dissenting:
I respectfully dissent. While I am inclined to agree that the admissibility of charging decisions may not be suitable for a rule of categorical exclusion and that the admissibility of adverse credibility findings in prior cases has been the subject of new law, I conclude:
A. With respect to the charging decisions, I would affirm on the district court’s alternative ruling that the probative value of such evidence in this case was substantially outweighed by the danger of confusing or misleading the jury; and in that way, I would avoid the vexed question that the majority opinion embraces.
B. As to the testimony of Detective Herrmann in a prior case, I agree with the majority opinion that the district court’s analysis should have considered the seven factors in Cedeño. But I would remand for the district judge to make that ruling in the first instance and in the ordinary course, subject to the usual appellate review for abuse of discretion.
Moreover, as the majority opinion observes, hearsay problems abound in this case; yet, since neither party raised these issues on appeal, the Court has not decided whether the charging documents or the prior credibility findings are admissible hearsay. The majority’s explicit recognition of that circumstance affords the district court a needed latitude on appeal, and prevents the opinion from being misread or overread.
A.
The majority opinion holds that there should be no categorical exclusion of charging documents from evidence. Maybe; but, in any event, this holding does little harm because trial judges would virtually always exclude such evidence for multiple and sound reasons. In my view, Judge Garaufis did not abuse his discretion in excluding the charging documents in this case.
In a nutshell, there were five people in one car with three guns. Under New York law, all five were in constructive possession of all three guns except for any gun “found upon the person.” See N.Y. Penal Law § 265.15(3)(a). A New York charging document charged all occupants with possession of all three guns. The state charge against White was dismissed, and White was charged in federal court with possessing one gun (on his person). He was convicted of that charge. White wanted to argue that if the gun had been on his person, the drafter of the state charging document would not have charged the other occupants with possessing it — assuming of course that the drafter understood that nuance of New York law.
The majority finds error in the district court’s apparent ruling that charging documents are categorically excluded from evidence. However, the district court prudently went on to find, in the alternative, that the “probative value” of the charging document would be “substantially outweighed by the danger of confusing or misleading [the] jury.” White, 2009 U.S. *254Dist. Lexis 113147, at *6. That seems to me an unexceptionable observation.
The majority rejects that finding on the extraordinary ground that the district court’s “prior finding that charging decisions are not proper subjects for cross-examination obviated the need for such balancing and east doubt as to the balancing made. In other words, the Rule 403 inquiry was tainted.... ” Op. at 247. I cannot agree that an alternative ground for an evidentiary ruling must be disregarded as “tainted” by a legal error (if there is one) in another cited ground for the ruling.
The majority goes on to do the balancing itself, and does a bad job of it, for several reasons.
First, the majority does not explain why we should not remand to the district court (by vacatur or under our Jacobson procedure) for it to do the balancing, now that we are all enlightened as to the law. See generally United States v. Jacobson, 15 F.3d 19, 22 (2d Cir.1994).
Second, in this case, the danger of confusing or misleading the jury is great: the underlying principles of New York law are complex; “possession” of a firearm is (as the district judge pointed out) “a legal term of art rather than a factual observation,” App. 23; the charging documents are steeped in hearsay; and explication would likely require testimony by state and federal prosecutors.
Third, the Rule 403 inquiry involves facts not before this Court. “The length of the chain of inferences necessary to connect the evidence with the ultimate fact to be proved necessarily lessens the probative value of the evidence, and may therefore render it more susceptible to exclusion as unduly confusing, prejudicial, or time-consuming....” Ravich, 421 F.2d at 1204 n. 10. The majority takes no account of the “length of the chain” connecting [i] the charges filed against the four women with [ii] the police officers’ observations. Were the charging decisions recommended by the police who made the direct observations? If not, what was the communication between the arresting officers and the individual who made the decisions? At a minimum, the question of admissibility should be remanded to the district court so that it can inquire into these issues, and any others that may bear on the balancing.
Fourth, the seeming breadth of the majority opinion may be expected to do harms in other eases. When there are several charging documents, and several defendants, the jury will be dealing with a variorum edition of the indictment. Charging documents may not have been drafted with the expectation that they would be final or authoritative. And drafts of charging documents may become the subject of motions under Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), Giglio v. United States, 405 U.S. 150, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972), or 18 U.S.C. § 3500. This is a can of worms.
B.
The second issue decided by the majority opinion concerns the relevance and probative value of findings in a prior unrelated case, in which the trial judge declined to credit certain testimony by a police officer. White wanted to use the earlier findings to cross-examine the police witness against him. The issue framed by the majority is whether this was “[e]vidence that might lead a jury to conclude that the officer was willing to lie in a similar case in order to secure á criminal conviction.” Op. at 239. The district judge relied on then-current precedent, to rule that such evidence must be excluded because the prior finding “was not a finding *255on [the officer’s] general veracity, and the [prior] case is entirely unrelated to this one.” White, 2009 U.S. Dist. Lexis 113147, at *9. As the majority opinion explains, those considerations became two of seven (non-exhaustive) considerations listed in the later-decided case of Cedeño.
So it would seem obvious that if this evidentiary ruling makes a difference, the proper disposition is to remand (either by vacatur or under our Jacobson procedure), for the district court to apply the seven-plus factors of Cedeño.
As Cedeño makes clear, whether a witness “was willing to lie in a similar case” is a finding with several component fact issues. Was the case similar? I am inclined to think so, and I would so decide if I were the district judge (which I am not). Was there a general finding as to veracity in the prior case? As the district court found, there was not. Were the statements that were not credited made as willing lies? That is a question that should be confided to the district court, which can supplement the transcript of the prior case with live testimony from the police officer. The district court found that the judge in Goines credited parts of the officer’s testimony and discredited other parts, and that the discredited parts could have been deficient by reason of “a combination of passage of time, confusion, [and] lack of firsthand knowledge.” White, 2009 U.S. Dist. Lexis 113147, at *9-10; see also Zaman v. Mukasey, 514 F.3d 233, 237 (2d Cir.2008) (per curiam) (emphasizing that a factfinder may “note[] problems with parts of [a party’s] story that go to the [party’s] credibility” without “mak[ing] the ultimate conclusion that the entirety of the testimony suffers from a lack of credibility”). On remand, the district judge could amplify those findings in light of Cedeño, review the Goines transcripts in order to make a more fine-grained set of findings, and elicit from the police officer an explanation for the inconsistencies noted by the Goines court that (as the majority opinion emphasizes) is absent from the record.
Instead, the majority (again) conducts its own evidentiary analysis, making findings as to the Cedeño factors and concluding that they “weigh strongly in favor of admissibility.” Op. at 250. The majority suggests that “should, more data come in on the various factors,” the trial court is free to conduct the evidentiary balancing “anew.” Op. at 251 n.8; 248 n.6. But it is the trial court — in the first instance — that should make findings on the Cedeño factors and conduct the evidentiary balancing. This Court’s role is to review the balancing, not to conduct it.1
In short, now that we have clarified the law (as the majority opinion undertakes to do), I would remand for the district court to apply it.
. In performing the trial court’s role, the majority discounts "the district court's fear that admitting the evidence would confuse the jury.” Op. at 251. The majority opinion posits that the defense "simply sought” to ask a single question on cross-examination, id.: "whether [the officer] previously gave testimony ... under oath, in another federal gun possession case, and whether the judge in that case refused to credit six different aspects of his testimony.” App. 257. But the question becomes a lot less simple when one considers that the prosecution would then get to conduct redirect, and it might in fairness be afforded the opportunity to set out the facts of Goines and elicit explanations from the police witness. See United States v. Diaz, 176 F.3d 52, 80 (2d Cir.1999) (“The scope of redirect examination is a matter entrusted to a trial judge’s broad discretion.”).