OPINION

CLAY, Circuit Judge.

Kelvin Mize, James Mize, and Jackie Mize (collectively "Defendants") were convicted, following a jury trial, of conspiracy to distribute and possession with intent to distribute oxycodone in violation of 21 U.S.C. §§ 846, 841(a)(1) and (b)(1)(C), and conspiracy to commit money laundering in violation of 18 U.S.C. §§ 1956(a)(1)(A)(i), (a)(1)(B)(i), and 1956(h). Those charges arose from the Defendants' involvement in a prescription pill conspiracy organized by Jackie Mize (the "Mize conspiracy"), the father of Kelvin and James. The Mize conspiracy operated similarly to a separate conspiracy organized by Kevin Bussell (the "Bussell conspiracy"). Because we find a prejudicial variance between the charges in the indictment and the proof offered at trial, we REVERSE and VACATE all three convictions and REMAND for a new trial.

I.

BACKGROUND

To explain how the Mize conspiracy began and operated, the government at trial presented evidence about the separate Bussell conspiracy. The Bussell conspiracy operated as follows: groups of individuals from Tennessee, many of whom were drug addicts, would travel to doctors' offices and pain clinics in Florida, visiting several doctors and pain clinicians at a time. Kevin Bussell,1 the conspiracy's ringleader, paid for their travel expenses, including hotel costs, and even provided them with drugs. These "doctor shoppers" would present false reports of pain to examining doctors in order to obtain prescription medication pills—mainly opiates and oxycodone. They would then bring those pills back to Tennessee, keep half of them, and give the other half to Bussell, who would sell his share for a profit.

The Bussell conspiracy was formed in 2009 when Bussell assembled a group of people to travel from Tennessee to Florida and obtain prescription medication pills. Bussell's first group of shoppers consisted of three people, one of whom was James Mize. But because James had to wait six months before he could refill his prescription, he did not accompany Bussell on the second trip to Florida. Instead, the trip was made by James' brother, Kelvin Mize.

Eventually, Bussell started doctor shopping—i.e., his doctor shoppers would visit multiple doctors in order to obtain multiple controlled substance prescriptions. This also allowed Bussell's doctor shoppers to circumvent Florida's six-month waiting time for filling new prescriptions. Because of the doctor shopping scheme, Bussell increased the number of shoppers he took on each trip to Florida. At its peak, the Bussell conspiracy sent forty people to Florida at a time, each of whom visited anywhere from two to five doctors per trip.

Both James and Kelvin went to Florida with Bussell as doctor shoppers. Jackie, on the other hand, was not one of Bussell's doctor shoppers. Instead, Jackie asked Bussell to take him on a trip to Florida to learn how the operation worked. Jackie accompanied Bussell on his next trip and Bussell showed him how the operation worked. But because Jackie was not one of Bussell's doctor shoppers, he paid for his own trip expenses.

After seeing Bussell and his shoppers in action, Jackie assembled his own group to travel to Florida and essentially do the same thing—i.e., obtain prescription pills for opiates and oxycodone and return to Tennessee to sell them on the black market. Jackie's group of doctor shoppers included Kelvin and his ex-wife Donna Webb, James, and Jackie's two other sons, Jonathan and Ryan.

The conspiracy organized by Jackie, the Mize conspiracy, would eventually grow to ten doctor shoppers, and did in fact operate just like the Bussell conspiracy. Upon returning from Florida, Jackie's doctor shoppers gave him half of their prescription pills. Either Jackie himself or Kelvin would then sell those pills for money in Tennessee. For example, Crystal Mason, one of Bussell's doctor shoppers, testified that she bought pills from Kelvin. On the other hand, Johnny Harvey, another one of Bussell's doctor shoppers, testified that he bought pills from Jackie. James, for his part, was mainly a drug user and doctor shopper—he did not sell very many pills.

According to the testimony at trial, Florida doctors and pharmacists started to adopt more stringent screening measures for prescription medication. One such measure was that clinics would not see patients, and pharmacists would not fill a prescription, absent proof of a valid Florida driver's license. Bussell responded by leasing residential property in Florida, which he used to establish residency in Florida for his doctor shoppers. This had the added benefit of housing his doctor shoppers when they travelled to Florida, thus eliminating hotel costs. Soon after Bussell entered into his lease, Jackie followed suit, leasing his own Florida property for purposes of establishing residency in Florida for his doctor shoppers.

In November 2010, law enforcement officers executed search warrants at both Bussell's rental houses in Florida and at Jackie's farmhouse in Harrogate, Tennessee. At trial, the items seized at Jackie's residence were admitted into evidence. Those items included prescription pill bottles for oxycodone in Jackie's name that had been filled at a pharmacy in Florida, and a folder containing several pages of the names and addresses of pain management clinics in Florida. Law enforcement officers also found a printout of Kelvin's patient drug history, which showed payment of $1,798.80 for 80 milligrams of oxycodone, and various pieces of notebook paper that contained handwritten notations about several kinds of drugs. There were handwritten notes about "TP," which was a reference to Xanax pills, and also about "30s," which were references to 30-milligram Roxycodone pills. One of those notes read: "I sold ten TPs plus three 30s. I sold one 30." (R. 115, Jury Trial Transcript, Volume 1, PageID# 468.) Another note read: "I paid dad 550 more dollars." (Id.)

Following the government's case-in-chief during the Mize conspiracy trial, Defendants moved for acquittal under Rule 29 of the Federal Rules of Criminal Procedure, arguing, among other things, that the government had not presented any physical evidence of pills or large amounts of cash. Kelvin also argued that the government improperly presented evidence about the separate Bussell conspiracy in an attempt to prove that he and James and Jackie were involved in a much larger drug-trafficking conspiracy. The district court denied Defendants' motions for acquittal. At the close of evidence, a jury convicted Defendants of both counts.

Over Defendants' objections, the district court adopted the presentence investigation report's findings with respect to the amount-of-loss and drug quantities attributable to Defendants. The court sentenced both Kelvin and Jackie to 360 months' imprisonment and sentenced James to 300 months' imprisonment. This appeal then followed. We discuss additional facts as necessary below.

II.

DISCUSSION

Defendants present three issues on appeal. First, Kelvin argues constructive amendment of the indictment and prejudicial variance in the proofs. He argues that the government's presentation of evidence about the separate Bussell conspiracy effectively amended the indictment, putting him on trial for a crime for which he was never formally charged. Second, he challenges the district court's calculation of the drug quantity attributable to him. And third, all defendants challenge the factual sufficiency of the evidence to support their convictions.

Because we reverse and vacate Defendants' convictions on the grounds that a prejudicial variance existed between the single conspiracy charged in the indictment and the government's proofs at trial, we need not reach Defendants' other contentions on appeal.

A. Reaching the Unpreserved Issue

We note initially that Kelvin is the only defendant to raise a constructive amendment and variance argument on appeal. And as discussed above, he also objected at trial to the government's presentation of evidence regarding the Bussell conspiracy, thereby preserving his claim for appellate review. While Jackie does not raise a constructive amendment and variance argument on appeal, he did, however, object at trial to the substantial amount of evidence about the Bussell conspiracy. James, on the other hand, does not raise this argument on appeal, nor did he raise this challenge at trial.

"The general rule of appellate procedure is that issues not presented in an appellant's initial merits brief are waived." Citizens Coal Council v. U.S. E.P.A., 447 F.3d 879, 905 (6th Cir.2006) (en banc) (citation omitted). But this rule is not jurisdictional, and we may choose to entertain arguments not raised by the parties when the failure to do so would constitute a miscarriage of justice. See Mayhew v. Allsup, 166 F.3d 821, 823-24 (6th Cir.1999) (holding that the court would consider the application of a statute helpful to the defendant despite his failure to address it either at trial or on appeal when the failure to do so would constitute a miscarriage of justice).

We recognize that the rule in Mayhew applies only to "exceptional" cases. See id. This case, however, meets that exception because, if successful, Defendants' appeal on this issue would result in reversal of their convictions. Moreover, because the issue has been briefed by both sides and discussed at oral argument, the government is not prejudiced by Jackie's and James' neglect in not raising the issue in their briefs.2 Finally, because Defendants were tried together for a single conspiracy, it would be a miscarriage of justice to affirm Jackie's and James' convictions while reversing Kelvin's conviction based on a constitutional error that tainted Defendants' joint jury trial. We will therefore exercise our discretion to reach the merits as to whether a constructive amendment or variance occurred as to all defendants, because we believe that a failure to do so would constitute a miscarriage of justice. See id.

B. The Indictment Versus the Proof at Trial

a. Standard of Review

Generally, we evaluate claims of constructive amendments to or variances from an indictment de novo. United States v. Prince, 214 F.3d 740, 756 (6th Cir.2000). However, when a defendant fails to preserve an argument for appeal, we review only for plain error. To obtain relief under that standard, a defendant must establish "(1) error, (2) that is plain, and (3) that affects substantial rights." Johnson v. United States, 520 U.S. 461, 466-67, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997) (internal citations and quotations omitted). If a defendant can show all three conditions, we will "exercise [our] discretion to notice a forfeited error, but only if (4) the error seriously affects the fairness, integrity or public reputation of judicial proceedings." Id. at 467, 117 S.Ct. 1544 (alteration in original). The defendant bears the burden of demonstrating that a constructive amendment or variance has occurred. United States v. Chilingirian, 280 F.3d 704, 712 (6th Cir.2002).

Because Kelvin preserved his claim of constructive amendment or variance at trial (and also on appeal), we review his claim de novo. We also apply this standard to Jackie because he objected at trial to the introduction of extensive evidence about the Bussell conspiracy, thus preserving his claim on appeal. But since James did not preserve this issue, we apply plain error review as to him. See United States v. Kuehne, 547 F.3d 667, 682 (6th Cir.2008) ("[W]here no specific objection is raised regarding a constructive amendment or a variance before the district court, we are limited to `plain error' review on appeal.").

b. Constructive Amendment of the Indictment/Variance

Defendants argue that the evidence introduced by the government at trial so diverged from the charges set forth in the indictment as to violate their rights under the Fifth and Sixth Amendments. "An indictment may be the subject of an actual amendment, a constructive amendment, or a variance." United States v. Budd, 496 F.3d 517, 521 (6th Cir.2007). Thus, constructive amendments and variances are two types of modifications to indictments that we have recognized. United States v. Hynes, 467 F.3d 951, 961 (6th Cir.2006). However, these two concepts—constructive amendments and variances—differ "with respect to the burden placed upon the defendant and the remedy mandated upon a showing that a constructive amendment or variance has occurred." Kuehne, 547 F.3d at 683.

A constructive amendment "results when the terms of an indictment are in effect altered by the presentation of evidence and jury instructions which so modify essential elements of the offense charged such that there is a substantial likelihood that the defendant may have been convicted of an offense other than the one charged in the indictment." United States v. Martinez, 430 F.3d 317, 338 (6th Cir.2005). Constructive amendments are "per se prejudicial because they infringe upon the Fifth Amendment's grand jury guarantee." Hynes, 467 F.3d at 962 (internal citations and quotations omitted). "Because of the constitutional injury that results from a constructive amendment, when proven, a defendant is entitled to a reversal of his conviction." Kuehne, 547 F.3d at 683 (citation omitted).

In contrast, a variance is "not per se prejudicial." Budd, 496 F.3d at 521. Instead, reversal is only warranted when a defendant proves that "(1) a variance occurred and (2) that the variance affected a substantial right of the defendant." Kuehne, 547 F.3d at 683 (citing Prince, 214 F.3d at 757). Generally speaking, a variance "occurs when the charging terms [of the indictment] are unchanged, but the evidence at trial proves facts materially different from those alleged in the indictment." Id. (alteration in original). A defendant's substantial rights are affected "only when the defendant shows prejudice to his ability to defend himself at trial, to the general fairness of the trial, or to the indictment's sufficiency to bar subsequent prosecutions." Id. (citations and quotations omitted).

c. Analysis

We are not convinced that a constructive amendment, as opposed to a prejudicial variance, occurred in this case. Although our precedent recognizes that the difference between the two is "shadowy," United States v. Barrow, 118 F.3d 482, 488 (6th Cir.1997), we have held that a constructive amendment requires a showing "that the important functions of an indictment were undermined by both the evidence presented and the jury instructions." Hynes, 467 F.3d at 962 (emphasis in original).

In this case, all of the harm was done by the introduction of extraneous and highly prejudicial evidence about the separate Bussell conspiracy. The jury instructions, on the other hand, do not support Defendants' claim that the indictment was constructively amended. In relevant part, the jury instructions provided as follows:

(R. 117, Jury Trial Transcript, Volume 3, PageID# 1110-11.)

While not necessarily a model of simplicity and clarity, these jury instructions do seem calculated to mitigate, to the extent possible, any potential for prejudice from the evidence about the Bussell conspiracy. The instructions specifically explained that the only way the jury could convict Defendants was to find beyond a reasonable doubt that they were members "of the conspiracy charged in the indictment"—i.e., the Mize conspiracy. Moreover, we have in the past approved of similar instructions in this context. See, e.g., United States v. Gioiosa, 924 F.2d 1059, 1991 WL 15149 at n. 4 (6th Cir.1991) (unpublished table opinion); United States v. Battista, 646 F.2d 237, 243 (6th Cir.1981). Therefore, under these circumstances, we find that there was no constructive amendment here.

We do, however, find that there was a prejudicial variance between the charges in the indictment and the evidence produced at trial. To determine whether reversal is required, we must engage in a two-step inquiry: was there a variance, and if so, was it prejudicial. Kuehne, 547 F.3d at 683 (citing Prince, 214 F.3d at 757). To determine whether a variance has occurred, we look to whether the evidence can "reasonably be construed only as supporting a finding of multiple conspiracies" rather than the single conspiracy alleged in the indictment. United States v. Warner, 690 F.2d 545, 548 (6th Cir.1982) (citing Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946)). Moreover, "defendants can establish a variance by referring exclusively to the evidence presented at trial." Hynes, 467 F.3d at 962.

The evidence in this case undoubtedly established two conspiracies, not just the one charged in the indictment. The government presented an extraordinary volume of evidence about the Bussell conspiracy. Even in its opening statement, the government began by giving the jury extensive information about the Bussell conspiracy. It then introduced testimony about the volume of drug transactions effected by the Bussell conspiracy. The government also presented a history of the investigation of the Bussell conspiracy which included wiretaps, surveillance, controlled buys, drug deals, and even surveillance photographs of members of the Bussell conspiracy. Even Bussell himself—along with several of his convicted collaborators—testified about the operation of his conspiracy. Bussell even went so far as to credit the government for his rehabilitation from being a drug user.

The government then argued to the jury that the Mize conspiracy operated in the same manner to commit the same crimes. By the close of its case-in-chief, the government had presented eleven witnesses—eight of whom discussed the Bussell conspiracy at some length. Indeed, during its direct examination of Drug Enforcement Administration ("DEA") Agent Bethel Poston, the government even acknowledged that the majority of Poston's testimony was about Bussell:

(R. 115 at 498.)

We think that if the testimony introduced by the government established anything, it was that Kevin Bussell operated a large scale drug trafficking organization responsible for the importation and distribution of hundreds of thousands of dollars' worth of pills from Florida to Tennessee. It is clear from our reading of the record that the government introduced significantly more evidence than it should have about the separate Bussell conspiracy. We therefore find that the government's evidence rises to the level of a variance. We turn then to whether the variance was prejudicial.

To demonstrate substantial prejudice, a defendant must show that the variance prejudiced either his ability to defend himself or the overall fairness of his trial. United States v. Manning, 142 F.3d 336, 339 (6th Cir.1998). Prejudice exists "where the defendant is unable to present his case and is taken by surprise by the evidence offered at trial"; "where the defendant is convicted for substantive offenses committed by another"; or "where spillover [occurs] because of a large number of improperly joined defendants." United States v. Swafford, 512 F.3d 833, 842-43 (6th Cir.2008) (internal citations and quotations omitted).

At least one of these conditions has occurred here. The primary risk that the variance doctrine is designed to alleviate is guilt transference—"that the appellant was convicted based on evidence of a conspiracy in which the appellant did not participate." United States v. Hughes, 505 F.3d 578, 587 (6th Cir.2007) (citing United States v. Blackwell, 459 F.3d 739, 762 (6th Cir.2006)). "This risk increases in direct proportion to the number of defendants, and the number of conspiracies demonstrated at trial." United States v. Caver, 470 F.3d 220, 237 (6th Cir.2006) (citing Kotteakos, 328 U.S. at 766, 66 S.Ct. 1239).

Even if a defendant can show that a variance resulted in guilt transference, "typically any danger of prejudice can be cured with a cautionary instruction to the jury that if it finds multiple conspiracies, it cannot use evidence relating to one conspiracy in determining another conspiracy." Hughes, 505 F.3d at 587 (citing Blackwell, 459 F.3d at 762). However, as we noted in Blackwell, "the more evidence presented at trial that is unrelated to the defendant's conduct, or a conspiracy in which the defendant took part, the less likely instructions are to cure the danger of guilt transference." 459 F.3d at 762.

"Whether or not a variance is prejudicial is a judgment that must be made on the facts of each case." United States v. Mills, 366 F.2d 512, 514 (6th Cir.1966) (citations and internal quotation marks omitted). This is an incredibly fact-intensive analysis requiring the consideration of what occurred in each case. And in this case, we do not see any other way but to find that Defendants were prejudiced by the material variance. They were forced to defend against a conspiracy, i.e., the Bussell conspiracy, that was totally separate from the conspiracy alleged in the indictment, i.e., the Mize conspiracy. The evidence from a different scheme was used to portray Defendants in a grossly prejudicial light before the jury. This error enabled the government to bombard the jury with evidence of the Bussell conspiracy and its cast of characters and recorded conversations and photographs, when, in fairness, those things had little to do with the charged Mize conspiracy. We do not see any reason why the government should not have been required to prove its charges against Defendants without all of this prejudicial and irrelevant evidence about a totally different conspiracy involving Bussell.

We think that we understand why the government did what it did—the theme of its case was that Defendants were inspired by the Bussell conspiracy to create their own conspiracy operating in a substantially similar manner. But this theme could have been told differently. The government could have easily explained to the jury that this case involves a conspiracy which was formed by Jackie Mize and that the idea for the conspiracy originated when Jackie learned of the Bussell conspiracy. That was really all that needed to be said about the Bussell conspiracy. Instead, the extensive proof presented by the government on the Bussell conspiracy likely distracted the jury from the relevant issues—all to Defendants' prejudice.

What this case really comes down to is balance—i.e., the balance between the government's need to introduce evidence about the Bussell conspiracy and its need to introduce evidence about the Mize conspiracy. And in trying to find that balance, the government leaned too heavily on establishing the existence and intricacies of the Bussell conspiracy, rather than focusing on the conspiracy for which Defendants stood trial. The government's extraordinary amount of evidence about the Bussell conspiracy enabled the jury to transfer the guilt of that conspiracy to the charged Mize conspiracy.

Moreover, we reject the government's argument that any error was harmless. "[T]his is not and cannot be the test." Kotteakos, 328 U.S. at 767, 66 S.Ct. 1239. Prejudice in this context means whether Defendants were found guilty of a different conspiracy from that charged in the indictment, not whether the evidence was sufficient to justify the verdict. And in any event, harmless error should not be what stands in the way of a seemingly unconstitutional conviction. We likewise reject the government's argument that any danger of prejudice was minimized by the district court's instructions to the jury. While that may be the general rule, it does not apply here. Because of the sheer volume of evidence introduced to establish the Bussell conspiracy, it is substantially less likely that the court's instructions cured the danger of creating unfair prejudice. See Blackwell, 459 F.3d at 762.

For all of these reasons, we find a prejudicial variance between the charges in the indictment and the proof offered at trial. Moreover, we find that the error was plain, that it affected the substantial rights of all three defendants, and that it seriously affected the fairness of the trial.

We therefore REVERSE and VACATE all three convictions and REMAND for a new trial.

KEITH, Circuit Judge, dissenting.

I respectfully dissent. The majority opinion is flawed in several respects, and I will highlight the most problematic areas here.

A. Reaching the Unpreserved Issue—Misconstrued Record & Case Law

As a preliminary matter, the majority misconstrues the record and the case law in their pursuit to reach the unpreserved issue with respect to James and Jackie. The majority's assertion that the "issue has been briefed by both sides and discussed at oral argument" is misleading. Neither Jackie's counsel nor James's counsel even attended oral argument, and neither raised the issue of a constructive amendment or fatal variance in their briefs. Therefore, the Government never had an opportunity to respond to any argument that a fatal variance occurred as to James's or Jackie's involvement in the offense. While asserting that the inquiry is "fact-intensive," the majority simultaneously applies a cookie-cutter approach to all three defendants, when in fact the circumstances for each defendant were different. For example, it is undisputed that Jackie was never a member of the Bussell conspiracy.

The majority relies on the case of Mayhew v. Allsup, 166 F.3d 821, 823-24 (6th Cir.1999), to justify addressing an issue that neither James nor Jackie raised. The majority asserts that Mayhew stands for the proposition that a court can "consider the application of a statute helpful to the defendant despite his failure to address it either at trial or on appeal when the failure to do so would constitute a miscarriage of justice." A close reading of Mayhew reveals that the majority misconstrued this case in much the same way it misconstrued the record. In Mayhew, the plaintiff, who was the appellant, failed to raise an argument before the trial court regarding a statute that was not enacted until after the trial court issued its judgment. See id. at 823. But the appellant did raise the issue before this court in its appellate brief. See id. The defendant, a corporate entity, was the appellee in that civil case. See id. It failed to directly respond to the issue raised by the appellant, instead arguing that the issue was not preserved. See id. This court determined that the issue could still be addressed in order to avoid a miscarriage of justice. See id. at 823-24. Mayhew is inapposite for many reasons, but mainly because the appellant in Mayhew, raised the issue on appeal, unlike Jackie or James here. See id.1 The majority misconstrues the record and Mayhew in an attempt to excuse its conduct of acting as defense counsel rather than impartial judges.2 Fundamentally, I believe we should refrain from overstepping our bounds under the guise of "justice."

Additionally, the majority purports to apply a different standard of review to the fatal variance claim it constructed for James, noting that "plain error" review was required. However, the majority, in perfunctory fashion, devotes one sentence to the "plain error" inquiry that amounts to nothing more than a bare recitation of the "plain error" elements without any analysis or citation to legal authority.

Because neither James nor Jackie raised the issue on appeal, the remainder of my dissent will address the fatal variance claim as raised by Kelvin Mize.

B. No Guilt Transference Occurred by Definition of the Term

I agree that a variance occurred with respect to Kelvin because multiple conspiracies were discussed at trial. However, as the majority points out, not every variance is a fatal one (i.e., not every variance requires reversal). See United States v. Budd, 496 F.3d 517, 522 (6th Cir.2007) (noting that a variance is not prejudicial per se, and thus, a variance will not always require reversal). A liberal reading of the majority opinion suggests that the majority believes guilt transference occurred with respect to Kelvin, rendering the variance fatal. I respectfully disagree. By definition, guilt transference occurs where there is a risk that the defendant was convicted "based on evidence of a conspiracy in which the [defendant] did not participate." United States v. Hughes, 505 F.3d 578, 587 (6th Cir.2007) (emphasis added). In other words, it occurs where guilt is transferred from one defendant to another defendant. See id; see also United States v. Caver, 470 F.3d 220, 237 (6th Cir.2006) ("The primary risk is the transference of guilt from defendants involved in one conspiracy to defendants in another conspiracy.") (emphasis added). In essence, when guilt transference occurs, a defendant is convicted of the "substantive acts of another" person. United States v. Bakri, 505 Fed.Appx. 462, 468-69 (6th Cir.2012).

Accordingly, there is no guilt transference where the defendant was a participant in all of the conspiracies presented at trial. See, e.g., Hughes, 505 F.3d at 587 (6th Cir.2007) (noting that to demonstrate a prejudicial variance, "[a] reviewing court looks to whether there is a danger that the appellant was convicted based on evidence of a conspiracy in which the appellant did not participate (guilt transference)") (emphasis added); United States v. Blackwell, 459 F.3d 739, 762 (6th Cir.2006) (same); see also United States v. Martin, 516 Fed. Appx. 433, 443-44 (6th Cir.2013) cert. denied sub nom. Brooks v. United States, ___ U.S. ___, 134 S.Ct. 301, 187 L.Ed.2d 215 (2013) (holding that no prejudicial variance occurred because even "assuming the existence of two separate conspiracies, the evidence overwhelmingly implicated [the defendant] in both"); Bakri, 505 Fed. Appx. at 468-69 ("Nor was there any possibility of [d]efendant being convicted for the substantive acts of another. Regardless of how one describes the conspiracy or the conspiracies, [d]efendant was an active participant."); United States v. Goff, 400 Fed.Appx. 1, 13 (6th Cir.2010) (holding that even though a variance occurred where evidence of two separate conspiracies was introduced, the variance did not warrant reversal because the defendant "was an integral participant in these other conspiracies"); United States v. Hettinger, 242 Fed.Appx. 287, 294 (6th Cir.2007) (holding that even assuming that multiple conspiracies existed, the defendant "was a member of each conspiracy and accordingly could not establish any `danger . . . that [he] was convicted based on evidence of a conspiracy in which [he] did not participate'"); United States v. Mitchum, 208 F.3d 216, 2000 WL 222578, *3 (6th Cir. 2000) (unpublished table decision) (holding that no fatal variance occurred because "no matter which conspiracy the jury believed existed, [the defendant] was involved in each"). Here, "[r]egardless of how one describes the conspiracy or the conspiracies, [Kelvin] was an active participant" in both of them. See Bakri, 505 Fed.Appx. at 468-69. Therefore, guilt transference did not occur. See id.

Even though a fatal variance did not occur, surely the government cannot introduce mounds of "extraneous and highly prejudicial" evidence, a term used by the majority, without some recourse for the defendant. Intuitively, one would think that something is amiss with the presentation of the evidence at trial. There is. Where the evidence does not rise to the level of a fatal variance, the recourse is an objection pursuant to the Federal Rules of Evidence. These rules govern the admissibility of prejudicial evidence. Fed. R.Evid. 403 (noting that in order for evidence to be excluded under this rule, the probative value of the evidence must be "substantially" outweighed by the danger of "unfair prejudice") (emphasis added). And the rules also govern the admission of evidence of other crimes or wrongs. Fed. R.Evid. 404(b). As an initial matter, it is not entirely clear what "highly" prejudicial means as the majority uses the term here. The majority does not define or quantify this term. If the majority is of the opinion that "highly" prejudicial actually means that the probative value of the evidence is substantially outweighed by the danger of unfair prejudice, then the majority's analysis demonstrates that the majority is actually applying a Rule 403 analysis to a fatal variance case.

The majority wrongly assumes that because "prejudice" to a defendant's substantial rights is a component of the fatal variance analysis, see Caver, 470 F.3d at 235-36, the introduction of "highly prejudicial" evidence must be sufficient for a fatal variance. But the majority attempts to fit a square peg into a round hole. In every case where there is a variance, by definition, there was evidence of a separate uncharged conspiracy, United States v. Warman, 578 F.3d 320, 341 (6th Cir.2009), which is undoubtedly prejudicial to some extent.

In essence, the majority attempts to fit an argument that sounds plainly in the admissibility of evidence (which is subject to abuse of discretion review)3 into a fatal variance analysis (subject to de novo review).4 As the majority opinion's loose language shows, the pieces do not fit. Cf. United States v. English, 785 F.3d 1052, 1059 (6th Cir.2015) (Clay, J., concurring) (concluding that the trial court's allowance of evidence that the defendant had been involved in two prior Medicare fraud schemes during a trial for Medicare fraud was error under 404(b), but nonetheless harmless error, even where the defendant raised the issue of a fatal variance).

C. The Majority's Impracticable Approach to the Volume of the Evidence Test

The majority opinion suggests that the "volume" of the evidence of the Bussell conspiracy makes this case unique and that the case "comes down to . . . balance." While balance is important, the majority again misses the mark in its analysis.

In United States v. Blackwell, 459 F.3d 739, 762 (6th Cir.2006), this court noted that where the sheer volume of evidence of a separate conspiracy is concerning, we can examine the following three factors to determine whether a variance was prejudicial: (1) the number of conspiracies the evidence establishes, (2) the number of non-conspiratorial co-defendants tried with defendant, and (3) the size of the conspiracy alleged in the indictment. Id. at 762. Indeed, in United States v. Hughes, 505 F.3d 578, 590-91 (6th Cir.2007), we held (after applying these factors) that a variance was not prejudicial where there were three separate conspiracies and six codefendants. In United States v. Osborne, 545 F.3d 440, 444 (6th Cir.2008), we held that even assuming a variance occurred, it did not require reversal where there were two conspiracies and only three defendants. In the present case, we similarly have two conspiracies and three defendants.

The majority seems to suggest that we need to abandon the three-factor test articulated in Blackwell, and instead employ a "this-feels-like-too-much-evidence" approach. This approach is problematic for at least two reasons. First, it advances a test that is non-quantifiable; in other words, it will be impossible for trial judges to know whether 50% or 60% or 70% of the evidence admitted is too much evidence of the uncharged conspiracy. Second, it is impracticable in reality. Trial judges would presumably be forced to keep tally marks of every piece of evidence of the uncharged conspiracy during the trial, and then presumably declare a mistrial midway through when the evidence crosses this arbitrary threshold of "too much"—even without defense counsel having raised the issue." I believe the problems associated with this approach are the very reasons this court usually employs a very quantifiable inquiry, such as the one articulated in Blackwell—the one the majority surreptitiously abandons here.

D. Juror Confusion

The majority also asserts that reversal is warranted because the jury was "distracted" by the evidence of the Bussell conspiracy. Perhaps the majority meant to rely on juror "confusion," which is applicable to the analysis for determining whether a variance was prejudicial. See United States v. Osborne, 545 F.3d 440, 444 (6th Cir.2008) (holding that even assuming a variance occurred, the evidence related to a second conspiracy was "clearly demarcated and not likely to confuse the jury."). At oral argument, Kelvin's counsel conceded that distinctions were drawn between the Bussell conspiracy and the Mize conspiracy at trial. This made it even less likely that the jury was confused. See id. Also, as we concluded in Caver, a trial of short duration also makes it less likely that the jury was confused. United States v. Caver, 470 F.3d 220, 237 (6th Cir.2006) (concluding that even assuming that the evidence at trial demonstrated only multiple conspiracies, there was no prejudice requiring reversal because, inter alia, the trial only lasted a week). In the present case, the trial only lasted four days, further minimizing any juror confusion.

E. Conclusion

In sum, the majority commits many errors: it confuses the admission of prejudicial evidence with the occurrence of a fatal variance; it ignores the definition of guilt transference (tellingly, the majority quoted the definition but never applied it to either of the defendants); it proposes a "this-feels-like-too-much" evidence test that is flawed in theory and impracticable in reality; and it robs the government of any meaningful opportunity to respond to the issues with respect to Jackie and James. In committing these errors, the majority reaches the wrong conclusion.

At first blush, it may appear that the majority reaches a "noble" outcome because the convictions of three defendants are reversed. However, courts should be ever mindful that true justice requires consistent application of the law for everyone. Undoubtedly, the "this-feels-like-too-much" evidence approach espoused by the majority will not be applied consistently in trial courts or even on appeal. Creating these arbitrary tests both invites and justifies discrimination among defendants, because what "feels like too much evidence" for one person will not "feel like too much evidence" for another. The majority opinion will perpetuate confusion and disorder in the courts, and thus undermine the public's confidence in the judiciary. As a court of appellate review, we strive to provide guidance that the district courts can consistently follow in a fair and meaningful way. But with today's opinion, the majority has abdicated that responsibility altogether.