dissenting:
I respectfully dissent. As the majority notes, Valdez-Novoa’s case turns on whether the Immigration Judge’s (“IJ’s”) failure to advise him of his apparent eligibility for voluntary departure was a due process violation and whether that violation was prejudicial. I write separately because the government earlier conceded that there was a due process violation, and because the majority elevates our benchmark for prejudice, the “plausibility” inquiry, to the higher standard of either preponderance or probability. Because it is plausible that the IJ would have exercised discretion to grant voluntary departure, I would reverse the district court’s judgment.
Before the district court, the government conceded that there was a “due process violation, in light of later — developing case law,” and focused its argument solely on the prejudice issue. The court implicitly accepted the violation as conceded, and it too addressed only prejudice. The government has now entirely changed its position, and maintains on appeal that there was no due process violation in the earlier proceedings.
The majority reasons that we may reach the government’s novel argument under an exception permitting us, in certain conditions, to reach “[ijssues not presented to the district court ... [and presented] for the first time on appeal.” United States v. Carlson, 900 F.2d 1346, 1349 (9th Cir.1990). But this is not an issue that is being raised for the first time on appeal. To the contrary, the issue was squarely before the district court, and the government affirmatively abandoned it. Waiver or judicial estoppel are the appropriate doctrines. To ignore those doctrines is to permit a party to relitigate an issue already abandoned.1
Nor does this situation fall, as the government suggests, under the intervening law exception. The government argues that the law changed between the district court proceeding in 2011 and this appeal, citing United States v. Vidal-Mendoza, 705 F.3d 1012 (9th Cir.2013). This argu*1036ment fails because Vidal-Mendoza did not “fundamentally alter[ ] the applicable law.” Indeed, the language the government cites from Vidal-Mendoza comes from United States v. Lopez-Velasquez, 629 F.3d 894 (9th Cir.2010) (en banc), which was decided three years earlier and before Valdez-No-voa was even indicted.2 The government should not be permitted to resurrect on appeal an argument that it affirmatively jettisoned before the district court.
We should not countenance a change in position simply because the government has changed its mind. To do so would make every appeal a jump ball situation in which the appeal is just another chance to raise a new issue and tip the litigation game in one party’s favor without regard to what happened in the trial court. This approach is particularly unfair when the appellee adopts a contradictory position on appeal, since the appellant has no warning and cannot tailor its opening brief to meet the argument. The circumstance here is nothing more than the government’s created “exigenc[y] of the moment.” New Hampshire v. Maine, 582 U.S. 742, 750, 121 S.Ct. 1808, 149 L.Ed.2d 968 (2001). To the extent we have discretion to entertain the government’s about-face, we should decline to do so here. To do otherwise makes a sham of the principles of waiver and estoppel.
The majority concludes, in the alternative, that Valdez-Novoa was not prejudiced by the due process violation. I likewise disagree with this conclusion. To establish prejudice, Valdez-Novoa “does not have to show that he actually would have been granted relief. Instead, he must only show that he had a ‘plausible’ ground for relief from deportation.” United States v. Ubaldo-Figueroa, 364 F.3d 1042, 1048 (9th Cir.2004) (quoting United States v. Arrieta, 224 F.3d 1076, 1079 (9th Cir.2000)).3
Although our cases reference the “plausibility” standard, we have yet to define it with precision. We have set a floor: “establishing ‘plausibility’ requires more than establishing a mere ‘possibility.’ ” United States v. Barajas-Alvarado, 655 F.3d 1077, 1089 (9th Cir.2011). We have also explained, in the context of a § 212(h) waiver, that while the evidence did not guarantee a waiver, “it provide[d] the ‘something more’ that ma[de] it plausible.” Arrieta, 224 F.3d at 1082-83. While citing earlier cases, the majority fails to articulate an actual plausibility standard. I agree with the majority that the standard is not, as Valdez-Novoa advocates, whether it would have been an abuse of discretion for an IJ to grant voluntary departure. But neither do our cases require a petitioner to find a case that is “on all fours” with his own to demonstrate plausibility. That would elevate our plausibility standard to a probability standard.
Two Supreme Court decisions underscore that the plausibility bar is low and, at the very least, that it does not rise to the level of “probability.” The Court’s formulation in Armour v. City of Indianapolis is instructive: a “plausible reason” *1037is one in which “there is any reasonably conceivable state of facts that could provide a rational basis for the classification.” — U.S. -, 132 S.Ct. 2073, 2080, 182 L.Ed.2d 998 (2012) (internal citation and quotation marks omitted). Similarly, in Ashcroft v. Iqbal, the Court specified that “[t]he plausibility standard is not akin to a ‘probability requirement,’ but it asks for more than a sheer possibility that a defendant has acted unlawfully.” 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Considering the Supreme Court’s formulation coupled with our precedent, I believe that our “plausibility” standard is best understood as more than a mere possibility but less than a probability: in short, as “reasonably conceivable or possible.”
In considering whether to grant voluntary departure, immigration judges are required, as the majority acknowledges, to “weigh favorable and unfavorable factors.” Campos-Granillo v. INS, 12 F.3d 849, 852 (9th Cir.1993). Favorable factors include
family ties within the United States; residence of long duration in this country, particularly if residence began at a young age; hardship to the petitioner or petitioner’s family if relief is not granted; service in the United States armed forces; a history of employment; the existence of business or property ties; evidence of value and service to the community; proof of rehabilitation if a criminal record exists; and other evidence attesting to good character.
Id. at 852 n. 8.
Valdez-Novoa met a number of these criteria in 1999. At that time, Valdez-Novoa had a plausible claim for relief based on his arrival in the United States as a nine-year-old boy and his continuous residence for sixteen years between his arrival and deportation in 1999; his three U.S. citizen siblings, his mother and father who held legal permanent resident (“LPR”) status and his five additional siblings, all of whom ultimately gained legal status; and his educational certification as a welder and his extensive employment history as a welder, field worker, and fence repairer.4
To be sure, Valdez-Novoa has an extensive criminal history, much of it stemming directly from his past abuse of alcohol. But the BIA has supported voluntary departure, or remanded to the IJ to consider whether to grant voluntary departure, for aliens with extensive criminal records, including alcohol-related crimes. The BIA has done so in cases involving, among others, six criminal convictions including battery, drunkenness, threatening, and driving under the influence, In re Pineda-Castellanos, 2005 WL 3833024 (BIA Nov. 16, 2005); attempted theft, In re Toledo-Alvarado, 2014 WL 1278406 (BIA Feb. 26, 2014); “Class A” assault with family violence, In re Tipaz-Poncio, 2014 WL 1401572 (BIA Mar. 19, 2014); two convictions for driving while intoxicated, one “recent,” In re Ortiz-Bustos, 2014 WL 1652408 (BIA Apr. 10, 2014); four convictions for assault, a conviction for resisting arrest, and numerous arrests, purportedly as a result of alcohol abuse, In re Gonzales-Figeroa, 2006 WL 729784 (BIA Feb. 10, 2006); a conviction for sexual battery, In re Villalongja Mante, 2007 WL 1676929 (BIA May 18, 2007); convictions for “two serious crimes[] involving controlled substances and robbery” and an arrest for driving under the influence, In re Guillermo Ramirez, 2005 WL 698425 (BIA Mar. 8, 2005); convictions for domestic violence, possession of a controlled substance, and *1038driving under the influence, In re Hernandez-Barreto, 2004 WL 2943517 (BIA Oct. 29, 2004); and convictions for driving under the influence, burglary, and disorderly conduct, In re Reyes-Jiminez, 2004 WL 2418597 (BIA Oct. 4, 2004). These cases underscore the wide range of cases in which the BIA has determined that it was appropriate to consider the exercise of discretion. They fall squarely in the category of plausible relief. The majority’s backhanded dismissal of these cases simply underscores that it is elevating the plausibility standard to one of preponderance or probability.
I highlight cases involving alcohol and violence like battery or domestic violence to illustrate that the BIA has exercised its discretion to grant voluntary departure in a wide range of cases involving criminal histories that equal or exceed that of Valdez-Novoa. Of course, we cannot know for certain whether the BIA would have done so here, but this is not our charge. The question before us is not whether Valdez-Novoa would probably or inevitably have been granted relief, but only whether, at the time his rights were violated, he had a “plausible” ground for relief. Detailing, as the majority does, cases in which the BIA chose not to exercise its discretion underscores only that the choice to grant or not lies with the BIA. Those cases do not prove, given the ranks of cases similar to Valdez-Novoa’s in which relief was granted, that relief was not “plausible.” When considered in light of the substantial positive factors, the combination of favorable and unfavorable factors easily satisfies a standard of “plausibility” for relief from deportation. Valdez-Novoa has established prejudice under Ubaldo-Figueroa, 364 F.3d at 1048.
The existence of such a wide range of cases — both granting and denying relief— only highlights that the BIA and IJs have taken a broad view of when to grant voluntary departure. I fear that the panel’s contrary result in this case turns the plausibility standard into something it was never intended to be: a preponderance or probability standard. The standard is low because the affront inherent in the violation is great. Without a true plausibility standard and the remedy of dismissal of the indictment, there would be no effective deterrent to due process violations in this context. I therefore respectfully dissent.
. Because, in my view, Valdez-Novoa had not "previously been lawfully ... removed,” I would likewise hold that the fourth element of the crime of illegal reentry here was not satisfied, United States v. Gracidas-Ulibarry, 231 F.3d 1188, 1196 (9th Cir.2000) (en banc), for purposes of the corpus delicti rule.
. Other circuits have adopted different standards for demonstrating prejudice. See, e.g., United States v. Charleswell, 456 F.3d 347, 362 n. 17 (3d Cir.2006) (adopting a "reasonable likelihood” standard); United States v. Aguirre-Tello, 353 F.3d 1199, 1209 (10th Cir.2004) (same); United States v. Copeland, 376 F.3d 61, 74 (2d Cir.2004) (adopting a "reasonable likelihood” or "reasonable probability” standard).
. Interestingly, though not part of a retrospective analysis, had the IJ taken a chance on Valdez-Novoa, it would have turned out to be a good one. Valdez-Novoa has been sober since 2007, and now has a U.S. citizen wife and three citizen children.