MEMORANDUM & ORDER
WILLIAM H. PAULEY III, District Judge:
Due process requires the Government to provide a criminal defendant with favorable material evidence in its possession. Brady v. Maryland, 373 U.S. 83, 86, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963). The Brady rule safeguards the fundamental principal that a trial is a search for the truth. It also protects the Government’s unique interest in a criminal prosecution “not [to] win a case, but [ensure] that justice shall be done.” Berger v. United States, 295 U.S. 78, 88, 55 S.Ct. 629, 79 L.Ed. 1314 (1935). When Brady material is withheld, the Government’s case is “much stronger, and the defense case much weaker, than the full facts would have suggested.” Kyles v. Whitley, 514 U.S. 419, 429, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995).
[233]*233Defendant Dwayne Thomas moves pursuant to Rules 29 and 33 of the Federal Rules of Criminal Procedure for a judgment of acquittal or a new trial. Thomas contends that the Government suppressed Brady material relating to the robbery of Blundy Jean Jacques (the “Jean Jacques robbery”) and that he suffered impermissible spillover prejudice on the other robbery charges. Thomas also asserts that the Government offered insufficient evidence to sustain the kidnapping conspiracy charge. For the following reasons, Thomas’s motions are granted in part and denied in part.
BACKGROUND
I. The Verdict
On May 2, 2013, a jury convicted Thomas of robbery conspiracy (Count 1), the robbery of the Magic Pot Bar on October II, 2008 and firearm possession in relation to that crime (Counts 2 and 4), the robbery of Blundy Jean Jacques on January 12, 2012 and firearm possession in relation to that crime (Counts 3 and 5), firearm possession while being a convicted felon on January 13, 2012 (Count 6), and kidnapping conspiracy (Count 7). The jury acquitted him of kidnapping and possessing a firearm in relation to that crime (Counts 8-9). Because Thomas’s motions related to Counts 1, 2, 4, 6 and 7 do not raise substantial issues, a more fulsome description of the facts relating to those crimes is unnecessary.
II. . The Jean Jacques Robbery
On January 12, 2012, two men robbed Jean Jacques and his cousin, Albert Remy, at gunpoint. Jean Jacques and Remy were home together that evening. At some point, Jean Jacques left the house and was accosted by two armed men. The robbers did not bother to cover their faces. Over the next few minutes, they forced Jean Jacques back into his home, where they robbed Jean Jacques and Remy, taking cellular telephones, jewelry, and shoes packaged for resale overseas. Detective Elaine Jenkins and her partner, Detective Michael Parchen, were assigned to investigate that robbery. After the robbery, Jean Jacques provided detailed physical descriptions of the robbers. But exactly what those descriptions were, formed the basis of the defense theory at trial. And to whom he gave them, spawned this post-trial motion.
From the outset, Thomas sought to attack the reliability of Jean Jacques’s identification of Thomas as one of the robbers. Thomas moved pre-trial to suppress Jean Jacques’s identification of him. At a suppression hearing, the Government called Jenkins as a witness. Jenkins testified about the identification procedure she used with Jean Jacques. She also testified about the descriptions of the perpetrators that Jean Jacques purportedly gave her on the night of the robbery. After hearing Jenkins’s testimony, this Court informed the Government that it would have to call “other witnesses,” because it “ha[d]n’t come close” to meeting its burden. (Hearing Transcript dated Sep. 11, 2012 (“9/11/12 Tr.”) at 87.) The Government elected to call Jean Jacques.
According to Jean Jacques’s testimony, six days after the robbery, she asked Jean Jacques to come to the precinct to view photographs in an effort to identify the perpetrators. After he. arrived at the precinct, Jean Jacques “accidentally” saw a single photograph of Thomas- — whom he identified as one of the robbers. Then, Jenkins showed Jean Jacques a single photograph of Kenton Russell — whom Jean Jacques identified as the second robber. By Order dated November 14, 2012 (the “November Order”), [234]*234this Court found the single-photograph identification procedure unduly suggestive. See United States v. Russell, No. 12-cr-174 (WHP), 2012 U.S. Dist. LEXIS 172206, at *1 (S.D.N.Y. Nov. 14, 2012). The Second Circuit has “consistently condemned the exhibition of a single photograph as a suggestive practice, and where no extenuating circumstances justify the procedure, as an unnecessarily suggestive one.” Mysholowsky v. New York, 535 F.2d 194, 197 (2d Cir.1976).
Nevertheless, Jean Jacques’s identification of Thomas was independently reliable. Russell 2012 U.S. Dist. LEXIS, at *27; see also Neil v. Biggers, 409 U.S. 188, 198-201, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972). To summarize, this Court found that: (1) Jean Jacques had an extensive opportunity to observe Thomas during the crime; (2) his attention was directed at the robbers during the crime; (3) he accurately described the robbers immediately after the crime to Jenkins; (4) identified Thomas with “100 percent certainty” during the identification procedure; and (5) identified Thomas only six days after the robbery. Russell 2012 U.S. Dist. LEXIS 172206, at *25-27.
Prior to the suppression hearing, the Government provided exhibits it would seek to introduce at the hearing as well as 3500 material for Jenkins.1 Among those materials were the following pertinent documents:
Complaint Report and Handwritten Notes: This report — produced as Exhibit 3501 — S-contained a typed narrative summary of the offense as well as handwritten notes on the last page. Jenkins made the handwritten notes, which included the following description of the robbers: “Perp # 1: M/B/20s (Jamaican) 5'7", approx. 160 lbs., Braids (cornrows), Black hoody — Blk Jeans, armed with 2 firearms (1) black revolver, (1) black semi-automatic. Perp #2: M/B/ 20s (Jamaican) 5'9"-10" approx. 160 lbs, Black Jacket and Blue Jeans, armed w/ 1 black semi-automatic.”
DD5 Follow-Up No. 1: Jenkins authored this report, which was produced as part of Jenkins’s 3500 material and marked as Exhibit 3501-C. It includes the following description of the robbers: “Perp # 1 M/B/20s 5/7", 140 lbs, dark skin, black hoodie and blue jeans, had black revolver and black semi-auto handgun. Perp #2 M/B/20s 5'9"-10", 160 lbs, medium skin, braids (cornrows) black jacket and blue jeans, had a black semi-auto handgun.”
DD5 Follow-Up Nos. 4 and 7: These reports were included in Jenkins’s 3500 material and produced together as Exhibit 3501-F. DD5 Follow-Up No. 4 memorializes an interview of Jean Jacques on the night of the robbery. According to the notes in this report, Jean Jacques described the robbers as follows: “Perp # 1 — M/B (Jamaican), 5'7", approximately 160 pounds, early twenties, braids, armed with two firearms — one black revolver and one black semi-automatic, wearing a black hoody and black jeans. Perp #2 — M/B (Jamaican) 5'9"-5'10", approximately 160 pounds, early twenties, armed with a black semi-automatic firearm, wearing a black jacket and blue jeans.” DD5 Follow-Up No. 7 memorializes Jenkins’s acquisition of a trap and trace pen register in an attempt to locate the stolen cellular telephones.
According to Jenkins’s handwritten notes (Exhibit 3501-S) and DD5 FollowUp No. 4 (Exhibit 3501-F), Jean Jacques told her that the shorter robber had the [235]*235braids or eornrows. But in Jenkins’s typed report, DD5 Follow-Up No. 1 (Exhibit 3501-C), she switches them, noting that Jean Jacques described the taller robber as having braids or eornrows. That distinction is vital, because Thomas and Russell were arrested the day after the robbery. At that time, Thomas — the taller of the pair — had eornrows while Russell did not.
But the discrepancy in the descriptions is not the only pertinent fact to be gleaned from the documents. Just as important is who made the reports and interviewed Jean Jacques. Because all of these documents were produced as part of Jenkins’s 3500 material and 18 U.S.C. § 3500 requires the Government to produce “any statement of [a witness it calls],” the memorialized descriptions appeared to be Jenkins’s statements.
And neither Jenkins nor the Government disabused defense counsel of the notion that Jenkins took the interview of Jean Jacques memorialized in the report. Indeed, Jenkins admitted the handwritten notes at the bottom of Exhibit 3501-S were hers. (9/11/12 Tr. at 54.) And Exhibit 3501-C indicates that she was the author of the report. The author of 3501-F was redacted from DD5 Follow-Up No. 4, but not for DD5 Follow-Up No. 7. Jenkins testified, and the Government stipulated, that she was the author of both documents included in Exhibit 3501-F:
Q: Let me also show you what’s been marked as 3501-F, for identification. Is this another report that you did, Detective Jenkins?
A: I assume I did this. This is blocked out.
MR. DEVITA: Can we stipulate that it is a report of Detective Jenkins, Mr. Kasulis?
MR. KAULIS: I believe so. Is the witness looking at the second — have you given her both pages?
MR. DEVITA: Yes.
MR. KASULIS: Then, yes, I think so.
A: Yes, this is.
(9/11/12 Tr. at 30.)
Jean Jacques also testified at the hearing. Importantly, Jean Jacques admitted that on the night of the robbery he described the robbers to law enforcement, but he also denied speaking to Jenkins at that time or ever telling her that “the shorter man had eornrows.” (Hearing Transcript dated Sep. 24, 2012 (“9/24/12/ Tr.,”) at 19-20, 24-25, 42.)
After the hearing, the parties briefed the issue, of whether the Government had earned its burden. In its brief, the Government sought to explain away the discrepancy between Jean Jacques’s hearing testimony regarding his description of the robbers and the description that Jenkins wrote down. In a footnote, the Government claimed that Jean Jacques “confirmed [ ]Jenkins’s suggestion that the one piece of paper describing the shorter man as having braids was a mistake in transcription, corrected when the formal NYPD reports were filed.” (Government’s Post-Hr’g Brief in Opp’n to Def.’s Mot. to Suppress Identification Evidence, Oct. 19, 2012, (“Gov’t Br.”) at 17 n. 5) (emphasis in original.) That assertion rests on the conclusion that the handwritten notes (Exhibit 3501-S) and DD5 Follow-Up No. 4 (Exhibit 3501-F) were both authored by Jenkins and are “one” document — and that the typed notes in DD5 Follow-Up No. 1 (Exhibit 3501-C) simply represents her transcription of her own notes.
Prior to trial, the Government again provided 3500 materials for Jenkins and Jean Jacques. The 3500 material produced for the hearing was re-produced for trial, us[236]*236ing different exhibit numbers (a confusing and unhelpful practice). Pertinent here are:
Trial Exhibits 3510-0 and 3510-L:
Produced as part of Jean Jacques’s 3500 material, these exhibits are both DD5 Follow-Up No. 4, previously produced at the hearing as Exhibit 3501-F. Two items were unredacted for the trial exhibits: the name of the interviewee (Jean Jacques) and the name of the reporter (Detective Parchen). DD5 Follow-Up No. 7 was not produced in connection with Exhibits 3510-0 or 3510-L. Trial Exhibit 3518-G: Produced as part of Jenkins’s 3500 material, this exhibit is the same DD5 Follow-Up Nos. 4 and 7 previously produced at the hearing as Exhibit 3501-F. The interviewer’s and reporter’s names are redacted, as they were for the hearing.
Trial Exhibit 3518-T: Produced as part of Jenkins’s 3500 material, this exhibit is the same Complaint Report and Handwritten Notes previously produced at the hearing as Exhibit 3501-S.
These documents were identical to the 3500 material produced for the suppression hearing, except that the Government produced DD5 Follow-Up No. 4 in unredacted form. That unredacted DD5 Follow-Up No. 4 indicates that its author was Parchen, not Jenkins. But the Government never corrected Jenkins’s testimony that she authored the report, its prior stipulation that Jenkins was the author, or its assertion that there was only “one piece of paper” with the inconsistent description.
Trial commenced on April 22, 2013. Evidence of the Jean Jacques robbery was overwhelming: Jean Jacques and Remy testified consistently and compellingly about the robbery and one of the stolen cellular telephones was recovered. Jean Jacques testified that “the taller [robber] had the braids,” (Trial Transcript (“T. Tr.”) at 606), and identified that man as Thomas. (T. Tr. at 611.) Jean Jacques testified further that six days after the robbery, he met with Jenkins, who showed him a photograph of Thomas followed by a photograph of Russell. Viewing those photographs separately, Jean Jacques identified Thomas as the taller robber with braids. (T. Tr. at 621-22.) Jean Jacques also denied speaking with Jenkins on the night of the robbery. (T. Tr. at 633-34.) When Thomas’s counsel challenged Jean Jacques’s recollection, Jean Jacques asserted that he was “[as] certain that [Jenkins] was not there on January 12 as [he was certain] that Dwayne Thomas was [there].” (T. Tr. at 635.)
Adhering to his earlier strategy, at trial, Thomas attacked the reliability of Jean Jacques’s identification of Thomas as one of the robbers. Thomas’s counsel cross-examined Jean Jacques vigorously, sought to highlight the discrepancy between the descriptions of the robbers purportedly given to Jenkins, and elicited expert testimony regarding factors affecting an eyewitness’s ability to reliably identify a perpetrator. That strategy was sound because the most important evidence establishing Thomas as one of the robbers was Jean Jacques’s testimony ... Nevertheless, the jury convicted Thomas of the charges related to the Jean Jacques robbery.
The Government called Parchen to testify solely about the tracing of Jean Jacques’s stolen cellular telephone. (Gov’t Br. at 14.) Parchen testified that he helped Jenkins track the stolen phone but also that he “interviewed one of the victims” at the scene. (T. Tr. at 972.) At that time, there was no follow-up by the Government on that fleeting statement. No detective reports were produced as 3500 material for Parchen. In fact, the only 3500 material produced for him was [237]*237one page of handwritten notes made in preparation for his testimony regarding the phone trace, which contains no reference to any interview Parchen did of Jean Jacques. The Government did not call Jenkins to testify at trial.
Instead, Thomas called Jenkins on the defense case. During that examination, Thomas’s counsel questioned Jenkins about Exhibit 3518-G. For the first time, Jenkins disclaimed authorship of that report:
Q: Let me show you what’s been marked for identification as 3518-G. Do you have that, Detective Jenkins? Now, this [referring to DD5 Follow-Up No. 4] is your typed report from your interview of Mr. Jean Jacques on the evening of January 12, 2012, isn’t it? If you look on the second page [referring to DD5 Follow-Up No. 7], it tells you who the reporting officer is.
A: On the second page [referring to DD5 Follow-Up No. 7] it is — it is my report.
Q: Okay. So this [referring to DD5 Follow-Up No. 4] is your report that you prepared of your interview with Mr. Jean Jacques. Is that right?
A: The second page [referring to DD5 Follow-Up No. 7], yes.
Q: And in that report [referring to DD5 Follow-Up No. 4] you repeat “Perp 1, male black, five-seven, approximately 170 pounds, early 20s, braids.” Right? Do you see that in that paragraph 2, third line and fourth line down?
A: Yes, I do. I’m just not sure if this [referring to DD5 Follow-Up No. 4] is my report.
(T. Tr. at 996-97.)
$ ‡ ‡ ‡
Then on redirect examination:
Q: Ok. Now, Ms. Moyne was going back over with you — can I see 3518-G, the sequence of your handwritten notes? Your handwritten notes reflect that the taller of the two individuals had — the shorter of the two individuals had the cornrows, right?
A: According to my handwritten notes, yes.
Q: Now, was Detective Parchen with you when you interviewed Mr. Jean Jacques?
A: He was at the location. Whether he was standing right with me—
Q: You don’t know one way or the other?
A: No.
Q: If he wasn’t there, where would he get the information that he put in report 3518-G? He’d get that from you, right?
A: No, he — he interviewed him as well.
Q: He interviewed Jean Jacques as well?
A: According to his report, he interviewed him as well.
(T. Tr. at 1043-44.)
And then, at sidebar:
MR. DEVITA: The name Detective Parchen is blanked out [of 3518-G], The only name on this report, and it’s produced as 3500 material for Detective Jenkins, is Detective Jenkins’s name. When Detective Parchen was here, we had one page of 3500 material, and it did not include this. Unless we want to recall Detective Parchen, I just want to bring out that his report reflects the fact that the taller — the shorter of the two individuals had the braids.
MS. MOYNE: Your Honor, may I just borrow this for [a] second to explain to Mr. DeVita? This report was also produced in an unredacted form. I have the citation there. It was produced as Mr. Jean Jacques’s 3500, it was unredacted, and it indicates that it was Mi[238]*238chael Parchen. So, first off, you have it in unredacted form. Secondly, we produced it as Detective Jenkins because we believed that Mr. — Detective Parchen may have received this description from Detective Jenkins. So in [an] abundance of caution, we produced it as her 3500 material and as Mr. Blundy Jean Jacques’s 3500 material.
(T. Tr. at 1045-46.)
Apparently, this was the first time Thomas’s counsel learned Parchen, not Jenkins, was the author of DD5 FollowUp No. 4. The Government had discovered that fact as it was “preparing for trial.” (Oral Argument Transcript, dated Aug. 8, 2013 (“8/9/13 Tr.”) at 21.) However, it never memorialized that information and, until the sidebar, the Government had not share it with Thomas’s counsel. Nor had the Government corrected Jenkins’s hearing testimony, its stipulation, or its assertion in a footnote to its brief — all of which identified Jenkins as the author.
After the revelation during Jenkins’s testimony that DD5 Follow-Up No. 4 memorialized Parchen’s interview of Jean Jacques on the night the robbery, Thomas recalled Parchen. During Thomas’s direct examination, Parchen acknowledged that he interviewed Jean Jacques on the night of the robbery and prepared DD5 FollowUp No. 4. (T. Tr. at 1053.) In that interview, Jean Jacques told Parchen that the shorter of the two perpetrators had the braids. (T. Tr. at 1054.) That description is consistent with Jenkins’s handwritten notes, but the important difference is that it was Parchen, not Jenkins, who conducted the interview.
DISCUSSION
I. The Brady Motion
The Government violates a defendant’s due process rights when it fails to disclose favorable material evidence to a criminal defendant. Brady, 373 U.S. at 86, 83 S.Ct. 1194; United States v. Agurs, 427 U.S. 97, 107, 96 S.Ct. 2392, 49 L.Ed.2d 342 (1976); see generally United States v. Meregildo, 920 F.Supp.2d 434 (S.D.N.Y. 2013). Thomas contends that the Government violated his due process rights because it suppressed evidence of an exculpatory or impeaching statement made by the lone identifying witness to the Jean Jacques robbery. To prove a Brady violation and obtain a new trial, Thomas must show: “(1) the Government, either willfully or inadvertently, suppressed evidence; (2) the evidence at issue is favorable to the defendant; and (3) the failure to disclose this evidence resulted in prejudice.” United States v. Coppa, 267 F.3d 132, 140 (2d Cir.2001); see also Moore v. Illinois, 408 U.S. 786, 794-95, 92 S.Ct. 2562, 33 L.Ed.2d 706 (1972).
A. Favorable Evidence
“Favorable evidence” is exculpatory evidence or impeachment material. Strickler v. Greene, 527 U.S. 263, 281-82, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999); United States v. Bagley, 473 U.S. 667, 676, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985); Giglio v. United States, 405 U.S. 150, 154-55, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972); see also United States v. Mahaffy, 693 F.3d 113, 127 (2d Cir.2012). Evidence that Jean Jacques — the sole identifying witness — provided a description of the perpetrators to Parchen that was inconsistent with his trial testimony (i.e., impeaching) or irreconcilable with Thomas’s and Russell’s actual heights (i.e., exculpatory) is favorable evidence for Thomas. See Leka v. Portuondo, 257 F.3d 89, 99 (2d Cir.2001) (evidence that undermines the testimony of eyewitnesses — who are central to a guilty verdict — is Brady material because it is “of a kind that would suggest to any [239]*239prosecutor that the defense would want to know about it”).
B. Knowledge of the Favorable Evidence
The duty to disclose favorable evidence to an accused is an affirmative one that extends “to material evidence ... known to the prosecutor.” United States v. Avellino, 136 F.3d 249, 255 (2d Cir.1998). A prosecutor is presumed “to have knowledge of all information gathered in connection with his office’s investigation of the case and indeed ‘has a duty to learn of any favorable evidence known to the others acting on the government’s behalf in the case, including the police.’ ” Avellino, 136 F.3d at 255 (quoting Kyles, 514 U.S. at 437, 115 S.Ct. 1555); see also United States v. Payne, 63 F.3d 1200, 1208 (2d Cir.1995). A prosecutor’s constructive knowledge extends to individuals who are “an arm of the prosecutor” or part of the “prosecution team.” United States v. Gil, 297 F.3d 93, 106 (2d Cir.2002); see also United States v. Morell, 524 F.2d 550, 555 (2d Cir.1975); United States v. Bin Laden, 397 F.Supp.2d 465, 481 (S.D.N.Y.2005).
Whether someone is part of the prosecution team depends on the level of interaction between the prosecutor and the agency or individual. See Morell, 524 F.2d at 555; see also United States v. Diaz, 176 F.3d 52, 106-07 (2d Cir.1999) (circumstances that make someone part of the prosecution team include whether the individual actively investigates the case, acts under the direction of the prosecutor, or aids the prosecution in crafting trial strategy). At issue here is whether Parchen is part of the prosecution team. He was Jenkins’s partner, assisted in the investigation of the Jean Jacques robbery, and testified for the Government at trial. That makes him part of the team. So, the Government’s constructive knowledge extends to whatever Parchen knew — including any description that Jean Jacques gave him of the robbers on the night of the robbery.
C. Suppression of Evidence
Evidence is suppressed when the prosecutor does not disclose it “in time for its effective use at trial.” Coppa, 267 F.3d at 135 (internal citations omitted); see also Avellino, 136 F.3d at 255. But “[e]vidence is not ‘suppressed’ if the defendant either knew, or should have known, of the essential facts permitting him to take advantage of any exculpatory evidence.” Leka, 257 F.3d at 100 (citing United States v. LeRoy, 687 F.2d 610, 618 (2d Cir.1982)). While the Brady rule requires the Government to turn over favorable information that it knows about, it does not require the Government to seek out such information like a “private investigator and valet ... gathering evidence and delivering it to opposing counsel.” United States v. Tadros, 310 F.3d 999,1005 (7th Cir.2002).
At the same time, the Government cannot hide Brady material as an exculpatory needle in a haystack of discovery materials. See United States v. Skilling, 554 F.3d 529, 577 (5th Cir.2009), aff'd in part and vacated in part on other grounds, 561 U.S. 358, 130 S.Ct. 2896, 177 L.Ed.2d 619 (2010) (suggesting that Brady violations related to voluminous open file discovery depend on “what the government does in addition to allowing access to a voluminous open file”); see also United States v. Hsia, 24 F.Supp.2d 14, 29-30 (D.D.C.1998) (“The Government cannot meet its Brady obligations by providing ... 600,000 documents and then claiming that [the defendant] should have been able to find the exculpatory information[.]”).
Viewing all of the facts together, the Government’s Brady disclosure of [240]*240Jean Jacques’s inconsistent description of the robbers to Parchen did not occur until the sidebar during Jenkins’s testimony on the defense case. That is too late. See, e.g., Leka, 257 F.3d at 101-03 (“Disclosure first made during the defendant’s case is often of lesser utility----The opportunity for use under Brady is the opportunity for a responsible lawyer to use the information with some degree of calculation and forethought.”); St. Germain v. United States, Nos. 03-cv-8006 (CM), 99-cr-339 (CM), 2004 WL 1171403, at *18 (S.D.N.Y. May 11, 2004) (“[D]efense strategies are largely formed prior to trial ... and the necessary predicate is that the strategies selected were chosen after careful consideration of all constitutionally-compelled disclosure.”).
Thomas’s defense strategy for the Jean Jacques robbery was to paint Jean Jacques’s identification of him as unreliable. He sought to convince the jury that the description memorialized by Jenkins, which would preclude Thomas as one of the robbers, was accurate. But that argument rested chiefly on the relative credibility of Jean Jacques and Jenkins.
As this Court observed in its November Order, Jenkins’s conduct in administering the identification procedures in this case was “unprofessional.” Russell, 2012 U.S. Dist. LEXIS 172206, at *28. She disregarded proper police procedure in administering the identification — or as Jenkins recasts it, she made a “mistake.” (T, Tr. at 1031.) This Court did not rely on her testimony at the hearing to resolve any disputed fact. Russell, 2012 U.S. Dist. LEXIS 172206, at *4 (citing Jenkins’s testimony only for the uncontested propositions that she recovered stolen cellular telephones and arrested someone). It is unlikely that a jury would find her testimony more reliable than Jean Jacques’s, who was a powerful witness. Parchen, however, did not suffer from the same credibility infirmities as Jenkins. But until the Government’s Brady disclosure on the defense’s case, Thomas could not have known that Parchen was an essential witness.
The Government contends that it provided Parchen’s report as part of 3500 material for other hearing and trial witnesses. While that is true, it glosses over the importance of the manner of the Government’s disclosure. See, e.g., Gil, 297 F.3d at 106 (labeling Brady evidence as 3500 material and producing it as part of a large 3500 production on the eve of trial constitutes suppression); United States v. Breit, 767 F.2d 1084, 1090 n. 4 (4th Cir.1985) (the Government may not discharge its Brady obligation merely by tendering a witness without providing any indication that the witness’s testimony may be helpful to defense).
At bottom, the Government’s argument is that the Brady evidence is not suppressed here because Thomas did not exercise due diligence. But that argument fails because the Government stipulated at the hearing that Jenkins authored the report and never corrected that error, even after it interviewed Parchen in preparation for trial. Penalizing Thomas for failing to fact check the Government’s erroneous stipulation would turn due process on its head. See Bagley, 473 U.S. at 682-83, 105 S.Ct. 3375 (“[T]he more specifically the defense requests certain evidence, thus putting the prosecutor on notice of its value, the more reasonable it is for the defense to assume from the nondisclosure that the evidence does not exist, and to make pretrial and trial decisions on the basis of this assumption.”); cf. Milke v. Ryan, 711 F.3d 998, 1017-18 (9th Cir.2013) (finding that the availability of court documents in the public record, which contained Brady material, did not diminish the government’s obligation to produce them); Payne, 63 F.3d at 1208-09 (reject[241]*241ing claim that the Government had “no duty to disclose” Brady material that was in public court records because defense counsel was not aware of facts “that would have required him to discover the [Brady material] through his own diligent investigation”).
The Government’s fallback position that “Parchen essentially copied [Jenkins’s handwritten notes in preparing [his] report,” (8/9/13 Tr. at 22), is indefensible. Parchen admitted authoring the report, interviewing Jean Jacques, and being told that the shorter robber had braids. (T. Tr. at 1053-54.) Jenkins also denied that Parchen got the information from her. (T. Tr. at 1043-44.) But “questions about the reliability of [] exculpatory information” are judgment calls for Thomas and his counsel, not the Government; “to allow otherwise would be to appoint the fox as henhouse guard.” DiSimone v. Phillips, 461 F.3d 181, 195 (2d Cir.2006).
The Government also argues that it had no obligation to produce the report as part of Parchen’s 3500 material because “[t]he subject matter of [his] testimony concerned tracing the [stolen cellular telephone] — and the Government produced 3500 material concerning that topic.” (Gov’t Br. at 23.) But on direct, the Government asked Parchen, “And what are some of the things that you did to assist Detective Jenkins?” And he responded, “At the scene I interviewed one of the victims. I also helped her track a stolen phone that was stolen in this case.” (T. Tr. at 972.) Thus, the Jencks Act required production of Parchen’s report of the victim-interview (DD5 Follow-Up No. 4) because it “relate[s] generally to the events and activities” of his testimony and is not “incidental or collateral.” United States v. Pacelli, 491 F.2d 1108, 1119 (2d Cir.1974) (internal citations omitted); see also 18 U.S.C. § 3500(b). The Government’s decision not to produce DD5 Follow-Up No. 4 as part of Parchen’s 3500 material also kept alive the notion that Jenkins authored the report.
The Government’s argument conflates its Jencks Act and Brady obligations. While those responsibilities overlap at times, they are distinct legal concepts. The Jencks Act is concerned with discovery to be produced by the Government. Brady is concerned with fairness. Whether the Government had a Jencks Act obligation vis-a-vis the Parchen report is of no moment to its due process responsibility to produce it.
D. Material Evidence and Prejudice
The only remaining question is whether the Government’s suppression of Brady material resulted in prejudice. In other words, is there “a reasonable probability that the Government’s suppression affected the outcome of the case” or could the suppressed evidence “reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict[?]” Coppa, 267 F.3d at 135 (internal citations omitted).
Thomas’s conviction on the Jean Jacques robbery charge hinges on the reliability of Jean Jacques’s identification of him. The credibility of witnesses is of paramount concern for any conviction. See Napue v. Illinois, 360 U.S. 264, 269, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959) (“The jury’s estimate of the truthfulness and reliability of a given witness may well be determinative of guilt or innocenee[.]”) Close scrutiny is invited for a “conviction based on a single witness’s identification.” Rodriguez v. Hoke, 928 F.2d 534, 537 (2d Cir.1991); see also Young v. Conway, 698 F.3d 69, 79-80 (2d Cir.2012) (acknowledging the “dangers of testimony based purely on eyewitness identification” but also noting that “much eyewitness identifi[242]*242cation testimony is reliable and is and should be routinely accepted by juries”); United States v. Petrillo, 821 F.2d 85, 90 (2d Cir.1987) (impeachment evidence is material if the witness whose testimony is attacked “supplied the only evidence linking the defendant ] to the crime”).
As this Court observed, Thomas’s strategy was to convince the jury that Jean Jacques’s identification was unreliable. But his proof depended on showing the jury that Jenkins was careless in following proper identification procedures, yet spot-on in her written notes regarding Jean Jacques’s description on the night of the robbery. Had Thomas been able to investigate Jean Jacques’s statement to Parchen and devise an informed trial strategy, he could have reasonably “put the whole case in ... a different light.” Coppa, 267 F.3d at 135. Thomas’s lack of an adequate opportunity to investigate or implement that strategy because of the Government’s Brady suppression undermines this Court’s confidence in the verdict on Counts 3 and 5.
While Thomas could have recalled Jean Jacques to question him about the description he gave to Parchen, that ad hoc improvisation involved substantial risks — including reinforcement of Jean Jacques’s powerful testimony — but uncertain reward. Thomas cannot be faulted for declining to engage in that perilous calculus. See Leka, 257 F.3d at 101 (“When such a disclosure is first made on the eve of trial, or when trial is under way, the opportunity to use it may be impaired. The defense may be unable to divert resources from other initiatives and obligations that are or may seem more pressing. And the defense may be unable to assimilate the information into its case.”); United States v. Washington, 294 F.Supp.2d 246, 250 (D.Conn.2003) (finding that the Government’s failure to disclose evidence impeaching the central witness until after the first day of trial prejudiced defendant because the late disclosure prevented defense counsel from investigating and planning overall trial strategy).
Recalling a witness after a late Brady disclosure is no substitute for thoughtful preparation and a considered strategy. Brady material must be provided to a defendant “in time for its effective use at trial.” Coppa, 267 F.3d at 135 (emphasis added); see also Grant v. Alldredge, 498 F.2d 376, 382 (2d Cir.1974) (refusing to “infer from the failure of defense counsel, when surprised at trial, to seek time to gather other information on [the suppressed witness], that defense counsel would have by-passed the opportunity had the prosecutor apprised him of the [evidence] at a time when the defense was in a reasonable pre-trial position to evaluate carefully all the implications of that information”).
Brady does not require a “strong” or “overwhelming” probability of a different outcome, only a “reasonable probability that the Government’s suppression affected the outcome of the case.” Coppa, 267 F.3d at 135. As the Court in Kyles noted,
a showing of materiality does not require demonstration by a preponderance that disclosure of the suppressed evidence would have resulted ultimately in the defendant’s acquittal____Materiality is a ‘reasonable probability’ of a different result, and the adjective is important. The question is not whether the defendant would more likely than not have received a different verdict with the evidence, but whether in its absence he received a fair trial, understood as a trial resulting in a verdict worthy of confidence.
[243]*243Kyles, 514 U.S. at 434, 115 S.Ct. 1555 (internal citations omitted). The Government’s suppression of critical impeaching evidence in this case meets that “reasonable probability” standard.
The Government’s Rule 29 argument illustrates just how its untimely Brady disclosure prejudiced Thomas. It argues that even without any identification of Thomas by Jean Jacques, the evidence is sufficient to sustain a conviction for that robbery. In support of its argument, the Government points to: (1) the recovery of the stolen cellular telephone days after the robbery at the home of one of the robbery crew members, (2) the testimony of cooperating witness Kirk Bent that Thomas boasted about his involvement in a robbery of someone who ships goods to another country, (3) Thomas’s arrest the day after the robbery with guns matching the description of the guns from the robbery, and (4) that this robbery fit the modus operandi of other robberies committed by Thomas’s robbery crew.
Without an identification of Thomas, at best the jury could infer that the cellular telephone was either stolen or purchased by one of the men who visited the home of a robbery crew member. That evidence alone would be insufficient to allow the jury to conclude Thomas was a robber, not a purchaser. Without an identification of Thomas, Bent’s testimony provides scant support that Thomas was boasting about this crime, that the firearms recovered the day after the robbery were used by Thomas the night before, or that the modus operandi was specific enough to conclude that Thomas was the robber. Thomas and his crew committed many robberies and each of those arguments could just as easily refer to another robbery or another robber as it could to Thomas. And while the modus operandi helps establish the identity of the robbery crew, the crew’s methods were not regimented enough for a rational trier of fact to conclude that Thomas always took an active role in their robberies.
With Jean Jacques’s identification of Thomas, these pieces of evidence are powerful corroboration of that identification. Without it, they are seemingly disconnected and random. While such evidence might pass Rule 29 muster, this Court harbors significant doubt that a jury would convict on that evidence alone. That doubt satisfies the standard for Brady prejudice to “put the whole case in such a different light as to undermine confidence in the verdict.” Coppa, 267 F.3d at 135. The result of this Brady motion should be noncontroversial — the Government’s suppression of favorable evidence undermining its key witness violates due process. And justice is not served by allowing such a conviction to stand.
II. Rule 29 Motions
A. Legal Standard
Rule 29 permits a court to set aside a conviction and “enter a judgment of acquittal of any offense for which the evidence is insufficient to sustain a conviction.” Fed.R.Crim.P. 29(a). A defendant “bears a heavy burden” on a Rule 29 motion. United States v. Ainar-Marshall, 336 F.3d 167, 171 (2d Cir.2003). If “after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt,” then the motion must be denied. Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979) (emphasis in original). “Pieces of evidence must be viewed not in isolation but in conjunction,” United States v. Matthews, 20 F.3d 538, 548 (2d Cir.1994), and when evaluating the sufficiency of the evidence, courts do not assess witness credibility, resolve [244]*244inconsistent testimony against the verdict, or otherwise weigh the significance of the evidence. See United States v. Autuori, 212 F.3d 105, 114 (2d Cir.2000). “The weight of the evidence is a matter for argument to the jury.” Matthews, 20 F.3d at 548.
B. The Jean Jacques Robbery Related Charges
Thomas does not contend that the trial evidence was insufficient as to the Jean Jacques robbery related charges. Instead, he argues that the Government failed to comply with its Brady obligations, and had it done so, this Court would have suppressed Jean Jacques’s identification of Thomas. Thomas avers the trial evidence without that identification would have been insufficient. The Second Circuit has suggested in dicta that Brady applies to suppression hearings. See, e.g., United States v. Nelson, 193 Fed.Appx. 47, 50 (2d Cir.2006) (“Whether Brady and its progeny require disclosures in advance of pre-trial hearings is an open question in this circuit____ [But] in cases that pivot on the outcome of the suppression hearing ... it seems clear that postponing disclosure until after that hearing will prevent the defense from having key information”).
Assuming that Brady applies to the hearing does not help Thomas. This Court presided over the suppression hearing and trial. Thus, it can determine whether the Brady violation had a “reasonable probability” of affecting the outcome of the suppression hearing. Coppa, 267 F.3d at 135. The Brady violation would not have altered this Court’s finding that Jean Jacques’s identification of Thomas was independently reliable. Because this Court again denies Thomas’s motion to suppress, his Rule 29 motion related to the Jean Jacques robbery is also denied.
C. Conspiracy to Kidnap Curtis Jackson
Thomas contends that the jury “signaled its rejection of the testimony of the alleged kidnapping victim, Curtis Jackson,” when it acquitted Thomas of the substantive kidnapping charged in Count 8. (Def.’s Mem. in Supp. of his Motion for Post-Conviction Relief Under Rules 29 and 33, at 15.) From that, he argues that this Court should exclude Jackson’s testimony when viewing the sufficiency of the evidence at trial. But speculating about why a jury reached its verdict is prohibited if the evidence in support of the conviction is legally sufficient. See United States v. Acosta, 17 F.3d 538, 545 (2d Cir.1994).
Rule 29 forbids a court from assessing witness credibility or weighing the significance of the evidence. “Sufficiency-of-the-evidence review involves assessment by [a] court[] of whether the evidence adduced at trial could support any rational determination of guilt[ ] beyond a reasonable doubt.” United States v. Powell, 469 U.S. 57, 67, 105 S.Ct. 471, 83 L.Ed.2d 461 (1984) (emphasis added). “[T]his review should be independent of the jury’s determination that evidence on another count was insufficient.” Powell, 469 U.S. at 67, 105 S.Ct. 471. A rational trier of fact could have found the Government proved the kidnapping conspiracy (Count 7) beyond a reasonable doubt.
III. Rule 33
Rule 33 permits a court to “vacate any judgment and grant a new trial if the interest of justice so requires.” Fed. R.Crim.P. 33(a). The rule gives a court “broad discretion ... to set aside a jury verdict and order a new trial to avert a perceived miscarriage of justice.” United States v. Sanchez, 969 F.2d 1409, 1413 (2d Cir.1992). To grant a Rule 33 motion, a trial judge “must harbor a real concern [245]*245that an innocent person may have been convicted.” United States v. Lin Guang, 511 F.3d 110, 119 (2d Cir.2007).
Thomas argues that because of the Brady violation on the Jean Jacques robbery, he deserves a new trial on every count. Evidence of the Jean Jacques robbery — and any prejudice relating to the Brady violation — is separate from the evidence of the Magic Pot Bar robbery and kidnapping conspiracy. As this Court instructed the jury, the Government has the burden of proving beyond a reasonable doubt not only that a charged crime was committed, but also that Thomas “is the person who committed that crime.” (T. Tr. at 1252.) The law presumes the jury follows a judge’s instructions. See Richardson v. Marsh, 481 U.S. 200, 206, 107 S.Ct. 1702, 95 L.Ed.2d 176 (1987); United States v. Whitten, 610 F.3d 168, 191 (2d Cir.2010).
Although the jury could have used evidence of the Jean Jacques robbery as proof of the Count 1 robbery conspiracy, the Indictment charged two overt acts as proof of that count: the Jean Jacques robbery and the Magic Pot Bar robbery. Because proof that Thomas committed the Magic Pot Bar robbery was overwhelming, this Court is not concerned that an innocent person was convicted of that robbery (Counts 2 and 4) or the robbery conspiracy (Count 1). See Lin Guang, 511 F.3d at 119.
CONCLUSION
For the foregoing reasons, Dwayne Thomas’s motion for a new trial on Counts 3 and 5 is granted. Thomas’s motions for a new trial on the remaining counts and a judgment of acquittal on Count 7 are denied. The Clerk of the Court is directed to terminate the motions pending at ECF No. 112.
SO ORDERED.