Jerry E. Smith, Circuit Judge

Petition for Writ of Mandamus to the United States District Court for the Western District of Texas, USDC No. 1:22-CV-909, David A. Ezra, U.S. District Judge, Michael J. Edney, Hunton Andrews Kurth, L.L.P., Washington, DC, for Petitioners. Martin B. White, U.S. Commodity Futures Trading Commission, Office of General Counsel, Washington, DC, Robert A. Schwartz, Anne Whitford Stukes, Deputy General Counsels, U.S. Commodity Futures Trading Commission, Office of General Counsel, Washington, DC, for Respondent. Before Smith, Southwick, and Wilson, Circuit Judges. Jerry E. Smith, Circuit Judge:

Seeking a writ of mandamus, petitioners challenge the district court's decision to transfer their suit against the Commodities Futures Trading Commission ("CFTC") to the U.S. District Court for the District of Columbia ("D.D.C."). They contend the district court plainly abused its discretion in its balancing of the public and private interest factors governing transfer.

Having reviewed the parties' submissions, as well as the district court's thoughtful and courteous supplemental response, we agree with petitioners. In essence, the district court transferred petitioners' case to D.D.C. because of court congestion. That is a clear abuse of discretion, for which petitioners have no other remedy, with sweeping implications for this circuit's § 1404(a) transfer standards. Accordingly, we grant the petition and direct the district court to request that the case be returned to the court à quo.

I.

PredictIt is an online platform run by the Victoria University of Wellington in New Zealand. Conceived as a data-gathering tool for academic researchers, the platform allows individuals to trade on the predicted outcomes of political events. Users of PredictIt make small investments based on predicting political events, such as future elections or the passage of federal legislation. Data gathered from those transactions is shared with academic researchers.

PredictIt began operating in 2014, after the CFTC Division of Market Oversight ("DMO") issued PredictIt's operator, Victoria University, a "no-action" letter. That letter stated that DMO would "not recommend that the Commission take any enforcement action," subject to Victoria University's representations to abide by certain terms and limitations relating to the operation of PredictIt.

In August 2022, DMO rescinded the 2014 no-action letter, stating that Victoria University had failed to operate PredictIt in compliance with the terms of the 2014 no-action letter. DMO therefore directed that "remaining listed contracts and positions comprising all associated open interest in such market should be closed out and/or liquidated no later than 11:59 p.m. eastern on February 15, 2023."

The mandamus petitioners (sometimes referred to as the "plaintiffs") are various third-party users of the PredictIt platform. They consist of market operators, traders, and academics who allege injury from the CFTC's rescission of the 2014 no-action letter. Five live and work in Austin, Texas: (1) Kevin Clarke, the lead plaintiff; (2) Michael Beeler; (3) Mark Borghi; (4) Josiah Neeley; and (5) Wes Shepherd.

II.

Plaintiffs sued the CFTC, alleging the agency acted arbitrarily and capriciously, in violation of the Administrative Procedure Act ("APA"), when it withdrew the 2014 "no-action" letter. They also claimed the revocation constituted a withdrawal of a license without the necessary procedural steps.

In light of the then-looming February 2023 market shutdown, petitioners moved for a preliminary injunction. The district court, by failing to rule on the preliminary injunction for three months, effectively denied it. A merits panel of the Fifth Circuit ("2023 merits panel") granted an injunction pending appeal in January 2023 and heard arguments a month later. In March 2023, the CFTC withdrew its August 2022 rescission of the no-action letter. It also issued a new letter that "determined as a preliminary matter" that the 2014 no-action letter was void and should be withdrawn. CFTC then moved to dismiss the appeal pending before the 2023 merits panel. That motion was denied. In July 2023, the Fifth Circuit reversed the district court's effective denial and granted petitioners' motion for a preliminary injunction.

On remand from the Fifth Circuit, the district court entered the preliminary injunction, and the parties submitted a scheduling order. The CFTC then moved to transfer the case to D.D.C. Without a hearing, the district court granted that motion. The order was signed on January 16, 2024, and the case was transferred that same day. Petitioners moved to stay the transfer three days later, to which the district court initially ordered the CFTC to respond. Before the CFTC responded, however, the court denied petitioners' motion on the ground that it lacked jurisdiction over an already-transferred case. Later that day, petitioners moved for the district court to request D.D.C. to return the case. That motion was denied.

Petitioners now seek mandamus relief.

III.

Transfer under 28 U.S.C. § 1404(a) is properly granted only if the moving party "clearly establishes good cause" by "clearly demonstrating that a transfer is for the convenience of parties and witnesses, in the interest of justice." Def. Distributed v. Bruck, 30 F.4th 414, 433 (quoting In re Volkswagen of Am., Inc., 545 F.3d 304, 315 (5th Cir. 2008) (en banc) ("Volkswagen II")) (cleaned up). It is the movant's burden—and the movant's alone—to "adduce evidence and arguments that clearly establish good cause for transfer based on convenience and justice." Id.

At minimum, showing "good cause" requires the movant "clearly [to] demonstrate" that its chosen venue is "clearly more convenient." Id. That standard is not met if the movant merely shows that the transferee venue "is more likely than not to be more convenient." Id. Likewise, "the fact that litigating would be more convenient for the defendant elsewhere is not enough to justify transfer." Id.

Accordingly, to establish "good cause," a movant must show (1) that the marginal gain in convenience will be significant, and (2) that its evidence makes it plainly obvious—i.e., clearly demonstrated—that those marginal gains will actually materialize in the transferee venue.

The district court—when assessing whether the movant clearly carried its burden—must consider the following public-interest and private-interest factors.

The private-interest factors are "(1) the relative ease of access to sources of proof; (2) the availability of compulsory process to secure the attendance of witnesses; (3) the cost of attendance for willing witnesses; and (4) all other practical problems that make trial of a case easy, expeditious and inexpensive." Volkswagen II, 545 F.3d at 315 (cleaned up). The public-interest factors are "(5) the administrative difficulties flowing from court congestion; (6) the local interest in having localized interests decided at home; (7) the familiarity of the forum with the law that will govern the case; and (8) the avoidance of unnecessary problems of conflict of laws [or in] the application of foreign law."

"No factor is of dispositive weight, and we have cautioned against a 'raw counting of the factors' that 'weigh[s] each the same.' " Id. (quoting Radmax, 720 F.3d at 290 n.8). Moreover, "[w]here there is no demonstration by the movant, let alone a clear one, the [district] court cannot weigh a factor against the non-movant and in favor of transfer." Def. Distributed, 30 F.4th at 434.

IV.

The district court granted CFTC's motion to transfer primarily on the grounds of two public interest factors—(A) court congestion and (B) local interests. The court stated that the remaining public-interest factors—familiarity with applicable law and avoidance of unnecessary problems of conflict of laws—were neutral. Lastly, the court found that the four private-interest factors were all neutral because the "case will be decided on the administrative record, and likely by cross-motions for summary judgment." A. Court Congestion

The district court found that court congestion weighed in favor of transfer. First, it noted that it had earlier failed to rule timely on plaintiffs' motion for preliminary injunction, thereby necessitating an interlocutory appeal to this court. Second, the court observed that it took approximately four months for it to rule on CFTC's present motion to transfer. It therefore reasoned that D.D.C. would be able to resolve this case in a more expeditious manner.

Petitioners disagree with the finding on court congestion. They claim that D.D.C. is more congested, citing a statistic showing that "cases in the Western District of Texas get to trial in approximately 26 months on average, compared to approximately 53 months on average in the District of Columbia."

In response, CFTC avers that time to trial is irrelevant for determining court congestion because APA cases are "highly unlikely to go to trial." CFTC contends that "other metrics favoring transfer are preferable in this case"—presumably alluding to the statistic it cited in district court that showed that "there are 801 weighted cases pending in the Western District versus 276 in the D.C. District Court."

Neither statistic is particularly relevant. In TikTok, we rejected petitioner's statistics on "time to trial," explaining that contentions relying on such statistics were "foreclosed by Planned Parenthood's recognition that, as a general matter, 'the district court is better placed' to evaluate its 'docket efficiency.' " TikTok, 85 F.4th at 363 n.12 (quoting In re Planned Parenthood Fed'n of Am., Inc., 52 F.4th 625, 631 (5th Cir. 2022)). Accordingly, petitioners' and CFTC's statistics carry little weight.

After evaluating its own docket efficiency, the district court reasoned that D.D.C. would be able to resolve this case in a more expeditious manner. It was not clearly an abuse of discretion for the court to find that congestion favored transfer. B. Local Interests

The district court stated in its order that the lack of local interest in the Austin Division of the Western District of Texas favored transfer. It so found for two reasons: First, it explained that "the two entity plaintiffs—Aristotle and PredictIt—are both based in D.C." Second, the court speculated that those two entities "possibly would suffer the greatest economic harm in this case"—"more so than the individual plaintiffs who reside in Austin."

Petitioners raise two challenges to the district court's finding that local interests favored transfer:

First, they fault the district court for " 'disregard[ing]' a plaintiff's interests—just because they are similar to those of other victims of government action in many or all other districts." They make much of the district court's statement that "local interests that 'could apply virtually to any judicial district or division in the United States' are disregarded in favor of particularized local interests" (quoting Volkswagen II, 545 F.3d at 318).

Responding to petitioners' accusation, CFTC avers that the district court "did not simply ignore the allegations of injury to traders and data users located in the Western District of Texas." In CFTC's view, the court merely "reasoned that any such injury did not single out the impact of the alleged wrongful regulatory action on the Western District of Texas from its impact on traders and data users in all of the other districts where PredictIt provides services and data."

CFTC's understanding of the district court's reasoning is more accurate. At bottom, the transfer factors are relative. For each factor, the defendant's proffered venue is measured against the plaintiff's chosen venue. A completely diffuse interest—that is, an interest that "could apply to virtually any judicial district or division in the United States"—cannot affect the local-interest determination. That's because the interest is the same in both the transferor and transferee venue. These completely diffuse local interests net out to zero—neither favoring nor disfavoring transfer. Accordingly, it was not error for the district court to say that local interests applicable to all judicial districts can be disregarded.

Second, petitioners contend the district court clearly abused its discretion by reasoning that the two D.C.-based entity plaintiffs "were more important and would suffer more monetary damages than others." They point to the district court's order, which explicitly compares the projected magnitude of the entity plaintiffs' injuries to those of the individual plaintiffs: "Aristotle and PredictIt . . . possibly would suffer the greatest economic harm in this case more so than the individual plaintiffs who reside in Austin."

CFTC defends the district court's decision by asserting that the two entity plaintiffs, as operators of the exchange, "are most directly affected by the alleged wrongful regulatory action." Citing Defense Distributed v. Bruck, 30 F.4th 414, 435-36 (5th Cir. 2022), CFTC notes that a key consideration in the local interest determination is "the location of the company . . . directly regulated by the government." Additionally, CFTC asserts that "lesser weight should be given to interests common to many or all judicial districts."

The district court's and CFTC's reasoning is error. To identify localized interests properly, "[w]e look not to 'the parties' significant connections to each forum . . . but rather the significant connections between a particular venue and the events that gave rise to a suit.' " TikTok, 85 F.4th at 364 (quoting Def. Distributed, 30 F.4th at 435). That point bears repeating: We focus on the events—not the parties.

We do not consider the parties' connections to the venue because the local interest analysis is a public interest factor. Accordingly, the local-interest inquiry is concerned with the interest of non-party citizens in adjudicating the case. Considerations such as "the location of the injury, witnesses, and the [p]laintiff's residence," Def. Distributed, 30 F.4th at 435, are useful proxies for determining what local interests exist in each venue. But proxies, no matter how useful in certain cases, can never subsume the ultimate inquiry.

The district court erred because its analysis focuses exclusively on the harms alleged by each of the petitioners. Its reasoning fails to consider any of the "events that gave rise to the suit." TikTok, 85 F.4th at 364 (cleaned up). Indeed, its reasoning considers only the relationship between the venue and the party.

CFTC's first assertion is error, as it mischaracterizes the holding in Defense Distributed. That case did not hold that local interests favored a Texas judicial venue merely because the regulation directly targeted a Texas-based company. Rather, Defense Distributed reasoned that, relative to their neighbors in Texas, New Jersey's citizens had little interest in a dispute that would not affect them:

• Even if Defense Distributed won on the merits, "the [New Jersey Attorney General] would not necessarily be prevented from enforcing the law in New Jersey." Def. Distributed, 30 F.4th at 435.

• Moreover, the events giving rise to that suit occurred in Texas. Defense Distributed "distributed digital firearms information via mail and at a brick-and-mortar public library in Austin." See id. at 434 n.25. Its "research, design, development, manufacturing, and publishing, occurred in and around Austin." Id. It used Texas-based servers to host files visitors could download from its website. Id.

• Further still, Texas courts, as non-litigants, had a "significant interest in assessing the constitutionality and extraterritorial impact of New Jersey's criminal law, especially a law that criminalizes speech." Id. at 435-36.

So, it is hard to blame New Jersey citizens for their relative lack of interest in a case that arose from out-of-state events; concerns an out-of-state plaintiff; and, regardless of outcome, had no effect on the legality of their own conduct. Contrary to CFTC's assertion, Defense Distributed does not suggest "that it would be an abuse of discretion not to transfer this case to the location of the businesses that operate the PredictIt exchange."

CFTC's second assertion is irrelevant. Why does it matter if a local interest is shared amongst many judicial districts? The transfer analysis is relative. That an interest is highly diffuse (but not completely diffuse) only increases the chance it is regarded as equally important by citizens in both the transferor and transferee districts (thereby netting out to zero).

Here, events giving rise to the suit can be separated into two categories: The first concerns individual traders who purchased contracts on the marketplace. Five of them are based in Austin, bought their contracts from Austin, and have been harmed in Austin. The second category deals with marketplace service providers—Aristotle and PredictIt. Based in D.C., they "expended significant resources to assist Victoria University in developing and operating the PredictIt Market" and "will be forced to incur massive administrative, labor, time, and other costs if forced to liquidate pending contracts prematurely."

Given those events, there is an obvious connection between the facts giving rise to this case and the Western District of Texas. And, if we assume that Aristotle's and PredictIt's development and operations activities occurred in D.C., there is also a factual connection with D.D.C.

Additionally, the effect of this suit is completely diffuse. Should plaintiffs prevail on their APA challenge, this court must "set aside" CFTC's ultra vires recission action, with nationwide effect. That affects persons in all judicial districts equally.

In sum, the local interest is, at most, neutral between the Western District of Texas and D.D.C. By finding that local interests "weighed heavily in favor of" transfer to D.D.C., the district court clearly abused its discretion. C. Private Interest Factors

The district court found that "the private interest factors in this case are neutral." But little reasoning supports that finding. Indeed, the order's discussion of the private-interest factors spans all of two sentences: In the first sentence, the court unquestioningly assumed "that this case will be decided on the administrative record, and likely by cross-motions for summary judgment." Then, in the second sentence, the court listed the four private interest factors and declared they were "neutral."

On appeal, petitioners contend the district court "clearly abused its discretion by punting on the so-called 'private interest factors' and claiming that everyone agreed they were 'neutral.' " Though petitioners agree that APA challenges are evaluated purely on the administrative record, they contend that this case is more complicated. Specifically, petitioners point to the various "threshold issues" CFTC has pressed—and contemplates pressing in the future. Resolving those threshold issues, petitioners aver, may require evidentiary hearings and testimony from Austin-based plaintiffs.

In response, CFTC asserts that there is little chance that evidentiary hearings will be needed to resolve issues of standing because "courts in [APA] cases routinely resolve standing issues based on declarations" (citing Sec. Indus. & Fin. Mkts. Ass'n v. U.S. Commodity Futures Trading Comm'n, 67 F. Supp. 3d 373, 401-12 (D.D.C. 2014)). So, based on its reading of the tea leaves, CFTC predicts that this case will be resolved without any future evidentiary hearings.

Petitioners are correct. The district court clearly abused its discretion by finding that all four private interest factors were "neutral." Section 1404(a)'s good-cause requirement prohibits a district court from relying on "conclusory assertions" in considering the transfer factors. But speculation is all the district court used to consider the private interest factors:

First, the court assumed that the case "w[ould] be decided on the administrative record." For starters, that assumption had already been disproven by the time the court issued its transfer order. Moreover, its reason for accepting that assumption relies on a mischaracterization of the plaintiffs' contentions.

According to the court, plaintiffs did "no[t] serious[ly] dispute" that the case would be "decided on the administrative record." Plaintiffs made no such concession. Instead, they flagged issues of "[s]tanding and injury" in their memorandum opposing transfer, expressly alerting the court to the existence of non-APA issues.

Second, the district court erred by uncritically accepting CFTC's claim that the case will not require evidentiary hearings. CFTC buttresses that assertion on appeal with nothing but a mere "see, e.g.," citation to one D.D.C. case. So, at most, the district court could have reasoned that evidentiary hearings were unlikely. But the court went all the way: By concluding that all private factors were "neutral," it found implicitly that there was zero chance that the case would need any evidentiary hearing. That finding "abuse[s] both logic and the court's discretion." Def. Distributed, 30 F.4th at 434.

Given the possibility of evidentiary hearings, we must assess the third private interest factor—the cost of attendance for willing witnesses. Testifying imposes myriad external costs on witnesses. The third factor attempts to internalize and minimize those costs by favoring the venue that is more convenient from the perspective of willing witnesses. See TikTok, 85 F.4th at 361; Volkswagen II, 545 F.3d at 317.

Our circuit assesses convenience as a function of distance. We begin by determining whether the transferee venue is more than one hundred miles from the transferor. Where that is so, the factor of inconvenience borne by witnesses positively and linearly scales with the additional distance they must travel. See TikTok, 85 F.4th at 361-62 & nn. 8-9.

Plaintiffs have identified Kevin Clarke, a resident of Austin, as the primary source for testimony regarding standing and injury. CFTC does not identify any relevant witness—much less one residing near the transferee venue. By transferring the case to D.D.C., Clarke—the primary witness in this case—must now travel over one thousand additional miles to testify. As should be plain, that is a massive inconvenience.

Accordingly, the third private-interest factor strongly favors the Western District of Texas over D.D.C. The district court clearly abused its discretion in deciding otherwise. D. Good Cause

The propriety of § 1404(a) transfer turns on CFTC's meeting its burden of "adduc[ing] evidence and arguments that clearly establish good cause for transfer based on convenience and justice." Def. Distributed, 30 F.4th at 433; see also Planned Parenthood, 52 F.4th at 629 ("The ultimate inquiry is whether the destination venue is clearly more convenient than the venue chosen by the plaintiff." (cleaned up)).

CFTC utterly fails to meet its burden "clearly [to] establish good cause." Id. Of the eight factors, six are neutral and therefore lend CFTC no support for transferring to D.D.C. Worse still, one of the factors—the cost of attendance for willing witnesses—reveals that CFTC's venue of choice results in a net loss of convenience. Assessing the public- and private-interest factors yields but one message: Beyond dispute is the fact that CFTC totally, absolutely, and unqualifiedly failed to show "good cause" for transferring to D.D.C.

Further, we struggle to comprehend how that message could have been any louder or clearer, given that none of the factors favors CFTC's venue of choice. Indeed, not even the court congestion factor supports transferring the case to D.D.C.—it would be a stretch to say that court congestion "favors D.D.C." and not just transfer "somewhere else."

It is well-settled law that § 1404(a) transfer cannot be granted solely because of court congestion. As a premier learned treatise forcefully states: "[R]educ[ing] congestion on the local docket[,] . . . [b]y itself, . . . is not an appropriate factor on a motion to transfer under Section 1404(a)." The conclusion is plainly obvious: "[D]ocket conditions . . . [are] not decisive." Id. (emphasis added). And any doubt to the contrary has long been eliminated in this circuit. In TikTok, we expressly held—in no uncertain terms—that "no factor is of dispositive weight." 85 F.4th at 358 (citing Radmax, 720 F.3d at 290 n.8). No actually means no.

Nor was that unequivocal holding in TikTok borne from whole cloth. The rationale underlying TikTok's holding has long existed in this circuit's caselaw. Court congestion alone cannot justify § 1404(a) transfer because it is speculative and ignores the plaintiffs' role as master of the complaint.

In Planned Parenthood, we expressly acknowledged holdings characterizing court congestion as a "speculative" factor. 52 F.4th at 631 (quoting In re Genentech, Inc., 566 F.3d 1338, 1347 (Fed. Cir. 2009) (internal quotation marks omitted)). We further declined to make any representation regarding "the extent docket efficiency can be reliably estimated" in Planned Parenthood. Id. That makes sense, for measuring congestion is easier said than done. True, we recognize that the district court is in a better position to estimate the efficiency of its own docket. See id. But that structural advantage disappears when it comes to estimating congestion in the transferee venue. Though the district court may, in exercising its discretion, conclude that its docket is more congested than the transferees', that conclusion—which, at its core, is an uninformed guess—alone cannot be dispositive.

Moreover, permitting § 1404(a) transfer solely because of court congestion would call into question the point of having different judicial districts at all. Indeed, those transfers would create but a poor approximation of a scheme dynamically to balance caseload amongst district courts nationwide. That cannot be, for we "must also give some weight to [a] plaintiff['s] choice of forum" and his role as master of the complaint.

Consequently, the district court clearly abused its discretion by transferring the case when none of the factors specifically weighed in favor of transferring the case to D.D.C., and one of the factors favored keeping the case in the Western District of Texas.

V.

There are three requirements for a writ of mandamus: First, "there must be 'no other adequate means to attain the relief . . . desire[d].' " TikTok, 85 F.4th at 357 (quoting Cheney v. U.S. Dist. Ct. for Dist. of Columbia, 542 U.S. 367, 380-81, 124 S.Ct. 2576, 159 L.Ed.2d 459 (2004)). Second, "the right to issuance of the writ must be clear and indisputable." Id. at 357-58 (cleaned up). Third, "the issuing court, in the exercise of its discretion, must be satisfied that the writ is appropriate under the circumstances." Id. at 358 (cleaned up).

The first requirement is met because there is no other avenue for petitioners to seek recourse for the erroneous transfer decision. See Radmax, 720 F.3d at 287 n.2.

The second requirement for mandamus is also satisfied. As explained, see supra part IV, the district court's transfer decision was patent error. And, as we observed in TikTok, we have granted mandamus for less egregiously erroneous transfers. Indeed, in Radmax, we granted mandamus when "three factors weighed in favor of transfer" and "five factors were neutral." TikTok, 85 F.4th at 358 (citing Radmax, 720 F.3d at 290).

Not a single relevant factor favors CFTC's chosen venue of D.D.C. So, petitioners' right to the writ is clear and indisputable.

That leaves the last mandamus requirement—"whether, in the exercise of our discretion, we are satisfied that the writ is appropriate under the circumstances." Id. at 367 (cleaned up). "Writs of mandamus are supervisory in nature and are particularly appropriate when the issues also have an importance beyond the immediate case." Id. (cleaned up).

TikTok is the last in a line of Fifth Circuit cases recognizing that "§ 1404(a) decisions often have importance beyond the immediate case because venue transfer decisions are rarely reviewed, and district courts have applied our tests with too little regard for consistency of outcomes." Id. (cleaned up). That reasoning continues to apply in full force to our instant petition. See id.

Petitioners meet all three requirements for mandamus. Accordingly, issuance of the writ is appropriate.

* * * * *

The district court transferred petitioners' case to D.D.C. essentially because of court congestion. That is a clear abuse of discretion, for which plaintiffs have no other remedy, with sweeping implications for our circuit's § 1404(a) transfer standards.

The petition for writ of mandamus is GRANTED. The district court is directed to request return of the case from D.D.C.