¶ 34. dissenting. In this case, the jury convicted defendant based on evidence that included testimony from a confidential informant who was terminated for unrelated reasons after participating in the controlled buy in this case. The majority concludes the trial court erred in limiting certain evidence surrounding the informant’s termination. But the jury had ample evidence with which to weigh the informant’s credibility and was not misled by the State. Defendant here asks this Court to second-guess the jury’s credibility determination. This we should not do. Therefore, I respectfully dissent.
¶ 35. I agree with the conclusion that the trial court properly excluded the informant’s termination report as well as defendant’s cross-examination questions of Detective LaChance regarding the subsequent and unrelated traffic stop that led to the informant’s termination. Defendant’s attempt to admit this evidence in either form was designed to elicit inadmissible hearsay, since the detective was not present at the traffic stop and therefore had no firsthand knowledge of the incident. See V.R.E. 602 (allowing exclusion of testimony of which a witness has no personal knowledge), 801 (defining hearsay). Moreover, the specifics of that incident were inadmissible even if the evidence had not been hearsay. Under Vermont Rule of Evidence 608(b), “[ejvidence of specific instances of conduct offered to attack the witness’s credibility are not admissible through other witnesses as a matter of law.” John A. Russell Corp. v. Bohlig, 170 Vt. 12, 22, 739 A.2d 1212, 1220 (1999) (citing 28 C. Wright et al., Federal Practice and Procedure § 6117, at 88 (1993)). Extrinsic evidence of the informant’s unrelated conduct could not be admitted through the detective to impeach the informant’s credibility — it had to be admitted through the informant herself, in the discretion of the trial court. See id.
¶ 36. The majority also properly rejects defendant’s argument that the informant’s termination was probative of her motive to fabricate her testimony under Rule 404(b). Defendant’s contention *247is simply illogical because the informant’s termination, which defendant claims gave rise to her motive to fabricate, occurred after the controlled buy from defendant and thus could not have influenced the events that day or her testimony consistent with those events at trial. Cf. In re A.B., 170 Vt. 535, 536-37, 740 A.2d 367, 369-70 (1999) (mem.) (reversing and remanding for new trial where defendant had demonstrated that alleged victim’s motive to fabricate might have arisen before victim reported sexual abuse, but was not permitted to explore the events giving rise to motive).
¶ 37. The majority is then taken, however, with defendant’s claim that the trial court erroneously refused to admit evidence of the reasons for informant’s termination — either through the informant termination form or the detective’s testimony about the contents of the form — when it concludes this evidence should have been admitted because it was “so interwoven with the crime charged it [could] not be separated without skewing the narrative.” John A. Russell Corp., 170 Vt. at 21, 739 A.2d at 1220. The majority reasons that, “[b]ecause the State elicited only part of the story concerning Detective LaChance’s relationship with and confidence in the informant, the jury was left with a clear misimpression that Detective LaChance trusted the informant and believed her to be truthful.” Ante, ¶ 30.
¶ 38. This analysis errs for several reasons. First, as the majority acknowledges, the traffic stop leading to the informant’s termination was not “interwoven with the crime charged” in the sense contemplated by John A. Russell Corp. In that case, we held that an employee’s previous termination for dishonesty was not “interwoven” with his later breach of contract claim against a different employer, and that evidence of the previous termination could “be excluded without affecting the narrative on the breach of contract at all.” Id. at 21-22, 739 A.2d at 1220. The majority fails to distinguish the. facts of John A. Russell Corp. from the instant case. Here the informant’s termination was, -without dispute, for an unrelated incident, which occurred long after the subject event and was completely unrelated to the controlled buy in this case.
¶ 39. Moreover, although the majority relies for its “context” proposition on State v. Findlay, the facts of that case are different. 171 Vt. 594, 765 A.2d 483 (2000) (mem.). There, prior to trial, the court excluded all evidence of prior controlled buys by the informant on the State’s motion, and, in addition, the infor*248mant was never called as a witness at trial and was therefore never subject to impeachment on cross-examination, a distinct difference. Id. at 596, 765 A.2d at 486-87. That difference is compounded by our later clarification that Findlay’s reasoning implicates cases involving “a total deprivation of the opportunity to show the witness’s bias.” State v. Brochu, 2008 VT 21, ¶ 87, 188 Vt. 269, 949 A.2d 1035. “By contrast, we have been particularly supportive of restrictions on cross-examination when the defendant wanted to explore details of criminal conduct or other misconduct that was irrelevant or only marginally relevant to the charges against the defendant.” Id. Defendant here clearly had the right to attack informant’s credibility, and she was not deprived of the opportunity to do so as in Findlay. The exclusion of other evidence of specific conduct of the informant after the controlled buy did not improperly restrict the opportunity for impeachment. The trial court applied the rules of evidence by limiting the scope of defendant’s efforts to impeach. The rules of evidence prohibit the introduction of unrelated instances of specific conduct because they “‘possess[] the greatest capacity to arouse prejudice’ . . . [and] can undermine accurate fact finding because of the tendency of juries to give [them] too much weight.” John A. Russell Corp., 170 Vt. at 23, 739 A.2d at 1221 (quoting Advisory Committee Note to F.R.E. 405). For this reason, courts should be especially wary of allowing otherwise inadmissible evidence of specific conduct to be introduced as “context.”
¶ 40. Second, the jury was not misled about the informant’s relationship with law enforcement. In making her context argument, defendant cherry-picks isolated testimony while ignoring the rest of the record before the jury. The jury learned of the origin of the informant’s deal with law enforcement: the informant initiated contact with police requesting to cooperate in exchange for consideration for leniency on several pending felony charges related to a burglary. The informant’s signed contract was admitted into evidence at trial. The jury heard that informant initially faced thirty-five years of imprisonment on her pending felony charges, but ultimately received a deferred eighteen-month sentence. Testimony was introduced that, as part of the deferred sentence agreement, she was required to testify in defendant’s trial and ultimately the charges would be removed from her record. As the majority concedes, “[t]he informant’s stake in her cooperation was large, and the jury knew of the informant’s *249ongoing risk if she violated the conditions of that cooperation.” Ante, ¶ 26. Moreover, the informant herself admitted in her testimony that she selected defendant as a target for investigation and that she did not personally like her.
¶ 41. The jury also heard evidence of law enforcement’s perception of the informant’s truth telling and the level of trust between them. The jury heard about the numerous steps taken to corroborate the information given by the informant according to standard operating procedure, including the signing of the confidential-informant contract outlining informant’s responsibilities, surveillance of informant by numerous officers at the time of the controlled buy, and thorough pat-downs both before and after the buy. The jury learned of the Drug Task Force’s plan to use informant’s controlled buy from defendant to establish probable cause for a wire warrant to record future buys, a plan thwarted by informant’s termination.
¶ 42. Most pertinently, the jury heard the details of the incident giving rise to the informant’s termination. Detective LaChance confirmed that the informant had not stopped cooperating voluntarily but had been terminated. Detective LaChance testified that although he trusted the informant at the time of the controlled buy, he ultimately chose to terminate her cooperation agreement several weeks afterward. He testified that he would never work with an informant he did not trust. He testified to the steps he took to terminate, including filling out a termination form to place in her informant file and informing the prosecutor on the informant’s case that she had been terminated. The jury heard from both Detective LaChance and the informant that the detective attempted to contact the informant to let her know she had been terminated, but that she did not respond to his calls.
¶ 43. Although defendant was not able to introduce all of the details of the informant’s traffic stop through the cross-examination of Detective LaChance, evidence about the stop eventually came in through the testimony of the informant herself on cross-examination by defendant’s attorney. Informant testified that she stopped cooperating with law enforcement simply because she “felt uncomfortable” and stopped returning Detective LaChance’s calls. But this opened the door for defendant to try to impeach her by showing that it was circumstances surrounding the traffic stop that led to her termination, as evidenced by the following exchange:
*250Defense counsel: Do you recall being stopped by the police on August 20th?
Informant: Yeah.
Defense counsel: And you were with a friend?
Informant: Yeah.
Defense counsel: On that date, isn’t it true that you lied to the police by telling them that you were working with the Task Force in the middle of a controlled buy?
Informant: No, I did not say that.
Defense counsel: Do you recall telling the officer that you were hanging out with this person because you thought they had drugs on them?
Informant: No.
Defense counsel: This is in the deposition, where you said, “I guess I said something about the reason I’m hanging out with this person is because I think he has drugs on him, yes.” In error?
Informant: Yes, I did. I did say that in the deposition, but with the whole — the police incident, I didn’t say that to the police officer.
¶ 44. Admittedly, the testimony from the informant at trial on cross-examination was disjointed, loaded with contradiction and at times difficult to follow. But this in itself is revealing. To summarize, several officers, including Detective LaChance, testified that the informant was terminated for reasons that were left unstated before the jury, but Detective LaChance also testified he would not work with an informant he did not trust. In contradiction, the informant testified that she stopped working with the Task Force because she became “uncomfortable.” On cross-examination, the informant agreed she purposely stopped returning the detective’s phone calls. She testified that about a month after defendant’s buy, she was stopped while driving without a license with a *251passenger in her car, and that drugs were found as a result of the stop. And while informant denied it when shown her prior deposition testimony stating that she told police at the traffic stop that she was working with the Task Force, she also confirmed that she had stated in her deposition testimony that she told the police officer during her stop that she was with that particular passenger because she believed he had drugs on him.
¶ 45. In total, this is not a record that requires a new trial. This is evidence of an evasive witness admitting to certain facts that go to the heart of defendant’s objections before us, sufficient for the jury to question the informant’s story and her reliability and to infer that the informant was terminated because she was considered unreliable and untrustworthy by law enforcement. The jury was not deprived of the opportunity to determine the informant’s credibility based on these facts; it simply made a credibility determination that was unfavorable to defendant. This credibility determination was soundly within the province of the jury and not for us to disturb on appeal. State v. Johnson, 2013 VT 116, ¶ 27, 195 Vt. 498, 90 A.3d 874 (“We are not triers of fact, and we will not substitute our judgment for that of the jury.”).
¶ 46. The majority errs for a third reason. If the jury had insufficient information or was somehow misled about the reason for terminating informant, it was due to defendant’s failure to ask the right questions on cross-examination, not due to the trial court’s evidentiary rulings. Defendant could have asked Detective LaChance his opinion of informant’s truthfulness at the time he terminated her. See V.R.E. 608(a) (“The credibility of a witness may be attacked ... by evidence in the form of opinion or reputation, but subject to [the limitation that] the evidence may refer only to character for truthfulness or untruthfulness . . . .”).
¶ 47. But defendant did not take this tack. Instead, she repeatedly questioned Detective LaChance about the specific circumstances surrounding the traffic stop. Even her inquiry, relied on by the majority, as to whether Detective LaChance believed the informant could “cooperate honestly and truthfully” in the investigation was unequivocally probative not of the informant’s character for truthfulness, but rather of her conduct during the traffic stop. Only in hindsight on appeal does defendant attempt to contort this line of questioning into one of opinion under Rule *252608.4 As the majority notes, defendant’s constitutional right to confront adverse witnesses does not provide her carte blanche to elicit inadmissible testimony. State v. Larose, 150 Vt. 363, 369, 554 A.2d 227, 231-32 (1988) (“ ‘Generally speaking, the Confrontation Clause guarantees an opportunity for effective cross-examination, not cross-examination that is effective in whatever way, and to whatever extent, the defense might wish.’ ” (quoting Delaware v. Fensterer, 474 U.S. 15, 20 (1985)). Defendant could have elicited her desired testimony in an admissible form but failed to do so.
¶ 48. In short, defendant did not have a constitutional right to introduce inadmissible testimony, and the trial court did not err in excluding it. The trial court’s ruling limiting cross-examination of Detective LaChance and refusing admission of the termination report was not error. If not clear through the cross-examination of the informant, defendant could have made her point — that informant was terminated due to her untrustworthiness — by eliciting Detective LaChance’s opinion under 608(a). Given these *253circumstances, even if the trial court had erred in its evidentiaryruling — which it did not — there was no prejudice to defendant and any such error would have been harmless beyond a reasonable doubt. See State v. Fellows, 2013 VT 45, ¶ 23, 194 Vt. 77, 76 A.3d 608 (reciting the standard for a harmless-error analysis). For these reasons, I would affirm the trial court.
¶ 49. I am authorized to state that Judge Crawford joins this dissent.
The majority claims that the question defendant was not permitted to ask — “You thought she couldn’t cooperate in this investigation honestly and truthfully?” — was “virtually indistinguishable” from a question regarding informant’s character for truthfulness. Ante, ¶¶ 31-32 & n.2. Nothing could be further from the truth, as defendant’s question was rooted exclusively in the informant’s conduct on a specific occasion, and in Detective LaChance’s decision to terminate her because she “couldn’t cooperate in this investigation honestly and truthfully” based on that specific incident.
Moreover, the majority claims that even if the excluded question related to informant’s specific conduct, it would nevertheless be admissible under Rule 608(b)(2) as evidence of specific conduct used to impeach the testimony given by a witness on cross-examination regarding another witness’s character for truthfulness or untruthfulness. In short, the majority essentially contends that Detective LaChance opened the door by “vouching” for the informant’s credibility on direct examination, so that it was fair game for defendant to impeach his testimony with specific instances of conduct on cross-examination. Ante, ¶ 32. However, it is a stretch to say that Detective LaChance vouched for informant’s credibility. The direct examination contains a brief exchange where, in response to the State’s question about whether the requirement that the informant sign a cooperation agreement means law enforcement does not trust her, Detective LaChance answered “I wouldn’t have worked with her [if I didn’t trust her].” This statement was elicited, not volunteered, and solely related to whether he trusted the informant while he worked with her — at the time of the controlled buy. Such a statement, without more, is hardly a resounding endorsement of the informant’s credibility in general, and in fact, the jury could easily infer that Detective LaChance no longer trusted her by virtue of the fact that he terminated her agreement several weeks later.