DECISION AND ORDER
HONORABLE RICHARD J. ARCARA, UNITED STATES DISTRICT JUDGEThe Defendant in this case is a Buffalo Police Officer who is alleged to have used excessive force against three individuals on three different occasions over a five-year period. The Defendant is also accused of making a false statement about one of those incidents in a Buffalo Police Department use-of-force form. The case is before the Court on several issues: (1) the Defendant’s objections to Magistrate Judge Schroeder’s Report and Recommendation, which recommends denying each of the Defendant’s motions to dismiss; (2) the Defendant’s motion for a hearing pursuant to Kastigar v. United States, 406 U.S. 441, 92 S.Ct. 1653, 32 L.Ed.2d 212 (1972), based on statements the Defendant made to the Buffalo Police Department’s Professional Standards Division; and (3) the Defendant’s motion for severance.
The Court adopts Magistrate Judge Schroeder’s Report and Recommendation in its entirety. The Court also concludes that the Defendant is entitled to a hearing pursuant to Kastigar v. United States. For the reasons explained below, the Court will hold a Kastigar hearing either during trial and/or after trial. Finally, the Court reserves decision on the Defendant’s severance motion. Closer to trial, the Court will decide, based on the factors discussed below, whether to sever trial of Counts 1, 2, and 3 from trial of Count 4.
BACKGROUND1
In the early-morning hours of November 27, 2014—Thanksgiving day—the Defendant was working in the entertainment district around Chippewa Street in downtown Buffalo. A local television news cameraman, who was in the area to film Thanksgiving-eve festivities, filmed what the Defendant refers to as a “physical confrontation” between him and D.F. Docket No. 14-1 at 1. The Defendant contends that the video of that incident “brought a wave of negative media attention to the Buffalo Police Department and Officer Krug” at a time when “there was a movement across the country to prosecute law enforcement for use of excessive force.” Id.
In December 2014, after the video was broadcast on local news stations, the Buffalo Police Department referred the matter to the Federal Bureau of Investigation (FBI). Docket No. 17 at 9. The FBI, in turn, presented the matter to the U.S. Attorney’s Office for the Western District of New York. And the U.S. Attorney’s Office then presented the matter to a grand jury impaneled in May 2015. As part of its investigation, the grand jury subpoenaed the Defendant’s Buffalo Police Department personnel file. According to the Government, the file contained, among other things, two “civilian complaints against [the Defendant] which alleged excessive force.” Id. The first of those incidents al*240legedly occurred on August 29, 2010, and the second allegedly occurred several months later, on February 2, 2011.
The Defendant’s motions to dismiss the superseding indictment are based largely on the manner and sequence in which the Government charged him with each of the three alleged excessive force incidents. On August 18, 2015, while the grand jury was still reviewing the 2014 incident, the Government filed a criminal complaint charging that incident violated 18 U.S.C. § 242. Then, nine days later, the grand jury handed up the first indictment in this case. That two-count indictment charged the Defendant with (1) a § 242 violation based on the 2010 incident, and (2) falsification of a record in a federal matter, in violation of 18 U.S.C. § 1519, based on the allegation that the Defendant made a false statement regarding the 2010 incident in a Buffalo Police Department use-of-force form.
One week later, on September 3, 2015, the grand jury handed up a superseding indictment, which is now the operative charging instrument in this case. The superseding indictment contains the same charges as the first indictment, but it also includes two additional § 242 allegations: one based on the 2014 incident, and one based on the 2011 incident. Thus, the Defendant is now charged in a four-count superseding indictment alleging falsification of a record in a federal matter and three § 242 violations: one each for the 2010, 2011, and 2014 incidents.
DISCUSSION
1. The Defendant’s objections to Magistrate Judge Schroeder’s Report and Recommendation
The Defendant objects to Magistrate Judge Schroeder’s recommendations to deny all three of the Defendant’s motions to dismiss. The Court therefore reviews Magistrate Judge Schroeder’s recommendations de novo. See 28 U.S.C. § 636(b)(1).
A. Motion to dismiss the superseding indictment for allegedly prejudicial pre-indictment delay
The Defendant first moves to dismiss the superseding indictment based on the Government’s allegedly prejudicial delay in charging him with the 2010 and 2011 excessive force incidents. The Defendant argues that the Government knew about those incidents at or around the time they occurred, but that “it was only when the government was trying to strengthen” its case for the 2014 incident “that [it] decided to prosecute both” the 2010 and 2011 incidents. Docket No. 26 at 7-8. According to the Defendant, this delay—which, in the case of the 2010 incident, was just two days short of the five-year statute of limitations—“gave the government a significant tactical advantage and unfairly prejudiced Officer Krug.” Id. at 8.
It is well settled that “statutes of limitations provide the primary protection against delay” in indictment, “with the Due Process Clause [acting] as a safeguard against fundamentally unfair prosecutorial conduct.” Betterman v. Montana, — U.S. —, 136 S.Ct. 1609, 1613, 194 L.Ed.2d 723 (2016). See also United States v. Lovasco, 431 U.S. 783, 789, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977) (noting that statutes of limitations “provide ‘the primary guarantee against bringing overly stale criminal charges,’ ... and that the Due Process Clause has a limited role to play in protecting against oppressive delay”) (quoting United States v. Marion, 404 U.S. 307, 322, 92 S.Ct. 455, 30 L.Ed.2d 468 (1971)). Given this high standard, “timely brought criminal prosecutions are only rarely dismissed.” United States v. Cornielle, 171 F.3d 748, 752 (2d Cir. 1999). And, when timely brought criminal prosecutions are dismissed, it is only where a defendant has met the “heavy burden of proving both that he suffered actual prejudice because of the alleged pre-indictment delay and *241that such delay was a- course intentionally pursued by the government for an improper purpose.” Id. (quotation marks omitted, emphasis in original).
In its papers and at oral argument, the Government largely conceded that, at least as to the 2010 incident, the Defendant has suffered prejudice in the form of “the loss of documentary evidence or the unavailability of key witnesses.” Cornielle, 171 F.3d at 752. See Docket No. 26 . at 9 (Def s Objections) (listing evidence that is missing from the 2010 incident); Docket No. 17 at 11 (Gov’t’s Response to Defs Mtn.) (“While the defense claims that evidence, including possible eye witnesses may have been lost, and witness recollection diminished, the government suffered the same, if not greater, loss.”); Docket No. 40 (Trans, of Oral Argument; hereinafter “Tr.”) at 2:4-14. But'even with this prejudice, the Defendant has not established that the Government’s delay in charging him with the 2010 and 2011 incidents “departs from fundamental notions of ‘fair play’ ” or was “a course intentionally pursued by the government for an improper purpose.” Cornielle, 171 F.3d at 752 (quoting Lovasco, 431 U.S. at 796, 97 S.Ct. 2044).
To try to- show otherwise, the Defendant notes that, for several years, the “Department of Justice has - been intimately involved in the ‘use of force’ policies and procedures of the Buffalo Police Department.” Docket No. 26 at 7 (citing Amended Memorandum of Agreement Between the United States Department of Justice and the City of Buffalo, et al., available at Docket No. 14-5). For example, the Defendant points to “systems [that] were implemented to enable the Department of Justice to supervise and review uses of force by Buffalo Police officers.” Id. Thus, the Defendant argues, “[i]t can be inferred that the Department of Justice had the ability to review and analyze Officer Krug’s Buffalo Police Department Use of Force Reports from his arrest of [M.W.] in 2010 and his arrest of [D.R.] in 2011.” Id. The Defendant argues that, as a result, the Court could infer that DOJ was “aware of both incidents and did not conclude that Officer Krug used unreasonable and unnecessary force, because he was never disciplined or criminally charged.” Id. Rather than charge the Defendant with the 2010 and 2011 incidents at or around the time they occurred, the Defendant argues that the Government waited until it “was trying to strengthen” its case as to the 2014 incident.
In response, the Government states that, although the 2010 and 2011 incidents were both nearly four years old when the Government began investigating them, “the actual FBI and Grand Jury investigations were completed in approximately 8 months after being initiated. The government was not responsible for the time period between the 2010 and 2011 events and the commencement of the investigations. Until the matter was referred to the [U.S. Attorney’s Office] in December 2014, the government had no knowledge of the events.” Docket No. 17 at 11. Magistrate Judge Schroeder found that the Government’s explanation for the delay in charging the Defendant with the 2010 and 2011 incidents “is legitimate and reasonable.” Docket No. 21 at 7. The Court agrees with Magistrate Judge Schroeder that the Government’s proffered explanation is both legitimate and reasonable.
The Court also agrees with Magistrate Judge Schroeder that the Defendant’s argument to the contrary relies largely on speculation. The Government, of course, could not have known in 2010 and 2011 that, nearly four years later, the Defendant would allegedly use excessive force on a third occasion.2 It therefore strains plau*242sibility to suggest that the Government declined to prosecute the 2010 and 2011 incidents based on the belief that those incidents might become useful, years later, in a prosecution of the Defendant for similar conduct. Thus, the Government’s delay in charging the 2010 and 2011 incidents could not have been “an intentional device to gain [a] tactical advantage” over the Defendant. Marion, 404 U.S. at 324, 92 S.Ct. 455.
To conclude otherwise on the facts of this case would inappropriately hamper the Government’s legitimate exercise of prosecutorial discretion in at least two ways. First, the Defendant’s argument would effectively impose on the Government a duty to seek an indictment shortly after an alleged crime occurs, at the risk of being unable to ever again seek an indictment. It is “obvious,” however, “that prosecutors are under no duty to file charges as soon as probable cause exists but before they are satisfied they will be able to establish the suspect’s guilt beyond a reasonable doubt.” Lovasco, 431 U.S. at 791, 97 S.Ct. 2044. Second, and relatedly, the Defendant’s theory would also “unnecessarily impair the performance of a core executive constitutional function”: deciding whether and when to seek an indictment. United States v. Armstrong, 517 U.S. 456, 465, 116 S.Ct. 1480, 134 L.Ed.2d 687 (1996). That decision “requires consideration of a wide range of factors in addition to the strength of the Government’s case, in order to determine whether prosecution would be in the public interest.” Lovasco, 431 U.S. at 794, 97 S.Ct. 2044. Thus, identifying whether and when a prosecution is appropriate is not a question a court may, or should, decide. Rather, such decisions belong—both constitutionally and institutionally—to the Attorney General and the United States Attorneys, who have “broad discretion to enforce the Nation’s criminal laws.” Armstrong, 517 U.S. at 464, 116 S.Ct. 1480 (quotation marks omitted).
Thus, the Court overrules the Defendant’s objection to Magistrate Judge Schroeder’s recommendation to deny the Defendant’s motion to dismiss for allegedly prejudicial delay in charging the Defendant. The Defendant’s motion to dismiss on that ground is therefore denied.
B. Motion to dismiss the superseding indictment for alleged grand jury taint
The Defendant’s second motion seeks dismissal of the superseding indictment for alleged grand jury taint. According to the Defendant, the Government improperly tainted the grand jury’s independence by filing a criminal complaint charging the Defendant with the 2014 incident while the grand jury was still considering whether to indict him for that incident. The Defendant argues that the Government’s strategy was to “file a criminal charge before the vote of the Grand Jury to advise the jurors that the government will prosecute regardless of the vote.” Docket No. 26 at 11. “Without question,” the Defendant argues, “this strategy influenced the jury. What is more, the government obtained increased media exposure compounding its influence on the Grand Jury.” Id.
Dismissal of an indictment is “the most drastic remedy” a Court may use to eliminate prejudice to a defendant or to disincentivize improper prosecutorial conduct. United States v. Dyman, 739 F.2d 762, 768 (2d Cir. 1984) (quotation marks omitted). The remedy of dismissal, therefore, “is rarely used.” Id. “[A]s a general matter, a district court may not dismiss an indictment for errors in grand jury pro*243ceedings unless such errors prejudiced the defendant ].” Bank of Nova Scotia v. United States, 487 U.S. 250, 254, 108 S.Ct. 2369, 101 L.Ed.2d 228 (1988). This is an exceedingly high standard: “dismissal of the indictment is appropriate only if [1] it is established that the violation substantially influenced the grand jury’s decision to indict, or [2] if there is grave doubt that the decision to indict was free from the substantial influence, of such violations.” Id. at 256, 108 S.Ct. 2369 (quotation marks omitted, emphasis added). Thus, the Defendant must show both that the Government’s decision to file a complaint before seeking an indictment was improper and that the Defendant was prejudiced by that decision.3
The Government’s alleged impropriety, the Defendant argues, was to file a criminal complaint concerning the 2014 incident before seeking an indictment; the result, according to the Defendant, was that the grand jury’s independence was compromised by the media coverage which resulted from the Defendant being charged. In response, the Government relies heavily on United States v. Silver, 103 F.Supp.3d 370, 378 (S.D.N.Y. 2015). In Silver, the government began its prosecution by filing a complaint against the defendant, which was followed by what the court called a “media blitz orchestrated by the U.S. Attorney’s Office.” Id. at 373. After the media campaign, the government presented its case to the grand jury, which handed up an indictment. The court strongly chastised the government’s conduct, but it ultimately denied the defendant’s motion to dismiss. The court found that the government had not engaged in misconduct by first charging the defendant via complaint, even if doing so resulted in significant media interest. As the court noted, “[t]he Government has discretion to proceed via complaint pursuant to the Federal Rules of Criminal Procedure.” Id. at 378.
The Defendant does not contest this point. Rather, at oral argument he attempted to distinguish Silver by noting that, in Silver, the grand jury was impaneled after the criminal complaint was filed, while in this case, the criminal complaint was filed while the grand jury was deliberating. See Tr. 10:17-11:6. But this difference is irrelevant to the first question the Court must answer before dismissing an .indictment: whether there was an error or other impropriety before the grand jury. See Bank of Nova Scotia, 487 U.S. at 256, 108 S.Ct. 2369. Put differently, the point in an investigation (if ever) at which the Government chooses to file a criminal complaint—even when, as in this case, the Government has little explanation for its decision—is not the sort of alleged misconduct which warrants dismissing an indictment. A district court’s “supervisory power” over the grand jury “can be used to dismiss an indictment because of misconduct before the jury, at least where that misconduct amounts to a violation of one of those few, clear rules which were carefully drafted and approved by [the Supreme Court] and by Congress to ensure the integrity of the grand jury’s functions.” United States v. Williams, 504 U.S. 36, 46, 112 S.Ct. 1735, 118 L.Ed.2d 352 (1992) (quotation marks omitted). The Defendant points to nothing suggesting that the Government’s decision to file a complaint before seeking an indictment violates one of those “few, clear rules.” The Defendant, *244therefore, can identify no misconduct of the sort that might justify dismissing the superseding indictment.
And even if the Government’s conduct was improper, “there is no evidence that” the Government’s decision to file a criminal complaint while the grand jury was deliberating “ ‘substantially] influenced’ the grand jury’s decision to indict.” Silver, 103 F.Supp.3d at 379 (quoting Bank of Nova Scotia, 487 U.S. at 255-56, 108 S.Ct. 2369). This conclusion is driven in large part by the fact that the grand jury’s function is quite distinct from that of the petit jury. In United States v. Nunan, 236 F.2d 576 (2d Cir. 1956), for instance, the Second Circuit considered whether the district court should have quashed an indictment based on a “series of sensational newspaper articles and radio and television publicity” that ran “both prior to and during the sessions of the Grand Jury,” Id. at 593. This publicity, the court noted, was “sensational in character,” and “much of it was unfair, misleading, and, at least to some extent, untrue and unwarranted.” Id. However, the court concluded that the publicity was insufficient to dismiss the indictment because, among other reasons, a grand jury “is not confined to a passive role,” as a petit jury is; a grand jury “may and often should proceed on its own initiative.” Id. Thus, the court had less reason to be concerned by whether the grand jury was affected by pre-indictment publicity: even if the grand jury “is induced to [indict] by newspaper reports,” that action “forms a continuum with [the grand jury’s] historic function of ferreting out crime and corruption, and is in no way inconsistent with its duty to decide on and in accordance with the evidence adduced before it.” Id. In other words, “the role of the grand jury historically has differed from that of a petit jury and the same freedom from outside influences” that a court must enforce as to a petit jury “is not required” as to a grand jury. United States v. Myers, 510 F.Supp. 323, 326 (E.D.N.Y. 1980) (quotation marks omitted).
Thus, in this case, even assuming that the Government’s decision to file a complaint before seeking an indictment was improper, the Defendant points to no evidence that the grand jury was improperly influenced by that decision, particularly in light of the principles discussed above. Thus, the Court overrules the Defendant’s objection to Magistrate Judge Schroeder’s recommendation on this issue. The Defendant’s motion to dismiss based on alleged grand jury taint is therefore denied.
C. Motion to dismiss Count 2 of the superseding indictment for facial insufficiency
The Defendant’s final motion to dismiss seeks dismissal of Count 2 for facial insufficiency. Count 2 charges the Defendant with a violation of 18 U.S.C. § 1519, which, in relevant part, states:
Whoever knowingly ... conceals, covers up, falsifies, or makes a false entry in any record, document, or tangible object with the intent to impede, obstruct, or influence the investigation or proper administration of any matter within the jurisdiction of any department or agency of the United States ... or in relation or contemplation of any such matter or case, shall be fined under this title, imprisoned not more than 20 years, or both.
Count 2 alleges that, following the 2010 incident, the Defendant made a false entry in a Buffalo Police Department use-of-force form “in that the defendant checked the ‘No’ box in response to whether he had used an impact weapon on [the victim], when in truth and in fact, as the defendant then and there knew, the defendant had struck [the victim] with an impact weapon.” Docket No. 5 at 3 ¶ 3. Count 2 further alleges that the FBI “was an agency of the United States Department of Justice and *245responsible for investigating the deprivation of rights secured and protected by the Constitution and laws of the United States by persons acting under color of law.” Id. ¶ 2.
An indictment must contain a “plain, concise, and definite written statement of the essential facts constituting the offense charged.” Fed. R. Crim. P. 7(c)(1). An indictment is therefore “sufficient if it, first, contains the elements of the offense charged and fairly informs the defendant of the charge against which he must defend, and, second, enables him to plead an acquittal or conviction in bar of future prosecutions for the same offense.” United States v. Alfonso, 143 F.3d 772, 776 (2d Cir. 1998). The indictment therefore “neecj do little more than to track the language of the statute charged and state the time and place (in approximate terms) of the alleged crime.” Id.
The Defendant first moves to dismiss Count 2 for facial insufficiency, arguing that he struck the victim in the 2010 incident -with a flashlight, and that a flashlight is not an “impact weapon,” as that term is used in the Buffalo Police Department’s use-of-force form. This argument is premature. It raises a factual question that goes to the heart of whether the Defendant made a false statement or otherwise violated § 1519. Federal Rule of Criminal Procedure 12(b)(3), however, provides that an indictment may be dismissed before trial only if “the motion can be determined without a trial on the merits.” Whether a flashlight is an “impact weapon,” as that term is used in the use-of-force form, is a question a jury must decide.
The Defendant’s second argument is that the Buffalo Police Department’s use-of-force form is not, and could not be, related to “the investigation or proper administration of any matter within the jurisdiction of any department or agency of the United States.” 18 U.S.C. § 1519. The statute the Defendant is alleged to have violated in Count 2, 18 U.S.C. § 1519, was enacted as part of the Sarbanes-Oxley Act of 2002. See Yates v. United States, — U.S. —, 135 S.Ct. 1074, 1079, 191 L.Ed.2d 64 (2015). Among the changes it made to then-existing obstruction law, § 1519 “expanded prior law by including within the provision’s reach ‘any matter within the jurisdiction of any department or agency of the United States.’ ” Id. at 1081 (quoting § 1519). Although § 1519 was “designed to protect investors and restore trust in financial markets,” id. at 1079, the Second Circuit has observed that the statute’s language has a much more expansive reach: it “ ‘is meant to apply broadly to any acts to destroy or fabricate physical evidence so long as they are done with the intent to obstruct, impede or influence the investigation or proper administration of any matter, and such matter is within the jurisdiction of an agency of the United States, or such acts done either in relation to or in contemplation of such a matter or investigation.’ ” United States v. Rowland, 826 F.3d 100, 109 (2d Cir. 2016) (quoting S. Rep. No. 107-146, at 14 (2002)) (emphasis omitted). Underscoring its broad sweep, § 1519 does not require “knowledge of a pending federal investigation or proceeding,” nor does § 1519 even “require the existence or likelihood of a federal investigation.” United States v. Gray, 642 F.3d 371, 378-79 (2d Cir. 2011). See also United States v. Moyer, 674 F.3d 192, 206-11 (3d Cir. 2012) (discussing the expansive reach of § 1519 in a case involving a police officer charged "with “knowingly falsifying police reports with the intent to impede, obstruct, or influence the investigation into a matter within the jurisdiction” of the FBI).
With this background, Count 2 easily satisfies Rule 7(e)(l)’s pleading standard. As noted, an indictment is facially sufficient under Rule 7 if it, among other *246things, “contains the elements of the offense charged and fairly informs the defendant of the charge against which he must defend.” Alfonso, 143 F.3d at 776. Count 2 does this by “tracking] the language of the statute charged,” and by “stat[ing] the time and place (in approximate terms) of the alleged crime.” Id. Count 2 alleges that the Defendant allegedly made a false statement in a Buffalo Police Department use-of-force form on or about August 29, 2010. And it alleges, as § 1519 requires, that the FBI “was an agency of the United States Department of Justice” that was and is “responsible for investigating the deprivation of rights secured and protected by the Constitution and laws of the United States by persons acting under color of law.” Docket No. 5 at 3 ¶2. The superseding indictment therefore adequately alleges that the Defendant made a false statement on a form that might be relevant to an investigation within the jurisdiction of the FBI—that is, an investigation of a violation of 18 U.S.C. § 242.
Thus, the Court overrules the Defendant’s objection to Magistrate Judge Schroeder’s recommendation to deny the Defendant’s motion to dismiss Count 2. The Court therefore denies the Defendant’s motion to dismiss Count 2.
2. The Defendant’s motion for a hearing pursuant to Kastigar v. United States
The Defendant’s next motion is for a hearing pursuant to Kastigar v. United States, 406 U.S. 441, 92 S.Ct. 1653, 32 L.Ed.2d 212 (1972), to determine whether any of the Government’s evidence was derived, either directly or indirectly, from the Defendant’s statements to the Buffalo Police Department’s Professional Standards Division (PSD). As the Government acknowledges, those statements were “obtained under threat of removal from office.” Garrity v. New Jersey, 385 U.S. 493, 500, 87 S.Ct. 616, 17 L.Ed.2d 562 (1967). They are therefore coerced statements. And, as a result, the Government may not use them against the Defendant “in subsequent criminal proceedings.” Id. The Government concedes this point. See Docket No. 28 at 9 (Gov’t response to Def s objections) (“Implicit in the government’s response” to the Defendant’s Ganity motion “is an acknowledgment that any such statements would not be admissible in its case-in-chief.”) Moreover, not only is the Government precluded from using the Defendant’s “immunized testimony itself,” but the Government is also prohibited from using “any evidence that was tainted ... by exposure to the immunized testimony.” United States v. Slough, 641 F.3d 544, 549 (D.C. Cir. 2011) (citation omitted). Kasti-gar, however, provides a means by which the Government may prosecute an individual despite Garrity’s restrictions. It does so by requiring the Government to show that “the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony.” Kastigar, 406 U.S. at 460, 92 S.Ct. 1653.
The Defendant is not entitled to a Kastigar hearing merely because he has, at some point in the past, provided Garrity statements. But, at the same time, the Defendant’s burden is minimal: he must show only that “in his immunized” statements—that is, in his statements to the Buffalo Police Department’s PSD—“he testified to ‘matters related to the federal prosecution.’ ” United States v. Blau, 159 F.3d 68, 72 (2d Cir. 1998) (quoting Kastigar, 406 U.S. at 460, 92 S.Ct. 1653). See also United States v. Mariani, 851 F.2d 595, 599 (2d Cir. 1988) (noting that Kasti-gar is implicated where a person “is later prosecuted for an offense which was the subject of his testimony given under immunity”) (emphasis added). Neither party suggests that this standard has not been met in this case. The Government, then, *247must “make an affirmative showing that evidence which [it] intend[s] to use at trial ... was derived from legitimate independent sources” other than the Defendant’s Garrity statements. United States v. Tantalo, 680 F.2d 903, 908 (2d Cir. 1982).
To attempt to make this showing, the Government has submitted an affidavit from FBI Special Agent (SA) Tina Taylor. In her affidavit, SA Taylor states that the agents investigating this case received the Defendant’s Buffalo Police Department personnel file, via compact disc, in response to a grand jury subpoena. Docket No. 37 ¶ 5. SA Taylor then states that “[t]he investigating agents immediately gave the compact disc to a paralegal in the office of the Chief Division Counsel of the Buffalo Field Office for that office to conduct a Garrity review to ensure that the investigating agents were not exposed to any Garrity materials.” Id. ¶ 6. Finally, “[a]fter the Garrity review was completed, the investigating agents were provided hard copies of the non-Garrity information.” Id. ¶ 7.
SA Taylor’s affidavit is not enough to meet the Government’s burden under Kastigar. To be sure, the Government may meet its burden with an affidavit, but that affidavit must be “non-conclu-sory in form,” and it must do more than “simply ask the court to rely on the government’s good faith.” United States v. Harloff, 807 F.Supp. 270, 282 (W.D.N.Y. 1992). See also United States v. Nanni, 59 F.3d 1425, 1432 (2d Cir. 1995) (noting that the Government may not meet its burden with a “mere assertion that the immunized testimony was not used”) (internal quotation omitted); Tantalo, 680 F.2d at 908 (holding that Kastigar is “not satisfied by the prosecution’s assertion that immunized testimony was not used”). The Court has absolutely no reason to doubt that the Chief Division Counsel of the FBI’s Buffalo Field Office conducted its Garrity review in good faith and that it provided the investigating agents only with such evidence that, in its view, appeared to be taint-free. SA Taylor’s affidavit, however, still largely “ask[s] the court to rely on the government’s good faith.” Id. That, of course, is insufficient to defeat a defendant’s request for a Kastigar hearing. Moreover, SA Taylor’s affidavit shows only that the investigating agents were not exposed to the Defendant’s Garrity statements. But Kastigar requires the Government to do more than show “the negation of taint.” Kastigar, 406 U.S. at 460, 92 S.Ct. 1653. Kastigar instead requires the Government to make an “affirmative” showing that all of its evidence was free of taint. Id.
Thus, based on the facts currently before the Court, the Defendant is entitled to a Kastigar hearing. The Court, however, has “discretion to determine whether the [Kastigar] hearing should be held before trial ...; during the course of trial as the evidence [is] presented; after verdict by way of a post-trial motion; or ... a combination of these alternatives.” Tantalo, 680 F.2d at 909. The Court will conduct a Kastigar hearing in this case either during and/or after trial. Conducting a Kastigar hearing in this way has several benefits. First, the hearing will likely be more efficient, as it will need to address only the evidence that the Government actually introduces at trial—not all the evidence in the Government’s possession that it might conceivably use at trial. Second, the Court will be able to better evaluate whether a hearing remains necessary, as the Court will then be able to tell “with the benefit of trial evidence whether there [is] a sufficient nexus between the immunized testimony and the federal prosecution to warrant a hearing.” United States v. Helmsley, 941 F.2d 71, 80 (2d Cir. 1991). And third, conducting a Kasti-gar hearing either during and/or after trial may allow the Court, in evaluating the *248source of the Government’s evidence, to rely at least in part on testimony elicited during trial, thus also shortening the Kastigar hearing. Closer to trial, the Court will decide whether the parties may, during trial and outside the presence of the jury, question relevant witnesses on Kastigar-related issues in order to further streamline any remaining post-trial Kasti-gar proceedings.
Thus, the Court grants the Defendant’s motion for a Kastigar hearing.
3. The Defendant’s motion for severance
The Defendant’s final motion seeks to sever the four-count superseding indictment into three trials: one trial concerning the events that occurred in 2010 (Counts 1 and 2);4 one trial concerning the events that occurred in 2011 (Count 3); and one trial concerning the events that occurred in 2014 (Count 4). In support of his motion, the Defendant first argues that, under Rule 8(a), Counts 1 and 2 were improperly joined with Counts 3 and 4. Likewise, the Defendant argues, Count 3 was misjoined with Count 4. In the alternative, the Defendant argues that, even if the superseding indictment’s four Counts were properly joined, he is still entitled to severance under Rule 14(a) in order to lessen the risk of spillover prejudice between the three excessive force Counts. Although they both allow a Defendant to break an indictment into smaller pieces, “[t]he question of the propriety of joinder under Rule 8 and ... relief from prejudicial joinder under Rule 14 are quite different in nature.” United States v. Werner, 620 F.2d 922, 926 (2d Cir. 1980) (Friendly, J,). The Court therefore addresses each argument in turn.
A. Whether joinder was proper under Rule 8(a)
An indictment properly charges a defendant with multiple offenses if those offenses are, among other things, “of the same or similar character.” Fed. R. Crim. P. 8(a). Thus, Rule 8(a)’s plain language provides that joinder is proper not only when multiple alleged crimes are of “the ‘same’ character but also” when multiple alleged crimes are of a “ ‘similar’ character.” Werner, 620 F.2d at 926. Crimes are of a “similar character” when then they bear a “resembl[ance] in many respects,” are “somewhat alike,” or when they otherwise “hav[e] a general likeness.” Id. (quoting Webster’s New Int’l Dictionary (2d ed.)). Thus, for example, in United States v. Werner, the Second Circuit held that four counts were properly joined under Rule 8(a) where the first two counts charged the defendant with theft from a cargo hold at John F. Kennedy International Airport in 1976, and the second two counts charged the defendant with theft from the same cargo hold at the same airport in 1978. Although the two thefts were carried out in different ways, the court held that they were nonetheless “similar” for purposes of Rule 8(a). To conclude otherwise, Judge Friendly observed, “would fail to give effect to the word ‘similar’ succeeding the word ‘same’ ” in Rule 8(a) and would “thus violate an elementary rule of statutory construction.” Id. at 926 (citations omitted).5 Thus, join-*249der under Rule 8(a) is proper as long as the counts “share a ‘general. likeness’ in terms of the conduct and events alleged.” United States v. Rivera, 546 F.3d 245, 253-54 (2d Cir. 2008) (holding that counts involving sexual exploitation of multiple children were properly joined under Rule 8(a) where, over a four-month period, the defendant exploited each child in a generally similar manner).
That standard is met in this case as to the 2010, 2011, and 2014 incidents. Counts 1, 3, and 4 each charge a violation of 18 U.S.C. § 242. And each Count alleges that the Defendant violated § 242 by using unreasonable and excessive force while acting under color of law. This is likely sufficient to conclude that the Counts were properly joined under Rule 8(a). Nonetheless, the Counts are even more similar because the Government’s anticipated proof for each incident “share[s] a general likeness.” Rivera, 546 F.3d at 253-54 (internal quotations omitted). As to the 2010 incident, the Government alleges, generally, that the Defendant struck M.W. in the head with a flashlight while the Defendant and other officers were responding to a report of a domestic disturbance. See Docket No. 17 at 3 (Gov’t’s Response to Defs Omnibus Motion). See also United States v. Gracesqui, No. 10-cr-74 (PKC), 2015 WL 5231168 at *2 (S.D.N.Y. Sept. 8, 2015) (“In its analysis [of ‘same or similar character’ under Rule 8(a)], a court may consider both what is alleged in the indictment as well as the Government’s pretrial representations of the evidence that will be presented at trial.”) (internal citations omitted). The basic facts underlying Count 2, which took place approximately six months later, have a “general likeness” to those in Count 1: the Government alleges that the Defendant, while responding to a call reporting a suspicious person near D.R.’s residence, stopped D.R. as he was walking home from work; that the Defendant placed D.R. against a police car, patted him down, and found a diet pill; that the Defendant kicked out D.R.’s legs, causing D.R. to fall; and that the Defendant then put his knee on D.R.’s chest and hit D.R. with a flashlight around D.R.’s right eye. Id. at 4-5. Finally, as to Count 4 (the 2014 incident), the Government alleges that after helping disperse a crowd around a nightclub, the Defendant threw D.F. against a car, and then to the ground; and that the Defendant then placed his knee on D.F.’s chest while striking D.F. with his night stick and yelling at D.F. to “get the f— up.” Id. at 6-7.
Each of these three alleged'incidents is sufficiently similar, both legally and factually, that joinder is proper under Rule 8(a). To be sure, the manner and circumstances in which the Defendant allegedly used excessive force against each of the three victims is slightly different. However, the three incidents still share a general likeness that justifies joinder under Rule 8(a). In each, the Defendant is alleged to have used excessive force by striking the putative victim with either a flashlight or a nightstick. And in each, the Defendant is alleged to have done so while carrying out his duties as a police officer. In short, the manner in which the Defendant allegedly violated § 242 is generally the same for each of the three excessive force counts. Compare United States v. Tubol, 191 F.3d 88, 94-95 (2d Cir. 1999) (holding that join-der of two robbery counts was improper where the perpetrator of each robbery used a different modus operandi and *250robbed “distinctly different victims”). All Rule 8(a) requires is that the different counts be of the “same or similar character”—not that there be a “precise ... identity” between the counts. Werner, 620 F.2d at 926. Counts 1, 3, and 4 meet that standard. Thus, the three counts were properly joined under Rule 8(a).6
B. Whether severance is warranted under Rule 14(a)
Rule 14(a) addresses a different concern than that at issue in Rule 8(a). “Even though distinct offenses have been properly joined under Rule 8, the court may order separate trials ... under Rule 14 if it appears that the defendant is prejudiced by the joinder.” Werner, 620 F.2d at 928. The question underlying Rule 14 is not whether a defendant “may have a better chance of acquittal in separate trials.” Zafiro v. United States, 506 U.S. 534, 540, 113 S.Ct. 933, 122 L.Ed.2d 317 (1993). Nor is the question (contrary to Rule 14(a)’s language) merely whether a defendant is prejudiced by joint trial. See Werner, 620 F.2d at 928-29 (noting that mere prejudice is insufficient to justify severance under Rule 14 because “any lesser showing would undermine the policies behind Rule 8 and essentially read that rule from the books”). The question, instead, is whether the defendant would suffer “substantial prejudice” by joint trial. United States v. Sampson, 385 F.3d 183, 190 (2d Cir. 2004) (citations omitted, emphasis in original).
The Defendant identifies two sources of potential substantial prejudice that, in his view, justify severance: (1) the possibility that he may want to testify as to some counts, but not others; and (2) the possibility that the jury will use evidence as to one count to infer the Defendant’s guilt on tly other counts. The Court addresses each in turn.
i. Potential prejudice from not testifying
The Defendant’s first argument is insufficient to justify severance. “ ‘[A] mere unexplicated assertion’ of the desire to testify on only one count is not enough to require severance.” Sampson, 385 F.3d at 191 (quoting Werner, 620 F.2d at 930). Instead, a defendant seeking a severance based on his desire to testify as to only some counts must “make[ ] a convincing showing that he has both important testimony to give concerning one count and strong need to refrain from testifying on the other.” Id. (internal quotations omitted). To make this showing, a defendant must “present enough information—regarding the nature of the testimony he wishes to give on one count and his reasons for not wishing to testify on the other—to satisfy the court that the claim of prejudice is genuine and to enable it intelligently to weigh the considerations of economy and expedition in judicial administration against the defendant’s interest in having a free choice with respect to testifying.” Id. (internal quotation marks omitted).
The Defendant has not met this standard. Rather than providing information “regarding the nature of the testimony he wishes to give on one count and his reasons for not wishing to testify on the other,” the Defendant states simply that he “may wish to testify in his defense for Counts 1 and 3, but not for Count 4.” Docket No. 14-1 at 20. Later, in his supplemental brief, the Defendant provides slightly more detail: the Defendant claims *251that he “will testify regarding Count 3 because only he and [D.R.] were at the scene but, on the other hand, Officer Krug will not testify regarding Counts 1 and 4 because other officers at the scene can provide similar testimony and a video of Count 4 speaks for itself.” Docket No. 33 at 4. Although the Defendant explains “his reasons for not wishing to testify on” some counts, Sampson, 385 F.3d at 191, he does not, as he must, provide information “regarding the nature of the testimony” he intends to provide. See also id. (providing an example of the type of proffered testimony that might justify a severance). Thus, the Court has no way of evaluating whether the Defendant would suffer substantial prejudice by testifying as to some counts but not others. The Defendant is therefore not entitled to sever the indictment based on his desire to testify as to some counts but not others.
ii. Potential spillover prejudice
The Defendant also argues that he is entitled to severance based on possible spillover prejudice. A defendant may be substantially prejudiced—and, therefore, entitled to severance under Rule 14(a)— where joint trial of multiple counts might “prevent the jury from making a reliable judgment about guilt or innocence.” Zafiro v. United States, 506 U.S. 534, 539, 113 S.Ct. 933, 122 L.Ed.2d 317 (1993). For example, if “evidence that the jury should not consider” as to certain counts were nonetheless admitted in a joint trial, the jury might improperly use that evidence to infer a defendant’s guilt as to counts on which the evidence is inadmissible. Id.
Whether severance is appropriate because of potential spillover prejudice in this case is a difficult question. At least three factors favor severing trial of the 2010 and 2011 incidents (Counts 1, 2, and 3) from trial of the 2014 incident (Count 4). First, the Government’s primary proof of the 2014 incident—video of the incident— has the potential to be much more vivid and memorable for the jury when contrasted with the testimonial evidence that appears to be the Government’s primary proof of the 2010 and 2011 incidents. There may be a risk that the jury will give the video undue weight or improper evidentia-ry value when considering the 2010 and 2011 incidents. Second, and relatedly, the Government acknowledges that at least some of its evidence as to the 2010 incident is missing. Particularly given the nature of the Government’s proof as to the 2014 incident, trying the 2010 Counts together with the 2014 Count might bolster potential weaknesses in the 2010 Counts. And third, the 2010 and 2011 Counts are temporarily distinct from the 2014 Count.7
But, on the other hand, several factors strongly favor trying all three excessive force incidents together. Federal courts have a “preference” for joint trial in part because of the obvious efficiency of trying multiple counts at once. Zafiro, 506 U.S. at 538, 113 S.Ct. 933. The Government states that a joint trial in this case would promote such efficiencies because, for each excessive force incident, the Government intends to “call witnesses from the Buffalo Police Department to explain the training and standards regarding use of force within the department.” Docket No. 17 at 20. The Government also states that it is “contemplating calling an expert witness regarding use of force.” Id. The efficiency of having these witnesses testify once, rather than two or three times, is undeniable. And if the Defendant chooses to introduce any witnesses on his behalf (for instance, a competing expert, or witnesses who will *252offer testimony under Rule of Evidence 404(a)(2)(A)), the efficiency of a joint trial is even more evident.
In addition to these efficiency benefits, a joint trial may also be appropriate if the Government is permitted, under Rule of Evidence 404(b), to use evidence of one excessive force incident to help prove another excessive force incident. See United States v. Nance, 168 F.Supp.3d 541, 553 (W.D.N.Y. 2016) (“[F]or purposes of Rule 14(a), a defendant cannot be unfairly prejudiced by evidence the Government offers to prove the crimes with which the defendant is charged.”) The Government states that, if the three excessive force counts were severed, it “would seek to offer evidence of the other incidents” under Rule 404(b). Docket No. 32 at 4. This issue, however, is better resolved closer to or at trial, when the Court can better evaluate the potentially-prejudicial nature of the Rule 404(b) evidence in light of its probative value. See United States v. Levy, 731 F.2d 997, 1002 (2d Cir. 1984) (“When the court has determined that the evidence is admissible under Rule 404(b), it must then balance the probative value of the evidence against its prejudicial impact to determine if the evidence is excludable under Rule 403”)
Thus, the Court cannot conclude on the current record whether the Defendant would be substantially prejudiced by joint trial. As discussed above, the answer to this question will turn heavily on facts and trial issues that, at this point, are still unsettled. The Court therefore reserves decision on whether to sever the 2010 and 2011 Counts from the 2014 Count. The Court will resolve the issue closer to trial, once the parties have better identified them anticipated evidence and witnesses, and once they have had an opportunity to fully brief (and once the Court has had an opportunity to fully consider) the Rule 404(b) issue.
CONCLUSION
For the reasons stated above, the Court adopts Magistrate Judge Schroeder’s Report and Recommendation in its entirety. See Docket No. 21. The Court therefore denies each of the Defendant’s motions to dismiss. See Docket No. 14. The Court grants the Defendant’s motion for a Kasti-gar hearing, which the Court will conduct either during trial and/or after trial. Finally, the Court reserves decision on whether to sever Counts 1, 2, and 3 from Count 4.
As noted, the question whether severance is appropriate in this case turns heavily on whether, if the excessive force Counts were severed, the Government would be permitted to introduce evidence of the other excessive incidents under Rule 404(b). The parties have, to a limited extent, addressed this issue in their prior briefing. However, given the importance of the Rule 404(b) issue to the Court’s severance decision, the Court requires further briefing. The Government shall file its brief on or before August 12, 2016. The Defendant shall file his brief on or before August 29, 2016. The Court will set a date for oral argument at a later date.
SO ORDERED.
. The facts presented below are taken from the Government’s and the Defendant’s briefs in support of, and in opposition to, the Defendant’s motions.
. The Court assumes for purposes of this decision that knowledge of those incidents-by the *242Justice Department’s Civil Rights Division is attributable to the Criminal Division and the U.S. Attorneys' Office for the Western District of New York.
. The Defendant does not argue, nor does the record suggest, that this is a case in which the Government’s alleged error is a "fundamental” one that affected "the structural protections of the grand jury” in a way that "render[s] the proceedings fundamentally unfair, allowing the presumption of prejudice." Bank of Nova Scotia, 487 U.S. at 256-57, 108 S.Ct. 2369.
. The Defendant does not challenge the join-der of, or request to sever, Counts 1 and 2 "inasmuch as they are based upon acts that allegedly occurred on the same date in connection with the same incident.” Docket No. 14-1 at 16.
. As Werner noted, this interpretation of' Rule 8(a) is consistent with United States v. Halper, 590 F.2d 422 (2d Cir. 1978), the case on which the Defendant largely bases his argument. Halper did not directly address the issue raised in this case: whether joinder was proper under Rule 8(a). Rather, Halper concerned whether two indictments, each charging distinct conduct, were properly joined under Rule 13. As Halper notes, although Rule 13 is informed by Rule 8, the two Rules have *249different standards. See Halper, 590 F.2d at 428 ("[E]ven if the offenses could have been joined in a single indictment under Rule 8(a), the trial court is not automatically free to order the indictments tried together under Rule 13.”) See also Werner, 620 F.2d at 927 (addressing the distinction between Werner and Halper ).
. In concluding that Counts 1, 3, and 4 were properly joined under Rule 8(a) because they are of a "same or similar character,” the Court need not address the fact that the other policies underlying Rule 8(a), such as efficiency, also support joinder in this case. See, e.g., United States v. Blakney, 941 F.2d 114, 116 (2d Cir. 1991) ("Joinder is proper where the same evidence may be used to prove each count.”).
. These same considerations favor trying the 2010 and 2011 incidents together. As the Defendant acknowledged at oral argument, there is very little basis for severing those Counts. See Tr. 30:15-17.