,⁵ though perhaps it could be deemed unfairly prejudicial under Rule 403.⁶ But most of Appellee's remaining statements were clearly admissible. They were statements about the offense or her reaction to being prosecuted that were relevant to her mental state at the time of the offense and to consciousness of guilt. These included the statement that she “wished her son was never born” (Finding 22), the statement that she “wished she had never gotten pregnant” (Finding 23), cursing the medical examiner (Finding 26), using cocaine (Finding 28), having anger and mental health issues (Finding 30), and uses of profanity that the trial court did not specify (Finding 32), but which included her saying that the State had nothing on her because she did nothing wrong and her saying that she used cocaine. And as the Court's opinion says, she admitted that the baby slept in the bed next to her, and her own expert acknowledged that it was unsafe for a young baby to sleep in an adult bed.

The trial court's error on the admissibility issue is important because the prejudice inquiry should in fact turn on the admissibility of the evidence and not merely on what the trial court would have done if Appellee had objected to the remainder of the call. In Fretwell, the Supreme Court addressed a situation where the defense attorney failed to lodge an objection that would have been meritorious under a court decision that existed at the time of trial.⁷ The case that supported the objection was later overruled.⁸ An intermediate federal appellate court concluded that prejudice had been shown because the objection would have been sustained under the caselaw that existed at the time.⁹ The Supreme Court, however, concluded that test for prejudice for an ineffective-assistance claim was more nuanced than a simple outcome determination.¹⁰ The Court concluded that a prejudice analysis must also look at whether the challenged conduct had an effect “on the reliability of the trial process.”¹¹ Showing prejudice requires showing that the result was “fundamentally unfair or unreliable.”¹² Because the case that would have supported a defense objection was overruled, there was in fact no error in the trial to object to, and prejudice does not occur when counsel's actions merely “deprive[ ] [Appellee] of the chance to have the state court make an error in his favor.”¹³

So the admissible portions of the phone call weigh in the State's favor, not Appellee's favor, under a prejudice analysis. Appellee characterized the recording of the phone call as “the most damning piece of evidence in the entire trial” but that “damning” nature was due largely to the admissible parts of the call—the various statements that Appellee made about her child and the impending proceedings. The Court says that the trial court's prejudice conclusion is supported by attorney Maesse's opinion testimony that the telephone recording annihilated the defense. But that opinion testimony assumes that the entire remainder of the phone call should be kept out, when much of it was in fact admissible. Although the mother's hearsay statements included the mother berating Appellee for not taking the proceedings seriously, Appellee's own statements would have conveyed the same impression to the jury.

Given that Appellee was convicted of a lesser-included offense rather than the charged offense, and given that much of the phone call was admissible, I would conclude, as the court of appeals did, that Appellee suffered no prejudice.¹⁴

I respectfully dissent.

FOOTNOTES

1.   The recording of the 911 call was not part of the court of appeals's record and has reportedly been lost during the appeal of this case.

2.   The defense attorney opened the door to testimony that Appellee was tested for drug use as part of the investigation, and the test showed she had used cocaine within 48 hours.

3.   State's Exhibit 51 is a four hour long recording consisting of multiple jail calls Appellee made. The call that was admitted at trial and played for the jury begins at 2:56:06 and lasts approximately 43 minutes.

1.   State v. Simpson, 488 S.W.3d 318, 322 (Tex. Crim. App. 2016) (citing State v. Thomas, 428 S.W.3d 99, 103 (Tex. Crim. App. 2014)).

2.   Id.

3.   See, e.g., Okonkwo v. State, 398 S.W.3d 689, 694 (Tex. Crim. App. 2013).

4.   Id.; see also Johnson v. State, 169 S.W.3d 223, 239 (Tex. Crim. App. 2005).

5.   Johnson, 169 S.W.3d at 239.

6.   Id.

7.   Id.

8.   Id. at 240. (“The trial court was not required to believe that the defendant would have testified at trial as to the more exculpatory version of events given at the motion for new trial hearing.”)

9.   Id. (“And even if the trial court believed that the more exculpatory version of events given by appellant at the motion for new trial hearing was in fact the version he would have given at trial, there still would not be a reasonable probability that the outcome would change.”).

10.   Kober v. State, 988 S.W.2d 230 (Tex. Crim. App. 1999).

11.   Id. at 233.

12.   Id.

13.   Guzman v. State, 955 S.W.2d 85, 87 (Tex. Crim. App. 1997) (noting that the amount of deference a reviewing court affords to a trial court's ruling on a ‘mixed question of law and fact’ is often determined by which judicial actor is in a better position to decide the issue) (citing Miller v. Fenton, 474 U.S. 104, 106, 106 S.Ct. 445, 88 L.Ed.2d 405 (1985)).

1.   506 U.S. 364, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993).

2.   See Tex. R. Evid. 801, 802.

3.   See id. 801(e)(2).

4.   See id. 404(b).

5.   See id. 408.

6.   See id. 403.

7.   506 U.S. at 366, 113 S.Ct. 838.

8.   Id.

9.   Id. at 368, 113 S.Ct. 838.

10.   Id. at 369, 113 S.Ct. 838 (“Thus, an analysis focusing solely on mere outcome determination, without attention to whether the result of the proceeding was fundamentally unfair or unreliable, is defective.”).

11.   Id.

12.   Id.

13.   Id. at 371, 113 S.Ct. 838.

14.   In Conclusion 16, the trial court painted the evidence of the culpable mental states of intent and knowledge (for the charged offense) as weak:Although a directed verdict was not granted, the State's evidence of conduct by omission and the requisite state of mind pled in the indictment of intentional and knowing was marginal.But Appellee was acquitted of the charged offense with those culpable mental states. She was, instead, convicted of the lesser-included offense of recklessly causing serious bodily injury. The findings and conclusions do not comment on the strength of the evidence to show recklessness.

Keel, J., delivered the opinion of the Court in which Hervey, Richardson, Yeary, Newell, Walker, and Slaughter, JJ., joined.

Newell, J., filed a concurring opinion in which Walker, J., joined. Keller, P.J., filed a dissenting opinion. McClure, J., concurred.