The defendant, Paul Poitras, appeals his convictions on one count of rape of his stepdaughter, a child under the age of sixteen (G. L. c. 265, § 22A); one count of assault on the same child with intent to commit rape (G. L. c. 265, § 24B); and one count of indecent assault and battery on the same child (G. L. c. 265, § 13J).1 Poitras’s chief arguments on appeal concern the admission of certain testimony by a Commonwealth expert on child sexual abuse and the application of stage two of the Bishop/Fuller procedure for privileged records. See Commonwealth v. Bishop, 416 Mass. 169, 181 (1993); Commonwealth v. Fuller, 423 Mass. 216, 225-226 (1996). Poitras has also appealed the denial without hearing of his postconviction motion for new trial, alternatively based on alleged newly discovered evidence or ineffective assistance of counsel. We are constrained to reverse Poitras’s convictions because of the clear error of law committed by the trial judge in allowing the expert to testify, over objection, to the characteristics of a child abuser, error that injuriously affected Poitras’s substantial rights.
We need not rehearse the sordid facts on which Poitras’s convictions were based. It is sufficient to note that they reflected many instances of sexual abuse against his stepdaughter, beginning when the complainant was seven and continuing until she was thirteen. The complainant did not reveal the abuse until three years after the last alleged act, there were no eyewitnesses to any abuse, there was no physical evidence of abuse, and the Commonwealth’s case turned entirely on the complainant’s credibility.
*693Against this background of a far from overwhelming Commonwealth case, the prosecution introduced the testimony of a licensed psychologist as an expert witness on the sexual abuse of children, including the general characteristics of sexually abused children.2 The bulk of the expert’s testimony dealt with the typical symptoms and behavioral characteristics of sexually abused children. Contrary to Poitras’s contentions that such portions of the testimony amounted to improper “vouching” for the credibility of the complainant and impermissible “mirroring” of the underlying facts about the complainant’s behavior, that portion of the expert’s testimony (including information about typical patterns of disclosure of abuse by such children) was properly admissible, see Commonwealth v. Dockham, 405 Mass. 618, 628-630 (1989); Commonwealth v. Colon, 49 Mass. App. Ct. 289, 291-293 (2000); it neither explicitly or implicitly linked the expert’s opinions to the complainant in this case, nor directly or indirectly compared the behaviors testified to with those of the complainant.3 Compare and contrast Commonwealth v. Ianello, 401 Mass. 197, 201-202 (1987); Commonwealth v. Montanino, 409 Mass. 500, 504-505 (1991); Commonwealth v. Federico, 425 Mass. 844, 848-851 (1997); Commonwealth v. Trowbridge, 36 Mass. App. Ct. 734, 743 (1994), S.C., 419 Mass. 750, 759-760 (1995); Commonwealth v. Rather, 37 Mass. App. Ct. 140, 148-149 (1994).
The expert fatally crossed the line, however, between proper “abused child” profile opinion and impermissible profile opinion when she testified at length, over twice-registered *694defense objection,4 about the typical attributes and characteristics of those most likely to abuse children. (That portion of the expert’s testimony is set out in the Appendix following this opinion, with emphases added.) That testimony amounted to a description of a child abuser profile and virtually “mirrored” prior prosecution evidence about Poitras that the jury had heard from the complainant and several other witnesses.5 Cf. Commonwealth v. Tanner, 45 Mass. App. Ct. 576, 581 (1998) (criticizing expert testimony “presented in conclusory form . . . [that] amounts to little more than vouching for the Commonwealth’s position”).
The admission of such testimony that effectively provides a profile of typical abusers or abuse situations has been condemned by the appellate courts of this Commonwealth as error, especially when the particular characteristics the expert testified to could be taken by the jury to identify the defendant as fitting the profile and therefore more likely than not to have committed the crime. See Commonwealth v. Day, 409 Mass. 719, 723-725 (1991); Commonwealth, v. LaCaprucia, 41 Mass. App. Ct. 496, 497-500 (1996); Commonwealth v. Roche, 44 Mass. App. Ct. 372, 380 (1998). Compare Commonwealth v. Goetzendanner, 42 Mass. App. Ct. 637, 644 (1997) (judge’s error in not striking expert’s batterer profile testimony deemed nonprejudicial because it was confined to a “single, fleeting reference” of a few words). Its admission “is inherently prejudicial to the defendant,” Commonwealth v. Day, supra at 723, and cannot be deemed harmless error when the Commonwealth’s case against the defendant is not strong, because of the reasonable possibil*695ity that it contributed to the jury’s conclusion that the defendant was guilty. See id. at 725-726; Commonwealth v. Federico, 425 Mass. at 852-853.
The harm is particularly pronounced when, as here, the issue of guilt ultimately turned on the complainant’s credibility, see id. at 852-853; Commonwealth v. LaCaprucia, supra at 502, and the prosecutor, in closing argument, reminded the jury of the very testimony by the expert that provided profile characteristics matching those of the defendant. See id. at 501-502 (“[t]he prosecutor’s remarks suggested to the jury that the defendant sexually abused his children simply because he fit the mold” described in the impermissible expert testimony). In the end, we cannot say with fair assurance that the jury would have convicted Poitras of even the minority of the charges against him (see note 1, supra) in the absence of the error in admitting the profile testimony, see Commonwealth v. Peruzzi, 15 Mass. App. Ct. 437, 445 (1983), and those convictions must, accordingly, be reversed.
Poitras’s remaining arguments, insofar as they may implicate evidentiary issues arising at a new trial, need only be briefly addressed:
1. Poitras’s challenge to a motion judge’s discretionary refusal to make an in camera examination of the complainant’s privileged counseling records, because (the judge concluded) Poitras had failed to make the requisite “relevance” showing at stage two of the applicable protocol to pierce the privilege, is without merit.6 The theories put forth by defense counsel to warrant such an examination were entirely speculative,7 lacking in any concrete or specific factual support, and far from the *696showing required by Commonwealth v. Fuller, 423 Mass. at 225-226. Moreover, information relevant to the articulated theories appears to have been otherwise available in nonprivileged documents, such as the complainant’s school records made available in discovery. See Commonwealth v. Neumyer, 432 Mass. 23, 32 (2000). Poitras has failed to demonstrate abuse of discretion on the part of the motion judge with respect to the complainant’s counseling records. See Commonwealth v. Pare, 43 Mass. App. Ct. 566, 572 (1997), S.C., 427 Mass. 427 (1998). See generally Commonwealth v. Pelosi, 55 Mass. App. Ct. 390 (2002).
2. We agree with the Commonwealth that the trial judge did not abuse his broad discretion in denying Poitras’s motions for a mistrial based upon a witness’s reference to allegations of abuse by Poitras against one of the complainant’s sisters. The judge’s explicit striking of the challenged testimony and twice-repeated curative instructions to the jury regarding it were sufficient to remove any prejudicial taint. A second testimonial reference claimed to constitute prejudicial uncharged misconduct by Poitras (the complainant’s statement on redirect examination that the reason for her disclosure of Poitras’s abuse was her “concern” and fear for her younger sister because “she spent a lot of time [alone at Poitras’s home]”) was, as the Commonwealth argues, permissible rehabilitation in light of the rigorous cross-examination of the complainant regarding her delay in disclosure. See Commonwealth v. Errington, 390 Mass. 875, 880-882 (1984).8
3. We also agree with the Commonwealth that Poitras’s motion for a new trial was properly denied (by the trial judge) without hearing. The several miscellaneous pieces of evidence offered in support of the motion cannot be characterized as *697newly discovered, as they were all known to (or reasonably knowable by) Poitras or his counsel at the time of trial, and, indeed, to some extent were introduced at trial. Nor does Poitras establish that failure by his counsel to have utilized that evidence constituted ineffective assistance of counsel; counsel’s conduct sufficiently reflected strategic decisions on her part, and her use of such essentially collateral or merely impeaching material could not have provided Poitras with a substantial ground of defense.
In conclusion, the judgments are reversed and the verdicts thereon set aside.
So ordered.
Appendix.
Prosecutor: “Doctor, do you have an opinion on whether there are common dynamics for children to come to be abused?”
Witness: “I do.”
Prosecutor: “And what is that opinion?”
Witness: “My opinion is that certain factors have to exist in order for child sexual abuse to occur. First of all, there has to be an individual with a motivation to abuse a child [emphasis added]. That motivation can be based on some —”
Defense Counsel: “Objection.”
The Court: “Overruled. Please continue.”
Prosecutor: “Please continue.”
Witness: “That motivation may have to do with a variety of factors, someone who primarily looks to children for sexual gratification, someone who as a result of certain circumstances in their life doesn’t have access to a partner and looks to a child for some kind of gratification. Inevitably, an abuse situation has to involve a significant power differential between a child and an alleged perpetrator, whereby virtue of a greater amount of power, of authority, of sophistication, an adult can enlist a child in activities that a child either doesn’t know whether they’re right or wrong, or doesn’t really have any ability to *698effectively resist. [Emphasis added.] There are certain general inhibitors that keep most kids safe, and those would have to be overcome as well both in terms of some sort of internalized sense that engaging a child in sexual abuse is wrong. That kind of inhibitor can be overwhelmed by substance abuse, by some kind of psychological disturbance, by some, you know, rationalization that this is not such a terrible thing for a child. The external inhibitors tend to be things like parental supervision, a close attachment relationship with a child. When those things don’t exist, a child may also be significantly at risk.”
Prosecutor: “Doctor, I realize you answered some of my next question, but can you describe in general how a child or adolescent becomes enlisted or engaged in the process of sexual abuse?”
Witness: “Most generally, a child is engaged in the process of sexual abuse by a known adult who has some sort of position of authority, is possibly a caregiver, has some kind of access and opportunity to engage a child. There are certain kinds of risk factors that may make children more vulnerable, among them being a disrupted attachment relationship with a parent, whereby a parent may not be as attentive to a child, may not be so aware of what is going on in a child’s life. There may also be some kind of particular vulnerability that a child has. Some kids have some sort of a disability that make[s] it more difficult for them to let people know if there is something distressing them. But generally what happens is an adult with access and an opportunity to have contact with a child engages a child in some kind of an activity that they don’t know how to consent to, have no ability to either consent to or refuse consent to.” (Emphasis added.)
Prosecutor: “Doctor, based on your training and experience, what dynamics play a role in the case of interfamilial [sz'c] sexual abuse?”
Witness: “Interfamilial [szc] sexual abuse can be a particularly painful and difficult situation for a child because the perpetrator is —”
Defense Counsel: “Objection.”
The Court: “Overruled.”
Witness: “Because in a situation in which abuse is going on within a family, an abuser is likely to have much more access to a child and much more opportunity to molest. Moreover, being molested within your own family means that there is a betrayal of what children are supposed to be able to believe about the context of the family and about parental relationships, that this is a place where they will be safe. This is a place where they will be taken care of. This is not a place where they are in danger. Moreover, someone within the family has *699already some land of a relationship with this child and is in a better position to engage the child. It is not unusual for any sexual abuse relationship to begin with the development of a non-sexual relationship. Caregivers and family members have a lot of authority over children. They are able to compel a child’s compliance or obedience without using extraordinary measures.” (Emphases added.)
Toitras was also convicted of one count of dissemination of pornography to a minor, G. L. c. 272, § 28, and acquitted of seven additional counts: three counts of rape of a child under the age of sixteen; three counts of indecent assault and battery on a child under the age of fourteen; and one count of indecent assault and battery. The Commonwealth has conceded error as to the pornography dissemination count, a concession we have (as we must) independently analyzed and confirmed as appropriate in light of the defense provided by G. L. c. 272, § 28 (for a defendant in a parental or guardianship relationship with the minor). See Commonwealth v. Clark, 23 Mass. App. Ct. 375, 379-380 (1987); Commonwealth v. McClary, 33 Mass. App. Ct. 678, 686 n.6 (1992), cert. denied, 510 U.S. 975 (1993).
In response to the Commonwealth’s pretrial notification of its intent to offer such expert testimony, defense counsel filed a motion in limine to exclude or limit it and for a voir dire of the proposed expert. After hearing, the judge denied the motion. Nothing in the record indicates that the expert was proffered to testify as to the characteristics of child abusers in addition to the characteristics of abused children.
The expert never referred to the complainant during her description of abused children’s characteristics. The expert made it clear (in both direct and cross-examination) that she had never met or evaluated the complainant, knew nothing of her history or her family dynamics, had never reviewed any reports relating to the case, and based her opinions solely on her training, experience, research, and familiarity with the literature in the area of sexual abuse.
Although the form of each was the single, inexplicit word, “Objection,” the Commonwealth makes no argument that Poitras’s objections did not adequately save the issue for appeal (but cf. Commonwealth v. Perryman, ante 187, 192-193 [2002]), or that the appropriate standard of review for the challenged testimony is other than the “prejudicial error” test for properly preserved trial errors. See Commonwealth v. Federico, 425 Mass. at 852-853.
The match between the expert’s testimony and the instant situation was striking. Poitras was a “known adult.” He was not dating anyone at the time of the abuse and therefore inferentially was without access to a partner. He was in a position of authority and was a caregiver who had access and opportunity to engage the complainant. There was manifestly a significant “power differential” between Poitras and the child complainant. Finally, there was extensive evidence that the complainant had a “disrupted attachment relationship” with her mother.
Commonwealth v. Fuller, 423 Mass. 216 (1996), which Poitras has conceded was the governing standard, made the relevance test originally announced in Commonwealth v. Bishop, 416 Mass. 169, 181 (1993), more stringent. The defendant must not merely set out the theories under which the records are “likely to be relevant” to an issue in the case, but he must “demonstrateQ a good faith, specific, and reasonable basis for believing that the records will contain exculpatory evidence which is relevant and material to the issue of the defendant’s guilt.” Commonwealth v. Fuller, supra at 225-226.
Those theories were conjecturally described as follows: (1) the complainant may have given a version of the abuse to the counselor that was inconsistent with other versions; (2) the complainant may have fabricated the *696abuse in order to gain sympathy from her mother who would then allow her release from her restrictive group home environment; and (3) the records might reveal that the complainant’s perception and memory were impaired.
The complainant’s credibility, which was central to the case, was challenged on cross-examination by questions probing why she had not disclosed the abuse earlier to various individuals, insinuating that she had disclosed it only when she was “stuck” in a group home. Any prejudicial effect of the complainant’s statements could only have been minimal, since she did not testify to having seen or been told of any actual abuse of the sister. See Commonwealth v. Aspen, 53 Mass. App. Ct. 259, 269 (2001).