*503Defendant Jeffrey G. has been confined to a state hospital for over 20 years. In 2015, he petitioned for transfer from the hospital to a conditional release program. At the hearing on his petition, defendant and a psychologist who had examined him testified regarding his likelihood of success in the unstructured environment of the conditional release program. In turn, the prosecution presented the testimony of three expert witnesses to contest his readiness. Under the law applicable at the time *504of the hearing, an expert witness was permitted to testify with respect to the hearsay evidence on which the expert based his or her opinion, regardless of whether there was competent evidence in the record to support that testimony. On the basis of such testimony, the trial court denied defendant's petition. A short time after the hearing, the Supreme Court issued People v. Sanchez (2016) 63 Cal.4th 665, 204 Cal.Rptr.3d 102, 374 P.3d 320 ( Sanchez ), which substantially limited expert testimony with respect to case-specific hearsay evidence. Had the hearing been conducted under Sanchez , it is certain that at least some of the testimony of the prosecution's experts would have been excluded.
We agree with defendant that we must apply the rule of Sanchez and reverse the trial court's denial of his petition for conditional release. Defendant's testimony and the testimony of his expert provided independent evidence to support some of the otherwise-hearsay testimony by the prosecution's experts, but a significant portion of their testimony was not anticipated by defendant's evidence. In part because the trial court found defendant's petition to present a close case, we conclude it is reasonably probable that the trial court would have granted defendant's petition in the absence of the expert testimony rendered inadmissible by Sanchez .
I. BACKGROUND
Defendant was committed to the State Department of State Hospitals (DSH) in 1990, after he was found not guilty by reason of insanity of a violent crime, and he has been confined to a hospital for most of the subsequent time. In May 2016, he agreed to extend his commitment for an additional two years, until June 2018. ( Pen. Code, § 1026.5, subd. (b).)
Notwithstanding his agreement, defendant petitioned to be transferred from a state hospital to a conditional release program. ( Pen. Code, § 1026.2, subd. (a).) At a bench trial on his petition, defendant testified and called two other witnesses, a consulting psychologist and a hospital employee who had observed defendant's conduct during his commitment. The prosecution called three witnesses, all psychologists or psychiatrists, two of whom had recent supervisory responsibility over defendant's confinement for periods of six months each.
Defendant is diagnosed with schizoaffective disorder, bipolar type, and narcissistic personality disorder. His symptoms have included aural and visual illusions and mental delusions. These are controlled by treatment, but they can be triggered by drug or alcohol use or the failure to maintain his medication regimen. Defendant has not suffered psychotic symptoms during his time in his current institution, a period of over 10 years. In testifying, he *505acknowledged his illness and recognized the necessity of maintaining sobriety and adhering to the medication regimen in order *91to avoid an onset of symptoms. He committed to continue both if conditionally released.
There was general agreement at trial that defendant, while in confinement, had complied with his medication regime and was free of psychotic symptoms. His three prior attempts to live outside the hospital under the conditional release program (CONREP), however, had ended in failure.1 His first attempt, in 1991, lasted only 18 days. The second attempt was more successful, lasting around two years before defendant violated the terms of the program by going absent without leave in 1996. The most recent attempt, in 2005, lasted only 12 days because an apparently ill-advised change in defendant's medication caused him to decompensate.
A primary focus of the testimony was defendant's relatively recent conduct in the hospital. Within the prior year, defendant had broken hospital rules by selling goods to other patients and having sexual contact with another patient in a public area of the hospital.2 Within the prior six months, he had a run-in with a staff person at the hospital. When the staff person told defendant to stop watering a plant before he killed it, defendant responded that he would kill the staff person before he killed the plant. There was also inconsistent testimony about defendant's attendance at treatment groups, which is discussed in more detail below.
Defendant's expert did not believe that these incidents disqualified him from conditional release. As the expert pointed out, the primary threat to defendant's mental health was substance abuse, which led to reduced medication compliance and a reappearance of his symptoms. Although drugs and alcohol were illicitly available in the hospital, the expert noted, defendant had avoided the temptation to use them for the duration of his confinement, suggesting to the expert that he could do the same on conditional release. The prosecution's experts, in turn, viewed defendant as a "moderate risk of danger," particularly if presented with stressful situations, and concluded on the basis of his rule violations, his interpersonal relationships, and his purportedly spotty attendance at treatment programs that he was not ready for conditional release.
The trial judge found the decision difficult, noting he was "very sympathetic to the position [defendant is] in." The court believed defendant *506"doesn't present much of a danger right now," but it recognized he would present a danger if he decompensated upon conditional release. In denying the petition, the court found decisive defendant's failure on three prior occasions to succeed upon conditional release, his failure to follow hospital rules, despite suffering discipline as a result of the violations, and his failure to attend treatment programs consistently. As to the latter, the court recognized that such programs could be "amazingly boring," but it believed that enduring the programs was critical because they would assist defendant in avoiding the temptations present in the world outside the hospital. Until defendant demonstrated that he could cope with reduced *92structure, the court concluded, he was not ready for conditional release.
II. DISCUSSION
Defendant's hearing occurred less than two weeks prior to the Supreme Court's issuance of Sanchez , which significantly changed the rules governing testimony by expert witnesses about the hearsay upon which they relied in forming their opinions.3 Under the law prevailing at the time of defendant's hearing, an expert was permitted to testify relatively freely about the content of hearsay evidence relating to the circumstances of the case at hand, if the evidence constituted a basis for his or her opinion. (E.g., People v. Montiel (1993) 5 Cal.4th 877, 919, 21 Cal.Rptr.2d 705, 855 P.2d 1277 ( Montiel ).) Sanchez now bars such testimony unless there is direct evidence of the matter discussed or the hearsay evidence has been admitted under an appropriate exception. ( Sanchez , supra , 63 Cal.4th at pp. 670-671, 204 Cal.Rptr.3d 102, 374 P.3d 320 ; see People v. Stamps (2016) 3 Cal.App.5th 988, 996, 207 Cal.Rptr.3d 828 [summarizing the holding of Sanchez ].) At defendant's hearing, the testimony of the prosecution experts was peppered with hearsay-based references to his history and his conduct during confinement. This testimony was unobjectionable under the law at the time of the hearing, but it would have been improper under Sanchez in the absence of independent evidence in the record regarding those incidents.
As it happened, the record did contain evidence of many of these incidents. Unlike at a typical criminal proceeding, in which the prosecution's witnesses testify first, defendant and his witnesses opened the trial. Both he and his expert discussed these various incidents, and their testimony was admitted without objection. Accordingly, although the prosecution experts' testimony was based on hearsay, there was evidence in the record to support much of this otherwise improper testimony under Sanchez . The primary issues in this appeal are therefore (1) the exact nature of the case-specific evidence to *507which the prosecution experts testified without evidentiary support and (2) whether this testimony was prejudicial to defendant.
A. Preliminary Issues
First, we conclude the rule of Sanchez applies to this appeal, rather than the law governing the admission of hearsay testimony by experts prevailing at the time of defendant's hearing. In general, "[a] new rule for the conduct of criminal prosecutions is applied retroactively to all cases pending on appeal or not yet final, even if the new rule presents a 'clear break' with the past." ( People v. Song (2004) 124 Cal.App.4th 973, 982, 22 Cal.Rptr.3d 118.) While this general principle was developed specifically with respect to new rules of law announced by the United States Supreme Court ( People v. Cage (2007) 40 Cal.4th 965, 974, fn. 4, 56 Cal.Rptr.3d 789, 155 P.3d 205 ), California similarly recognizes that " '[a]s a rule, judicial decisions apply "retroactively." [Citation.] Indeed, a legal system based on precedent has a built-in presumption of retroactivity.' " ( People v. Guerra (1984) 37 Cal.3d 385, 399, 208 Cal.Rptr. 162, 690 P.2d 635.) Because Sanchez does not "expand criminal liability or enhance punishment," we apply it here. ( Correa v. Superior Court (2002) 27 Cal.4th 444, 463, fn. 5, 117 Cal.Rptr.2d 27, 40 P.3d 739 ; see *93People v. Welch (1999) 20 Cal.4th 701, 732, fn. 4, 85 Cal.Rptr.2d 203, 976 P.2d 754 ["In the criminal law, similarly, the rule is that 'convictions should ordinarily be tested on appeal under the law then applicable, not the law prevailing at the time of trial.' "].)
Second, defendant's trial counsel did not forfeit this legal claim by failing to object to the prosecution experts' testimony on this ground at the hearing. There is little doubt that objecting would have been futile. Because the experts' testimony was unobjectionable under the law prevailing at the time of the hearing, any objection would presumably have been overruled.4 " '[R]eviewing courts have traditionally excused parties for failing to raise an issue at trial where an objection would have been futile or wholly unsupported by substantive law then in existence.' " ( People v. Brooks (2017) 3 Cal.5th 1, 92, 219 Cal.Rptr.3d 331, 396 P.3d 480.) In addition, parties are generally not required to anticipate rulings that significantly change the prevailing law. Our Supreme Court has consistently entertained claims premised on Crawford v. Washington , supra , 541 U.S. 36, 124 S.Ct. 1354, despite a defendant's failure to object on that ground, if the hearing occurred prior to Crawford 's *508issuance. As the court explained in People v. Banks (2014) 59 Cal.4th 1113, 176 Cal.Rptr.3d 185, 331 P.3d 1206, overruled on other grounds in People v. Scott (2015) 61 Cal.4th 363, 391, footnote 3, 188 Cal.Rptr.3d 328, 349 P.3d 1028, "[b]ecause Crawford 'was a dramatic departure from prior confrontation clause case law,' a defendant's failure to raise a Crawford claim in a pre- Crawford trial 'is excusable because defense counsel could not reasonably have been expected to anticipate this change in the law.' " ( Banks , at p. 1167, 176 Cal.Rptr.3d 185, 331 P.3d 1206 ; see similarly People v. Harris (2013) 57 Cal.4th 804, 840, 161 Cal.Rptr.3d 364, 306 P.3d 1195 ; People v. Kitchens (1956) 46 Cal.2d 260, 263, 294 P.2d 17 ["A contrary holding would place an unreasonable burden on defendants to anticipate unforeseen changes in the law and encourage fruitless objections in other situations where defendants might hope that an established rule of evidence would be changed on appeal. Moreover, in view of the decisions of this court ..., an objection would have been futile, and 'The law neither does nor requires idle acts.' "].) While Sanchez might not have been as dramatic a departure from prior law as Crawford , it certainly worked a significant change.
In arguing for forfeiture, the Attorney General points out that the basis for much of the testimony of the prosecution's experts was in all likelihood admissible hearsay, including hospital records and defendant's own statements in the course of interviews by the experts. Had an appropriate objection been made, it is argued, the prosecution would have laid an evidentiary foundation for the experts' testimony. While the failure to provide an opportunity to cure would argue in favor of forfeiture under ordinary circumstances, it fails to overcome the futility concern present here. Even if defense counsel had interposed appropriate hearsay objections, the objections would undoubtedly have been resisted by the prosecution and overruled by the *94court, which would have left the experts' testimony unchanged and lacking the foundation required by Sanchez . Only if the trial court had refused to follow applicable precedent would the prosecution have been forced to lay a Sanchez -appropriate foundation.
The Attorney General also argues that if trial counsel had paid more attention to our Supreme Court's docket and recent developments in the law governing experts from other jurisdictions, counsel could have anticipated that Sanchez would one day issue and change the law. We decline to require such prescience.5
*509B. Application of Sanchez
In Sanchez , an expert witness testified about the defendant's gang affiliation.6 ( Sanchez , supra , 63 Cal.4th at p. 670, 204 Cal.Rptr.3d 102, 374 P.3d 320.) As the court explained, expert witnesses are provided greater latitude than lay witnesses in presenting hearsay testimony, particularly with respect to generalized knowledge in their area of expertise. ( Id. at pp. 675-676, 204 Cal.Rptr.3d 102, 374 P.3d 320.) Yet, the court noted, experts had "traditionally" been precluded from testifying to "case-specific" facts, defined as "those relating to the particular events and participants alleged to have been involved in the case being tried," since experts rarely have first-hand knowledge of the events underlying the cases in which they testify. ( Id. at p. 676, 204 Cal.Rptr.3d 102, 374 P.3d 320.) Instead, in former times experts would be asked to apply their generalized knowledge to the case-specific facts through the use of appropriate hypothetical questions, premised on matters established through independent competent evidence. ( Id. at pp. 676-677, 204 Cal.Rptr.3d 102, 374 P.3d 320.) More recently, however, "the line between [expert testimony about general background knowledge and case-specific facts] has ... become blurred" ( id. at p. 678, 204 Cal.Rptr.3d 102, 374 P.3d 320 ), as exemplified by Montiel , in which the court allowed expert testimony about case-specific facts under a limiting jury instruction. The holding was premised on the conclusion that the expert's testimony was not offered for its truth and therefore did not constitute hearsay. ( Montiel , supra , 5 Cal.4th 877, 919, 21 Cal.Rptr.2d 705, 855 P.2d 1277 ; see Sanchez , at p. 678, 204 Cal.Rptr.3d 102, 374 P.3d 320.)
In Sanchez , the court rejected the legal fiction that an expert's testimony about case-specific facts is not offered for its truth. Finding support in a concurring opinion of Justice Clarence Thomas in Williams v. Illinois (2012) 567 U.S. 50, 108 and footnote 3, 132 S.Ct. 2221, 183 L.Ed.2d 89, the court recognized that the validity of an expert's opinion depends on the assumption that the hearsay underlying the opinion is true, since "[i]f the hearsay that the expert relies on and treats as true is not true, an important basis for the opinion is lacking." ( Sanchez , supra , 63 Cal.4th at pp. 682-683, 204 Cal.Rptr.3d 102, 374 P.3d 320.) If the underlying hearsay is not true, the opinion is rendered irrelevant to the case at hand. In acknowledgment of this conclusion, the court held, the law regarding an expert's use of hearsay must *95be changed. ( Ibid. ) "Any expert may still rely on hearsay in forming an opinion, and may tell the jury in general terms that he did so. Because the jury must independently evaluate the probative value of an expert's testimony, Evidence Code section 802 properly allows an expert to relate generally the kind and source of the 'matter' upon which his opinion rests.... [¶] What an expert cannot do is relate as true case-specific *510facts asserted in hearsay statements, unless they are independently proven by competent evidence or are covered by a hearsay exception." ( Id. at pp. 685-686, 204 Cal.Rptr.3d 102, 374 P.3d 320.) "Like any other hearsay evidence, [case-specific evidence considered by an expert] must be properly admitted through an applicable hearsay exception. Alternatively, the evidence can be admitted through an appropriate witness and the expert may assume its truth in a properly worded hypothetical question in the traditional manner." ( Id. at p. 684, 204 Cal.Rptr.3d 102, 374 P.3d 320 fn. omitted.)
We evaluate prejudice resulting from the allowance of expert testimony in violation of Sanchez under the standard of People v. Watson (1956) 46 Cal.2d 818, 299 P.2d 243, which requires reversal if "it is reasonably probable that a result more favorable to the appealing party would have been reached in the absence of the error." ( Id. at p. 836, 299 P.2d 243 ; see People v. Ochoa (2017) 7 Cal.App.5th 575, 589, 212 Cal.Rptr.3d 703.)
The parties to this proceeding understandably failed to anticipate the change in the law created by Sanchez . In his opening brief, defendant presents a long list of case-specific matters to which the prosecution's witnesses testified, all of which are improper under Sanchez without independent evidentiary support in the record. The Attorney General counters with a demonstration that evidence of many of these matters was introduced into the record through the testimony of defendant's witnesses. If prior unobjected testimony supported the prosecution experts' case-specific testimony, the testimony was not objectionable under Sanchez . We have compared the parties' lists and examined the trial transcript to determine the evidentiary support for the testimony of the prosecution's witnesses. We present our conclusions without recounting the details of our examination.7
The prosecution experts' testimony about many of the specific events during defendant's detention was supported by evidence in the record. For example, both defendant and his expert testified about his sexual contact with another patient in a public area, his resale of goods to obtain money, his three prior failures at conditional release, his purported threat to a staff person, an altercation with another patient in 2012, and his long history of detention in a treatment facility. The prosecution experts, however, sometimes testified to specific details that were not recounted by defendant's witnesses or characterized the incidents in an unsupported way. For example, one expert testified that defendant had a "lengthy history" of rule violations, but the record *511reveals only a few isolated violations over the past two or three years. *96There was also testimony that defendant was moved briefly to a discharge unit in 2013 before he "decompensated," but there is no evidence in the record of any decompensation since 2005. More significantly, there is little or no evidence in the record to support the prosecution experts' characterizations of defendant's interpersonal relations. The experts testified that defendant had difficult personal relationships, leading him to react with hostility and aggression. Beyond the specific incidents mentioned above, there is little evidence in the record about defendant's personal relationships, let alone of aggression and hostility toward others. Defendant was similarly said to have a history of being too aggressive and argumentative in getting his needs met, but, again, nothing in the record supports this characterization by the prosecution experts. Finally, and most importantly, there was no other testimony to support the prosecution experts' testimony about defendant's inconsistent attendance at treatment programs. Although the prosecution witnesses asserted that defendant attended less than 85 percent of his treatment groups in recent months, had particularly low attendance in a substance abuse treatment group, and stopped attending treatment groups altogether at some point, the defense witnesses testified that defendant's attendance at treatment groups met or exceeded 85 percent, which is considered an acceptable level by CONREP.
We conclude that introduction of the unsupported testimony described above was prejudicial. The trial court found this to be a close call. As the court noted at the outset of its oral decision, "I don't find this a very easy decision on either side, and I've been on both sides of it and I may be on a different side of it by the time I articulate it." For this reason, relatively small changes in the record could be important. Here, the trial court found three factors critical to its conclusion: defendant's prior failures on conditional release, his recent rule violations, and his spotty attendance at treatment programs. One of those factors, defendant's purported irregular attendance, was unsupported by evidence in the record. Under the rule of Sanchez , the trial court would have known only that defendant's attendance at treatment programs was adequate. Given the close nature of the court's decision, it is "reasonably probable" that the court would have found defendant suitable for release, notwithstanding his prior failures and rule violations, if it concluded he had worked diligently at treatment. Adding to our conviction is other unfavorable evidence that would have been excluded under Sanchez , such as an expert's unsupported characterization of defendant's purported "lengthy history" of rule violations, his purported decompensation in 2013, and his hostile and aggressive interpersonal relations. Because this evidence reinforced the trial court's decision, its absence would have made a different decision more likely.
*512III. DISPOSITION
The trial court's order denying defendant's petition for conditional release under Penal Code section 1026.2 is reversed, and the matter is remanded for a new hearing on the petition.
We concur:
Humes, P.J.
Banke, J.
According to DSH's Web site, CONREP is the conditional release program operated by the department for persons found not guilty by reason of insanity, incompetent to stand trial, or to be mentally disordered offenders. Its goal "is to ensure public protection in California communities while providing an effective and standardized outpatient treatment system." (See < http://www.dsh.ca.gov/Forensics/CONREP.aspx> [as of July 13, 2017].)
While sexual relations between patients is permitted, it must occur in a private place.
Sanchez also addresses issues raised by Crawford v. Washington (2004) 541 U.S. 36, 124 S.Ct. 1354, 158 L.Ed.2d 177. Because those issues are not pertinent here, we have not discussed this aspect of Sanchez .
There is no dispute that Sanchez materially changed the law governing expert testimony in effect at the time of the hearing. The Sanchez court expressly disapproved six prior Supreme Court decisions, noting, in particular, "We also disapprove People v. Gardeley [ (1996) ] 14 Cal.4th 605 [59 Cal.Rptr.2d 356, 927 P.2d 713], to the extent it suggested an expert may properly testify regarding case-specific out-of-court statements without satisfying hearsay rules." (Sanchez, supra, 63 Cal.4th at p. 686, fn. 13, 204 Cal.Rptr.3d 102, 374 P.3d 320.)
Because we find the Sanchez issue preserved, we need not address defendant's claim of ineffective assistance of counsel resulting from the failure to object on hearsay grounds.
Sanchez was a criminal case, but the court did not limit the application of its expert testimony ruling to criminal cases. (See People ex rel. Reisig v. Acuna (2017) 9 Cal.App.5th 1, 34, 214 Cal.Rptr.3d 781 [Sanchez applies to civil cases]; People v. Burroughs (2016) 6 Cal.App.5th 378, 383, 211 Cal.Rptr.3d 656 [applying Sanchez in the context of a sexually violent predator hearing].)
Defendant contends we should not consider the hearsay-based testimony by his expert to constitute independent evidence because the expert "discussed certain incidents without asserting they were true." We do not accept the premise of the argument. None of the experts who testified were in a position to "assert" the hearsay evidence to be true, since they did not have direct knowledge of it. All of the experts assumed it to be true and based their opinions on that assumption. Defendant's expert's testimony was no different. He assumed the evidence to be true yet found it no bar to defendant's conditional release.