HOLMES, Circuit Judge.

Grant R. Smith, Assistant Federal Public Defender (Virginia L. Grady, Federal Public Defender, with him on the briefs), Office of the Federal Public Defender, Denver, Colorado, for Defendant-Appellant.

Stuart S. Healy, III, Assistant United States Attorney (L. Robert Murray, Acting United States Attorney, with him on the brief), Office of the United States Attorney, District of Wyoming, Cheyenne, Wyoming, for Plaintiff-Appellee.

Before HOLMES, Circuit Judge, LUCERO, Senior Circuit Judge, and EID, Circuit Judge.

HOLMES, Circuit Judge.

Christopher Dominguez appeals from the district court's denial of his motion to withdraw his guilty plea. He contends that his plea is invalid for two reasons: first, he did not knowingly and intelligently plead guilty, and second, he did not receive the requisite "close assistance" of counsel in making his plea. Based on these contentions, he asks that we reverse the district court's decision and remand with instructions to vacate his plea. We decline to do so. Despite his arguments to the contrary, Mr. Dominguez does not convince us that he unknowingly and unintelligently pleaded guilty or that he did not receive "close assistance" of counsel. Accordingly, exercising jurisdiction under 28 U.S.C. § 1291, we affirm the district court's judgment.

I

This case arises from Mr. Dominguez's involvement in a series of robberies in New Mexico and Wyoming. On December 3, 2016, Mr. Dominguez was arrested along with two other suspects, Antoine Mitchell and Moses D. Dickens III, for the armed robbery on that same day of the Medicine Shoppe Pharmacy in Raton, New Mexico. There were a number of salient features of the robbery. The robbery was committed by only three men. Prior to the robbery, the robbers had stolen a vehicle, which they used during the robbery and abandoned thereafter. During the robbery, the robbers wore face coverings, displayed firearms, and ordered the pharmacy employees to load certain drugs into black trash bags that the robbers supplied. The robbers identified the requested drugs by name, specifically mentioning "Oxy 30," by which they apparently meant Oxycontin and Oxycodone.

After Mr. Dominguez and his cohorts were arrested, the Raton Police Department posted information concerning the robbery, including photos, on the Department's Facebook page. This post drew the attention of a pharmacist who had been shot during a robbery of the Medicap Pharmacy in Cheyenne, Wyoming, on October 6, 2016. The salient features of that robbery closely resembled those of the Raton, New Mexico, robbery. However, notably, the robbers in the Wyoming robbery of the Medicap Pharmacy obtained the stolen vehicle used in the robbery by carjacking a female victim. Moreover, not only did they display their firearms, they also engaged in a gun battle with the pharmacist before fleeing the scene.

After seeing the Raton Police Department Facebook information, the Medicap pharmacist alerted law enforcement, which began investigating whether the suspects in the New Mexico robbery had also committed the prior Wyoming robbery. Law enforcement would eventually directly connect Messrs. Mitchell and Dickens to the Wyoming robbery through DNA uncovered in the carjacking victim's stolen vehicle and through that victim's eyewitness identifications of the two men. Furthermore, one of the firearms that law enforcement seized from a vehicle involved in the New Mexico robbery was forensically shown to have fired shells recovered at the scene of the Wyoming robbery. Though Mr. Dominguez was arrested with Messrs. Mitchell and Dickens after the New Mexico robbery, there was no evidence directly connecting him to the Wyoming robbery. However, law enforcement made note that the vehicle in the driveway of Mr. Dominguez's girlfriend matched the description of the vehicle that the Wyoming robbers used during the victim's carjacking and to flee the robbery scene.

Federal grand juries in New Mexico and Wyoming returned multi-count indictments charging Mr. Dominguez and the other two men with various crimes related to these robberies. Mr. Dominguez pleaded not guilty to all charges.

Among Mr. Dominguez's charges were three separate violations of 18 U.S.C. § 924(c) —one in relation to the New Mexico robbery, and two in relation to the Wyoming robbery. Under § 924(c),

any person who, during and in relation to any crime of violence or drug trafficking crime ... uses or carries a firearm, or who, in furtherance of any such

crime, possesses a firearm, shall, in addition to the punishment provided for such crime of violence or drug trafficking crime—(i) be sentenced to a term of imprisonment of not less than 5 years; (ii) if the firearm is brandished, be sentenced to a term of imprisonment of not less than 7 years; and (iii) if the firearm is discharged, be sentenced to a term of imprisonment of not less than 10 years.

18 U.S.C. § 924(c)(1)(A). Notably, as to the New Mexico robbery, Mr. Dominguez's only § 924(c) charge alleged that he brandished a firearm and thus invoked the seven-year mandatory minimum sentence; and, as to the Wyoming robbery, his two § 924(c) charges each averred that Mr. Dominguez discharged a firearm, which invoked the ten-year mandatory minimum sentence.

Section 924(c) requires that each term of imprisonment for a violation of the statute run consecutively to each of a defendant's other terms of imprisonment. See id. § 924(c)(1)(D)(ii) ("[N]o term of imprisonment imposed on a person under [ § 924(c) ] shall run concurrently with any other term of imprisonment imposed on the person, including any term of imprisonment imposed for the crime of violence or drug trafficking crime during which the firearm was used, carried, or possessed."). Additionally, at the time Mr. Dominguez was charged , § 924(c) contained a "stacking provision," which mandated mandatory minimum sentences of twenty-five years for second or subsequent § 924(c) violations. See id. § 924(c)(1)(C) (2006) ("In the case of a second or subsequent conviction under [ § 924(c) ], the person shall ... be sentenced to a term of imprisonment of not less than 25 years ...."), amended by First Step Act of 2018, Pub. L. No. 115-391, § 403(a), 132 Stat. 5194 (Dec. 21, 2018). Consequently, under the version of § 924(c) in effect when he was indicted, Mr. Dominguez faced a mandatory minimum sentence of sixty years if convicted on all three § 924(c) violations with which he was charged.

Against this backdrop the parties began plea negotiations in November 2018. By mid-December, the parties reached an agreement, which was reduced to writing and signed by Mr. Dominguez on December 20, 2018. Under the agreement's terms, Mr. Dominguez would plead guilty to three counts from the Wyoming indictment—including one § 924(c) violation—and one count from the New Mexico indictment. Mr. Dominguez would also waive his right to appeal his convictions and sentences. In exchange, the government would dismiss the remaining charges against him, including the two additional § 924(c) violations with which he was charged—thereby resolving both the New Mexico and Wyoming cases. Further, the government agreed not to charge Mr. Dominguez with any conduct for which he was being investigated related to a bank robbery that occurred in 2015 in Albuquerque, New Mexico.

For the four charges to which he would plead guilty, the parties agreed in the written plea agreement, pursuant to Federal Rule of Criminal Procedure 11(c)(1)(C), that Mr. Dominguez would serve a total of twenty-eight years' imprisonment. Specifically, with respect to the Wyoming charges, Mr. Dominguez agreed to serve eight years for Count 2, Carjacking and Aiding and Abetting, in violation of 18 U.S.C. §§ 2, 2119 ; ten years for Count 5, Attempted Robbery Involving Controlled Substances and Aiding and Abetting, in violation of 18 U.S.C. §§ 2, 2118(a)(1)(3)(c)(1) ; and ten years for Count 7, Using, Carrying, and Discharging a Firearm During and in Relation to a Crime of Violence, in violation of 18 U.S.C. § 924(c)(1)(A)(iii) —all to run consecutively. As well, with respect to the New Mexico case, Mr. Dominguez agreed to serve an eighteen-year sentence for Count 1, Hobbs Act Robbery, in violation of 18 U.S.C. § 1951(a), which would run concurrently with his sentences for the Wyoming charges.

Notably, in the plea agreement, Mr. Dominguez averred at numerous points that he understood the nature of the charges against him, what his sentences would be, and that he entered into the agreement's terms knowingly and voluntarily. See R., Vol. III, at 15–16 (Plea Agreement, filed Dec. 20, 2018) (stating that Mr. Dominguez "understands," inter alia , the minimum and maximum sentences associated with the charges to which he was pleading guilty); id. at 17 (stating that he "understands that by entering this agreement and pleading guilty he waives certain rights"); id. (acknowledging that "he understands the nature and elements of the offenses he is pleading guilty to"); see also id. ("The Defendant acknowledges he is entering this agreement and pleading guilty freely and voluntarily because he is, in fact, guilty.").

On December 21, 2018, Mr. Dominguez appeared in district court for his change-of-plea hearing. During a thorough Rule 11 colloquy conducted by the court, Mr. Dominguez, under oath, represented (among other things) that he (1) was "fully satisfied with the advice and representation [he had] received ... from [his] attorney"; (2) understood his plea agreement's terms and discussed them with his counsel; (3) signed the agreement voluntarily, and suffered no threats or coercion; and (4) understood the myriad consequences of pleading guilty, including the loss of certain fundamental rights. See id. , Vol. V, at 58–69 (Tr. Change of Plea Hr'g, dated Dec. 21, 2018). He also provided a factual basis for, and affirmatively pleaded guilty to, each of the four charges covered by the plea agreement. Following the colloquy, the district court conditionally accepted the plea agreement, finding that Mr. Dominguez was "fully competent and capable of entering an informed plea," "aware of the nature of the charges" against him, and "aware of the consequences of a plea of guilty to each of those four charges," and also finding "that his pleas of guilty [were] knowing and voluntary pleas, supported by an independent basis of fact sufficient to satisfy all the essential elements of the four offenses to which [he] ... pled." Id. at 82.

But unbeknownst to the parties, the law changed the day of the hearing. Specifically, on December 21, 2018, the President signed into law the First Step Act, a wide-ranging package of criminal justice reform measures that had passed both houses of Congress only days prior. Of note, the First Step Act altered the "stacking" provision of § 924(c) ; now, subsequent to the Act, the twenty-five-year mandatory minimum provision applies only to "violation[s] of [ § 924(c) ] that occur[ ] after a prior [ § 924(c) ] conviction ... has become final ," 18 U.S.C. § 924(c)(1)(C) (2018) (emphasis added), rather than applying to any "second or subsequent [ § 924(c) ] conviction," as it had previously, id. § 924(c)(1)(C) (2006).

This statutory change had significant implications for Mr. Dominguez's potential sentencing exposure: whereas before the First Step Act, he faced a sixty-year mandatory minimum sentence were he to be convicted on all three § 924(c) charges brought in the New Mexico and Wyoming indictments, after the First Step Act, conviction on all these charges would bring only a mandatory minimum sentence of twenty-seven years. See id. § 924(c)(1)(A), (D)(ii) (noting that a defendant shall be sentenced to a seven-year mandatory minimum if he "brandishe[s]" a firearm during the commission of a qualifying crime, and a ten-year mandatory minimum if he "discharge[s]" a firearm during such commission, and that such sentences shall run consecutively); see also Pub. L. No. 115-391, § 403(b), 132 Stat. at 5222 (noting that this amendment "shall apply to any offense that was committed before the date of enactment of this Act, if a sentence for the offense has not been imposed as of such date of enactment"). In Mr. Dominguez's estimation, "the First Step Act significantly undercut the government's bargaining power to convince [him] to plead guilty." Aplt.'s Opening Br. at 8–9.

In light of this purported shift in the parties' bargaining positions, Mr. Dominguez moved to withdraw his guilty plea on February 4, 2019. The district court held a hearing on the motion a few weeks later. At that hearing, Mr. Dominguez's counsel acknowledged that, at the time the parties were negotiating the plea agreement, and "up until the day that [Mr. Dominguez] actually pled guilty," the law "was that you [c]ould stack 924(c)s." R., Vol. V, at 92 (Tr. Hr'g on Def.'s Mot. to Withdraw Guilty Plea, dated Feb. 26, 2019); see id. at 95 (Mr. Dominguez's counsel noting that "[t]he law at the time ... [the parties] negotiated the plea agreement said [the government] c[ould] stack the 924(c)s").

Nonetheless, Mr. Dominguez contended that he should be permitted to withdraw his plea because he had not made the plea knowingly and intelligently and because he had not received "close assistance" of counsel. As to whether his plea was knowing and intelligent, Mr. Dominguez's counsel noted that the evidence against his client in the New Mexico case was "substantial," and that "there was no doubt in [his] mind ... that Mr. Dominguez was likely to be convicted" on the New Mexico charges. Id. at 92–93. Given this virtual certainty, Mr. Dominguez was concerned that, should he also be convicted on the Wyoming charges, he would face a decades-long mandatory minimum sentence, pursuant to the pre-First Step Act version of § 924(c)(1)(C). Yet, he reasoned that, once § 924(c)(1)(C)'s stacking provision was no longer in the picture—due to the First Step Act becoming law—the potential consequences of any New Mexico convictions were substantially less severe. More specifically, following the First Step Act, Mr. Dominguez faced only a minimum sentence roughly half the length of what he faced prior to the Act's passage, even if he was convicted on all of the § 924(c) charges.

Based on this material change that he was not aware of at the time of his guilty plea, Mr. Dominguez maintained his guilty plea was not made knowingly and intelligently. See id. at 107–08 (Mr. Dominguez's counsel defining "knowing" as, inter alia , "understanding the entire legal structure that gives [a defendant] confidence that [he is] standing [before the court] making an intelligent decision [in pleading guilty]," and arguing that, if "knowing" is defined in this way, "then [Mr. Dominguez's plea] wasn't a knowing plea"). And, as to whether Mr. Dominguez received the requisite "close assistance" of counsel, his attorney argued his failure to recognize the potential impact of the First Step Act on Mr. Dominguez's sentencing exposure fell below professional standards. See id. at 103 (Mr. Dominguez's counsel expressing his "belie[f] that [he] should have ... recognized that there was an existing law ... likely to be signed ... the same day that Mr. Dominguez [was] changing his plea ... and figured out whether or not that affected Mr. Dominguez's substantial rights and whether it would have affected his ability to change his plea"); id. at 106 ("Had I [Mr. Dominguez's counsel] read the First Step Act and understood what the stacking provisions meant, [Mr. Dominguez] wouldn't have changed his plea."). Accordingly, Mr. Dominguez argued that his motion to withdraw was justified and should be granted.

However, the district court disagreed and denied the motion. As is germane to this appeal, the district court rejected Mr. Dominguez's "assert[ion] that [his plea] was not knowing because [he] lacked the pertinent information regarding the stacking of the § 924(c) counts that would have altered his decision to plead in this case." Id. , Vol. I, at 43 (District Ct. Order Den. Def.'s Mot. to Withdraw Guilty Plea, filed Feb. 28, 2019). In light of the court's Rule 11 colloquy, in which Mr. Dominguez stated he understood his plea agreement's terms and consequences, the court reasoned that the fact Mr. Dominguez "was not aware of potential statutory changes d[id] not provide a sufficient basis ... to find that his guilty plea was not knowingly entered." Id. at 44.

The court also found unpersuasive Mr. Dominguez's argument that he was denied "close assistance" of counsel, in no small part because Mr. Dominguez "continually stated [at his change of plea hearing] that he was satisfied with the advice of counsel he received throughout the entire process." Id. at 43. Moreover, Mr. Dominguez's attorney "provided [Mr. Dominguez] with advice based on the law that was in place at the time of the negotiations," when "passage of the First Step Act was uncertain"; his attorney's "[f]ailure to fully consider th[e] legislation" until after it was signed into law "d[id] not constitute ... lack of close ... assistance of counsel." Id. at 42–43; see also id. at 43 ("[Mr. Dominguez] is entitled to close assistance of counsel, not clairvoyant nor perfect assistance, and certainly not assistance based on 20/20 hindsight."). The district court, accordingly, denied Mr. Dominguez's motion to withdraw his guilty plea.

Mr. Dominguez now appeals from that denial, asserting two bases for reversal: (1) his plea was not knowingly and intelligently made, and (2) he was denied "close assistance" of counsel. II

We review the district court's denial of Mr. Dominguez's motion to withdraw his guilty plea under our "deferential" abuse-of-discretion standard. United States v. Byrum , 567 F.3d 1255, 1259 (10th Cir. 2009) ; accord United States v. Yazzie , 407 F.3d 1139, 1142 (10th Cir. 2005) (en banc); cf. United States v. Siedlik , 231 F.3d 744, 748 (10th Cir. 2000) ("Defendants do not have an absolute right to withdraw a guilty plea."). To withdraw his guilty plea, Mr. Dominguez "bears the burden of establishing a fair and just reason" for his request. United States v. Marceleno , 819 F.3d 1267, 1272 (10th Cir. 2016) (quoting United States v. Hamilton , 510 F.3d 1209, 1214 (10th Cir. 2007) ); see FED. R. CRIM. P. 11(d)(2)(B) ("A defendant may withdraw a plea of guilty ... after the court accepts the plea, but before it imposes sentence if: ... the defendant can show a fair and just reason for requesting the withdrawal.").

In determining whether Mr. Dominguez has presented a "fair and just reason" for his withdrawal request, we have typically considered the following factors:

(1) whether the defendant has asserted his innocence; (2) whether withdrawal would prejudice the government; (3) whether [he] delayed in filing his motion, and if so, the reason for the delay; (4) whether withdrawal would substantially inconvenience the court; (5) whether close assistance of counsel was available to [him]; (6) whether [his] plea was knowing and voluntary; and (7) whether the withdrawal would waste judicial resources.

Yazzie , 407 F.3d at 1142 (quoting United States v. Sandoval , 390 F.3d 1294, 1298 (10th Cir. 2004) ); see also United States v. Sanchez-Leon , 764 F.3d 1248, 1258 (10th Cir. 2014) (framing these "seven factors" as follows: "(1) whether the defendant has asserted his innocence, (2) prejudice to the government, (3) delay in filing defendant's motion, (4) inconvenience to the court, (5) defendant's assistance of counsel, (6) whether the plea is knowing and voluntary, and (7) waste of judicial resources." (quoting Hamilton , 510 F.3d at 1214 )). Here, Mr. Dominguez challenges the district court's denial of his motion to withdraw his guilty plea based on two of these factors: (1) whether his plea was knowingly and intelligently made and (2) whether he received "close assistance" of counsel. Notably, Mr. Dominguez does not assert that he is innocent of the charged crimes (i.e., the first Yazzie factor), nor does he allege that his plea was involuntary (i.e., part of the sixth Yazzie factor).

Beyond these specific factors, we have acknowledged that, though "motion[s] to withdraw a plea prior to sentencing should be freely allowed," United States v. Garcia , 577 F.3d 1271, 1274 (10th Cir. 2009) (quoting Hamilton , 510 F.3d at 1213–14 ), "[w]hether to permit withdrawal ‘always and ultimately lies within the sound discretion of the district court,’ " Sanchez-Leon , 764 F.3d at 1259 (quoting United States v. Soto , 660 F.3d 1264, 1267 (10th Cir. 2011) ). Within our overarching abuse-of-discretion analysis, "[w]e review legal conclusions de novo, such as whether the plea was made knowingly and voluntarily or whether counsel was ineffective." Sanchez-Leon , 764 F.3d at 1259 ; see also Hamilton , 510 F.3d at 1215–16. "An error of law is per se an abuse of discretion." United States v. Lopez-Avila , 665 F.3d 1216, 1219 (10th Cir. 2011) ; accord Sanchez-Leon , 764 F.3d at 1262.

Recall that Mr. Dominguez does not assert his innocence as a "fair and just reason" for withdrawing his guilty plea. In the context of the two objections that he does raise, this is consequential because we have held that, "[i]f the defendant fails to carry his ... burden on asserted innocence, validity of the plea (whether it was given knowingly and voluntarily), and [also] ineffective assistance of counsel, [we] need not address ‘the remaining factors ... because the[y] ... speak to the potential burden on the government and the court, rather than the defendant's reason for withdrawal.’ " Sanchez-Leon , 764 F.3d at 1258 (second ellipsis in original) (citation and emphasis omitted) (quoting Hamilton , 510 F.3d at 1217 ); see Marceleno , 819 F.3d at 1272 ("If the assertion-of-innocence, knowing-and-voluntary, and ineffective-assistance-of-counsel factors all weigh against the defendant, a district court need not consider the remaining four factors."). Accordingly, if Mr. Dominguez has failed to carry his burden of proof regarding his knowing-and-intelligent and "close assistance" challenges, then we may conclude that Mr. Dominguez has not shown a "fair and just reason" to withdraw his plea because he has made no assertion of his innocence of the charged offenses.

III

Mr. Dominguez claims that the district court erred in denying his motion to withdraw his guilty plea for two reasons: (1) his plea was not knowingly and intelligently made, and (2) his lawyer did not provide "close assistance." However, for the reasons discussed below, we conclude that Mr. Dominguez has failed to demonstrate that his guilty plea was unknowing and unintelligent and that he lacked "close assistance" of counsel. More specifically, as to the latter, we hold that Mr. Dominguez has not demonstrated that he was prejudiced by any deficient performance of his counsel. Consequently, we determine that the district court did not abuse its discretion in denying Mr. Dominguez's motion to withdraw his guilty plea. We turn now to address Mr. Dominguez's arguments for reversal.

A

Mr. Dominguez first argues that his guilty plea is invalid because it was not knowingly and intelligently made. We first outline the governing law and then address the merits of this claim.

1

"A plea of guilty is constitutionally valid only to the extent it is ... ‘intelligent.’ " Bousley v. United States , 523 U.S. 614, 618, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998) (quoting Brady v. United States , 397 U.S. 742, 748, 90 S.Ct. 1463, 25 L.Ed.2d 747 (1970) ). This constitutional protection stems from the Due Process Clause. See, e.g. , United States v. Avila , 733 F.3d 1258, 1261 (10th Cir. 2013) (noting that a defendant may challenge the knowing and intelligent nature of his guilty plea under the Due Process Clause); see also Fields v. Gibson , 277 F.3d 1203, 1212–13 (10th Cir. 2002) ("The Due Process Clause of the Fourteenth Amendment requires that a defendant knowingly ... enter a plea of guilty.").

"The longstanding test for determining the validity of a guilty plea is ‘whether the plea represents [inter alia ] a[n] ... intelligent choice among the alternative courses of action open to the defendant.’ " United States v. Carr , 80 F.3d 413, 416 (10th Cir. 1996) (quoting Hill v. Lockhart , 474 U.S. 52, 56, 106 S.Ct. 366, 88 L.Ed.2d 203 (1985) ); accord Sanchez-Leon , 764 F.3d at 1259. That is, for a plea to be knowing and intelligent , the defendant "must have ‘a full understanding of what the plea connotes and of its consequence.’ " Marceleno , 819 F.3d at 1276 (quoting United States v. Hurlich , 293 F.3d 1223, 1230 (10th Cir. 2002) ); see also United States v. Gigot , 147 F.3d 1193, 1199 (10th Cir. 1998) ("A plea must ‘constitute a deliberate, intelligent choice between available alternatives’ in order to be knowingly and intelligently made." (quoting United States v. Rhodes , 913 F.2d 839, 843 (10th Cir. 1990) )).

More particularly, a defendant knowingly and intelligently pleads guilty if he understands his plea's "direct consequences," even if he does not also "understand every collateral consequence of the plea." Hurlich , 293 F.3d at 1230–31 (emphases added); see United States v. Muhammad , 747 F.3d 1234, 1239–40 (10th Cir. 2014) ("[T]here is no requirement that a defendant be advised of all potential collateral consequences of a guilty plea in order for that plea to be ... knowing[ ] and intelligent." (alteration in original) (quoting United States v. Krejcarek , 453 F.3d 1290, 1296 (10th Cir. 2006) )).

Rule 11 of the Federal Rules of Criminal Procedure "prescribes procedures designed to ensure that pleas are entered knowingly and voluntarily." United States v. Gomez-Cuevas , 917 F.2d 1521, 1525 (10th Cir. 1990) ; see United States v. Gordon , 4 F.3d 1567, 1573 (10th Cir. 1993) (noting that district courts "ensure[ ] that ... plea[s] [are] voluntary and knowing by meticulously complying with the requirements of Fed. R. Crim. P. 11"); see also Padilla v. Kentucky , 559 U.S. 356, 391 n.1, 130 S.Ct. 1473, 176 L.Ed.2d 284 (2010) (Scalia, J., dissenting) (noting that the "required colloquy between a federal district court and a defendant required by Federal Rule of Criminal Procedure 11(b) ... approximates the due process requirements for a valid plea" (citing Libretti v. United States , 516 U.S. 29, 49–50, 116 S.Ct. 356, 133 L.Ed.2d 271 (1995) )); Gigot , 147 F.3d at 1197 (reviewing de novo "[w]hether a district court has complied with Rule 11 in accepting a defendant's plea and, accordingly, whether the defendant's plea was knowing, intelligent, and voluntary"). Pursuant to Rule 11(b), the district court, "[b]efore ... accept[ing] a plea of guilty," "must address the defendant personally in open court" and ensure that he is aware of and understands, inter alia , his rights to plead not guilty, to a jury trial, and to counsel—and his waiver of such rights in the event of a guilty plea; "the nature of each charge to which [he] is pleading"; all mandatory maximum and minimum penalties related to the charges to which he is pleading; and "the court's obligation to calculate the applicable sentencing-guideline range and to consider ... other sentencing factors under 18 U.S.C. § 3553(a)." FED. R. CRIM. P. 11(b)(1) ; see also United States v. Rollings , 751 F.3d 1183, 1188 n.2 (10th Cir. 2014) (discussing the requirements of Rule 11 ).

Where the district court complies with Rule 11 and conducts a thorough plea colloquy, in which the defendant affirmatively indicates he understands his plea and its direct consequences—as occurred in Mr. Dominguez's case—we have recognized that such "solemn declarations made in open court carry a strong presumption of verity." See Sanchez-Leon , 764 F.3d at 1259 (quoting Tovar Mendoza v. Hatch , 620 F.3d 1261, 1269 (10th Cir. 2010) ); see also Blackledge v. Allison , 431 U.S. 63, 74, 97 S.Ct. 1621, 52 L.Ed.2d 136 (1977). Indeed, we have concluded that "the truth and accuracy of [a defendant's] statements made at the Rule 11 proceedings should be regarded as conclusive in the absence of a believable, valid reason justifying a departure from the apparent truth of his Rule 11 statements." Hedman v. United States , 527 F.2d 20, 22 (10th Cir. 1975) (per curiam); accord United States v. Weeks , 653 F.3d 1188, 1205 (10th Cir. 2011).

2

Mr. Dominguez argues that his plea was not knowingly and intelligently made and is, therefore, invalid because "he was not aware of his available alternatives." Aplt.'s Opening Br. at ii. Notably, he does not contest the sufficiency of the court's Rule 11 colloquy. However, Mr. Dominguez asserts that he pleaded guilty because he "believed the bargained-for benefit to him was that a 50-year mandatory minimum sentence was taken off the table," when "unbeknownst to [him], the law had changed"—through the passage of the First Step Act—"and at the time he ple[aded] guilty, he was no longer subject to what he believed was the applicable mandatory minimum sentence." Id. at 12. Accordingly, "it was impossible for Mr. Dominguez to make ‘a deliberate, intelligent choice between available alternatives,’ as he was operating under a misunderstanding as to what the available alternatives actually were"—i.e., what the potential sentencing exposure was, if he elected to go to trial rather than plead guilty. Id. at 15 (quoting Gigot , 147 F.3d at 1199 ); see also id. at 12 (arguing that "it was impossible for Mr. Dominguez to make a deliberate choice between alternatives as he was wrongly informed what the alternatives were"). We are not persuaded by Mr. Dominguez's argument. In effect, Mr. Dominguez contends that, in order to have knowingly and intelligently entered a plea agreement, he needed to understand not just the penalties and direct consequences associated with the charges to which he pleaded guilty, but also any penalties or potential consequences associated with charges to which he did not plead guilty—charges that the government actually dismissed here as part of the plea agreement. In this regard, recall that the First Step Act's amendment of the stacking rules only impacted the penalties and consequences of charges to which Mr. Dominguez did not plead guilty. It was those charges that created the specter of the additional fifty-year mandatory minimum sentence that allegedly troubled Mr. Dominguez. The First Step Act's stacking amendment did not impact the charges to which Mr. Dominguez pleaded guilty and as to which the district court undisputedly conducted a thorough Rule 11 colloquy. In light of these circumstances, we are hard pressed to conclude other than that Mr. Dominguez was fully apprised of the direct consequences of his proposed guilty plea and thus knowingly and intelligently made it. See, e.g. , Hurlich , 293 F.3d at 1230–31 ; Muhammad , 747 F.3d at 1239–40.

Not surprisingly, Mr. Dominguez struggles to find authority supportive of his novel position. He offers none that is controlling. And the authority that he does identify is not persuasive. Mr. Dominguez relies heavily on the Fifth Circuit's decision in United States v. Guerra , 94 F.3d 989 (5th Cir. 1996), to argue that his plea was not knowingly and intelligently made. See Aplt.'s Opening Br. at 15–16; Oral Arg. at 9:28–9:39 (Mr. Dominguez's counsel citing Guerra when asked what was his "best case close to these facts"). He claims that, "[j]ust like ... in Guerra , [he] entered into a guilty plea erroneously believing that his sentencing exposure if he went to trial was far greater than it actually was," and so, "like Guerra , [his] conviction is ‘invalid and must be set aside.’ " Aplt.'s Opening Br. at 15–16 (quoting Guerra , 94 F.3d at 995 ). But, as we shall detail, Guerra is patently distinguishable. Stated briefly, in Guerra "the district court did not properly inform the defendant of the maximum penalty he faced and thus did not comply with the mandate of Rule 11," 94 F.3d at 995 ; these important circumstances are not present here, and thus Guerra is patently distinguishable.

A further examination of salient details of the case reinforce the point. There, the district court informed a defendant "that, because of his prior drug convictions, he was subject to enhanced criminal penalties under the federal statutes as a repeat offender .... [and] [t]hus instead of facing a possible 30-year term for both [of his drug-trafficking] counts, the court was of the view that a 60-year term was possible." Guerra , 94 F.3d at 991. However, "the district court was mistaken. The enhanced sentences for repeat offenders were applicable only where a defendant had previously been convicted of federal drug offenses. [The defendant] had been convicted of drug offenses in the courts of Illinois and Texas, but not of the United States." Id. "Unaware of the district court's error," the defendant pleaded guilty to one of the charged drug-trafficking offenses, in exchange for the government's agreement to dismiss the other. Id.

The defendant thereafter sought to challenge the validity of his plea in post-conviction proceedings under 28 U.S.C. § 2255. Notably, in making the threshold determination that the defendant had established prejudice to overcome an alleged procedural bar, the court made the following observation:

We do not find it difficult to imagine that the district court's erroneous statement

to [the defendant] that he faced a possible sentence of 60 years in prison upon conviction on both counts might have led him to enter into plea negotiations for a reduced sentence on a single count, rather than go to trial and face maximum exposure on both counts.

Id. at 994. Having found prejudice to overcome the bar, the Fifth Circuit with little difficulty determined that the record showed that the district court's misstatements concerning "the maximum penalty he faced," were he to be found guilty of the charged offenses, effected a Rule 11 violation. Id. at 995. And, though the court noted that this did not necessarily mean that the defendant "will receive the relief for which he asks" because he still was obliged to show that the Rule 11 violation "constitute[d] a constitutional violation," the court concluded that the defendant made this constitutional showing. Id. In short, it reasoned that because of the "erroneous information" that he possessed "as to the possible penalty he faced," the defendant was "unaware of the true nature of the options"—that is, "[h]e did not know that going to trial would only put him at risk of half the possible sentence he was informed he would face." Id. Consequently, the Fifth Circuit determined that the defendant's plea "was made unintelligently and is therefore invalid." Id.

As these details should underscore, Guerra is patently distinguishable and does not even marginally advance Mr. Dominguez's cause. The Fifth Circuit's decision there turned on the fact that the district court expressly made inaccurate statements to the defendant regarding the penalties that he faced, which effected a Rule 11 violation and dramatically distorted the penalty picture the defendant considered in entering into plea negotiations. But the district court here made no such misstatements.

The district court undisputedly conducted a thorough Rule 11 colloquy that accurately informed Mr. Dominguez of the potential penalties and direct consequences that he faced with respect to the charges to which he pleaded guilty. And, importantly, though the First Step Act's stacking amendment did affect the dismissed charges to which Mr. Dominguez did not plead guilty, there is nothing in the record to indicate that, in the colloquy, the district court said even a word about the penalties associated with those charges or about the potential interplay between those dismissed charges and the ones to which Mr. Dominguez pleaded guilty. And Mr. Dominguez does not argue to the contrary. See Oral Arg. at 13:05–13:18 (The court: "Let me be clear, do I understand correctly, at the time that [Mr. Dominguez] had his Rule 11 colloquy, was there any discussion in the ... colloquy about the 60-year mandatory minimum that he was facing?" Appellant's counsel: "No."). Therefore, unlike in Guerra , it is "difficult to imagine" how the district court's actions could have pressured Mr. Dominguez "to enter into plea negotiations" to avoid the penalties associated with those charges that were ultimately dismissed. See Guerra , 94 F.3d at 994. Therefore, Mr. Dominguez's reliance on Guerra is unavailing.

After oral argument, pursuant to Federal Rule of Appellate Procedure 28(j), Mr. Dominguez also identified the Seventh Circuit's decision in United States v. De La Torre , 940 F.3d 938 (7th Cir. 2019), contending that it supports his "position that his plea was not intelligently made because he was misinformed about the mandatory minimum sentence he would receive if he went to trial and was found guilty." Aplt.'s 28(j) Letter at 1 (filed Mar. 18, 2020). However, even if we put aside the fact that De La Torre is out-of-circuit authority that is not binding on us, it cannot offer Mr. Dominguez succor. Most fundamentally, this is so because the relevant question in this case is not whether a defendant must be aware of any mandatory minimum penalties he might face in order to knowingly and intelligently enter a guilty plea—or, more precisely, whether mandatory minimum penalties always constitute direct consequences of a defendant's plea. Rather, the relevant question in this case is whether mandatory minimum penalties that relate to charges for which the defendant does not plead guilty can be deemed direct consequences of the defendant's plea, especially when the district court does not comment on—let alone misrepresent—the nature of those non-plea mandatory minimums. And De La Torre utters nary a word regarding these circumstances. Accordingly, Mr. Dominguez fares no better with De La Torre than he does with Guerra .

In sum, Mr. Dominguez does not demonstrate how any misunderstanding he may have had about the First Step Act's potential effects on mandatory minimum penalties for charges to which he did not plead guilty are of direct consequence with respect to the charges to which he did plead guilty, such that his purported misunderstanding regarding these effects would render his plea unknowing and unintelligent. Accordingly, we conclude that the district court did not err in rejecting Mr. Dominguez's contention that his plea was not knowing and intelligent.

B

Mr. Dominguez also argues that his plea is invalid because he did not receive "close assistance" of counsel during the plea bargaining process. More specifically, Mr. Dominguez contends that his plea is invalid because his counsel rendered constitutionally deficient assistance of counsel by failing to account for the impending passage of the First Step Act during plea negotiations. He is quick to point out that his trial counsel "admitted complete ignorance of the First Step Act's impact on [his potential sentence] and conceded that he failed to perform any research on the point until after Mr. Dominguez had changed his plea," thereby "candidly conced[ing] that he did not provide ... ‘close assistance of counsel.’ " Aplt.'s Opening Br. at 21. Mr. Dominguez further claims that, had his counsel informed him of the First Step Act's potential impact on his sentence—in particular, the Act's amendment to § 924(c)'s stacking provision—there is a reasonable probability that the outcome of the plea proceeding would have been different because he either would have declined to plead guilty and gone to trial or at least negotiated a better plea agreement. And, as evidence of his claim that he would have rejected the plea agreement, Mr. Dominguez cites the very fact that he filed a motion to withdraw his guilty plea.

The parties frame this "close assistance" of counsel issue as one raising an ineffective-assistance-of-counsel question under the familiar, two-pronged test of Strickland v. Washington , 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). We have previously applied Strickland 's ineffective-assistance standard to the "close assistance" issue when considering a motion to withdraw a guilty plea. See Marceleno , 819 F.3d at 1272 (discussing the "ineffective-assistance-of-counsel factor[ ]" and asking "whether counsel provided effective assistance"); Hamilton , 510 F.3d at 1216 ("When a defendant's challenge to a guilty plea is based on ineffective assistance of counsel, we apply the two-part [ Strickland ] test ...."); Gordon , 4 F.3d at 1570–73 (reviewing a defendant's "challenge to a guilty plea based on a claim of ineffective assistance of counsel" under Strickland and holding that the defendant's "contention ... that original counsel's ineffective assistance is a fair and just reason for withdrawal of his guilty plea ... fail[ed] in light of our conclusion ... that [he] did not receive ineffective assistance of counsel"). Accordingly, for purposes of this appeal, we will view Mr. Dominguez's claim that he did not receive "close assistance" of counsel through the prism of Strickland and its progeny.

1

Under Strickland , Mr. Dominguez must demonstrate both that (1) "his counsel's performance ‘fell below an objective standard of reasonableness’ " and (2) "the deficient performance prejudiced [his] defense." Byrd v. Workman , 645 F.3d 1159, 1167 (10th Cir. 2011) (quoting Strickland , 466 U.S. at 687–88, 104 S.Ct. 2052 ). We "may address the performance and prejudice components [of the Strickland test] in any order, but need not address both if [Mr. Dominguez] fails to make a sufficient showing of one." Cooks v. Ward , 165 F.3d 1283, 1292–93 (10th Cir. 1998) ; see Byrd , 645 F.3d at 1168 ("Courts are free to address these two prongs in any order, and failure under either is dispositive."). Indeed, "in Strickland , the Supreme Court emphasized that ‘if it is easier to dispose of an ineffectiveness claim on the ground of lack of sufficient prejudice, ... that course should be followed.’ " Littlejohn v. Royal , 875 F.3d 548, 552 (10th Cir. 2017) (omission in original) (quoting Strickland , 466 U.S. at 697, 104 S.Ct. 2052 ). And we elect to do so here: that is, because we may readily dispose of Mr. Dominguez's contention of ineffective assistance of counsel on the basis of insufficient prejudice, we do so here.

2

To satisfy Strickland 's prejudice prong, a defendant "must show that there is a reasonable probability that, but for [his] counsel's unprofessional errors, the result of the proceeding would have been different." Cullen v. Pinholster , 563 U.S. 170, 189, 131 S.Ct. 1388, 179 L.Ed.2d 557 (2011) (quoting Strickland , 466 U.S. at 694, 104 S.Ct. 2052 ). In the guilty-plea context, Strickland 's prejudice prong ordinarily requires a defendant to demonstrate "there is a reasonable probability that, but for [his] counsel's errors, he would not have pleaded guilty and would have insisted on going to trial." Hill , 474 U.S. at 59, 106 S.Ct. 366 (emphases added); accord United States v. Kramer , 168 F.3d 1196, 1201 (10th Cir. 1999) ("To show prejudice in the guilty plea context, the defendant must establish that there is a reasonable probability that, but for counsel's errors, the defendant would not have pleaded guilty and would have insisted on going to trial.").

"A reasonable probability is a probability sufficient to undermine confidence in the outcome," which necessitates a " ‘substantial,’ not just ‘conceivable,’ likelihood of a different result." Cullen , 563 U.S. at 189, 131 S.Ct. 1388 (first quoting Strickland , 466 U.S. at 694, 104 S.Ct. 2052, then quoting Harrington v. Richter , 562 U.S. 86, 112, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011) ). "This does not require a showing that counsel's actions ‘more likely than not altered the outcome,’ but the difference between Strickland 's prejudice standard and a more-probable-than-not standard is slight and matters ‘only in the rarest case.’ " Richter , 562 U.S. at 111–12, 131 S.Ct. 770 (quoting Strickland , 466 U.S. at 693, 697, 104 S.Ct. 2052 ); accord Frost v. Pryor , 749 F.3d 1212, 1225 (10th Cir. 2014).

Though "[u]ltimately ... our task is to make a holistic inquiry into all of the" facts bearing on whether a particular defendant suffered prejudice, Heard v. Addison , 728 F.3d 1170, 1183 (10th Cir. 2013), we have stressed that the defendant's " ‘mere allegation’ that he would have insisted on trial but for his counsel's errors, although necessary, is ultimately insufficient to entitle him to relief," Miller v. Champion , 262 F.3d 1066, 1072 (10th Cir. 2001) ; see Heard , 728 F.3d at 1184 ("[W]e remain suspicious of bald, post hoc and unsupported statements that a defendant would have changed his plea absent counsel's errors ....").

Consistent with the proposition that a defendant's say-so is not enough, we have indicated that this prejudice inquiry is largely an objective one. See Heard , 728 F.3d at 1183 (noting that "[o]ur assessment of this [prejudice] prong will of necessity ‘depend in large part’ on objective factors such as whether an unmade evidentiary or legal discovery ‘likely would have changed the outcome of a trial,’ or whether a defense about which the defendant was not advised ‘likely would have succeeded at trial,’ " but "[i]t is uncontroversial to say that this assessment can include ‘objective ’ facts specific to a petitioner such as his age, the length of the sentence he faced under the terms of the plea deal , the prospect of minimizing exposure to other charged counts , and so on" (emphases added) (quoting Hill , 474 U.S. at 59, 106 S.Ct. 366 )); cf. Bonney v. Wilson ("Bonney I "), 754 F.3d 872, 884 (10th Cir. 2014) (stating that "we have been unable to locate any Supreme Court precedent, regardless of factual context, that clearly holds the Strickland [prejudice] standard—in whole or in part—is subjective"; opining that "one may reasonably read Hill as endorsing an objective prejudice standard"; and noting that the Court has cited Strickland "in support of an objective prejudice standard").

In this vein, the Supreme Court in Padilla v. Kentucky advised that, to obtain relief on an ineffective-assistance-of-counsel claim under Strickland , a defendant challenging his guilty plea "must convince the court that [his] decision to reject the plea ... would have been rational under the circumstances ." 559 U.S. at 372, 130 S.Ct. 1473 (emphasis added). We have subsequently incorporated and subsumed this "rationality" factor into Strickland 's reasonable-probability prejudice inquiry. See Heard , 728 F.3d at 1184 (reading Padilla 's "state[ment] that proof of prejudice requires a [defendant] to show that ‘a decision to reject the plea bargain would have been rational under the circumstances’ " as "suggest[ing] an objective floor " or "threshold" on the prejudice inquiry—"somewhere below Hill 's more demanding requirement that the defendant show ‘a reasonable probability that’ he would have gone to trial absent counsel's errors"—and further opining that, if the defendant satisfies this threshold, "we see no reason to blind ourselves to the individual defendant's statements and conduct" in determining whether the reasonable-probability standard is met (first quoting Padilla , 559 U.S. at 372, 130 S.Ct. 1473, then quoting Hill , 474 U.S. at 59, 106 S.Ct. 366 )); see also Bonney v. Wilson ("Bonney II "), 817 F.3d 703, 712 (10th Cir. 2016) ("To show prejudice (the second prong) in the context of a plea agreement, [a defendant] ha[s] to ‘convince the court that a decision to reject the plea bargain would have been rational under the circumstances.’ " (quoting Padilla , 559 U.S. at 372, 130 S.Ct. 1473 )); Bonney I , 754 F.3d at 884 ("[T]he [ Padilla ] Court explained that to establish prejudice based on counsel's failure to advise his client of certain consequences of a guilty plea, a defendant ‘must convince the court that a decision to reject the plea bargain would have been rational under the circumstances .’ " (quoting Padilla , 559 U.S. at 372, 130 S.Ct. 1473 )).

Consonant with Supreme Court precedent, our cases make clear that "it is often quite difficult for [defendants] who have acknowledged their guilt," like Mr. Dominguez, "to satisfy Strickland 's prejudice prong." Bonney I , 754 F.3d at 886 (quoting Padilla , 559 U.S. at 371 n.12, 130 S.Ct. 1473 ); see Lee v. United States , ––– U.S. ––––, 137 S. Ct. 1958, 1967, 198 L.Ed.2d 476 (2017) (" ‘Surmounting Strickland 's high bar is never an easy task,’ and the strong societal interest in finality has ‘special force with respect to convictions based on guilty pleas.’ Courts should not upset a plea solely because of post hoc assertions from a defendant about how he would have pleaded but for his attorney's deficiencies. Judges should instead look to contemporaneous evidence to substantiate a defendant's expressed preferences." (citations omitted) (first quoting Padilla , 559 U.S. at 371, 130 S.Ct. 1473, then quoting United States v. Timmreck , 441 U.S. 780, 784, 99 S.Ct. 2085, 60 L.Ed.2d 634 (1979) )). 3

Mr. Dominguez's prejudice theory is somewhat elusive. That is, Mr. Dominguez measures prejudice in his opening brief in terms of the denial of his right to a trial, see Aplt.'s Opening Br. at 23 (arguing that it is "more than reasonably probable" that "Mr. Dominguez would not have changed his plea," and, therefore, would have gone to trial, "if not for the erroneous advice that he would receive a 60-year mandatory minimum sentence if he was convicted of the three § 924(c) offenses"); however, in his reply brief, he modifies his measure of prejudice to focus on the purported denial of a potentially better (i.e., more favorable) plea agreement, see Aplt.'s Reply Br. at 13, 15 (questioning "whether Mr. Dominguez would have entered into the same plea agreement with the government had he been properly informed of the available alternatives" and arguing that, but for his counsel's errors, he "might still have pleaded guilty, but under a more beneficial plea agreement " (emphases added)). As we will see, under either theory, Mr. Dominguez's prejudice showing is inadequate and, therefore, his ineffective-assistance claim fails.

a

Mr. Dominguez's denial-of-trial-right theory falls squarely within Hill 's rubric. "In cases where a defendant complains that ineffective assistance led him to accept a plea offer as opposed to proceeding to trial, the defendant will have to show ‘a reasonable probability that, but for counsel's errors, he would not have pleaded guilty and would have insisted on going to trial.’ " Missouri v. Frye , 566 U.S. 134, 148, 132 S.Ct. 1399, 182 L.Ed.2d 379 (2012) (quoting Hill , 474 U.S. at 59, 106 S.Ct. 366 ). Under this rubric, it is clear that Mr. Dominguez's mere insistence that he would have maintained his plea of not guilty and proceeded to trial is not enough to show Strickland prejudice. See, e.g. , Miller , 262 F.3d at 1072. And we conclude that Mr. Dominguez's prejudice showing is otherwise insufficient. In particular, he has failed to demonstrate that his rejection of his plea agreement would have been rational under the circumstances. See, e.g. , Heard , 728 F.3d at 1184.

The potential sentencing alternatives drive this point home. Most significantly, as the government stresses, Mr. Dominguez's twenty-eight-year sentence under his plea agreement was very advantageous to him because it was seventy-six months below the low end of the applicable Guidelines range—even when we consider only the offenses to which he pleaded guilty. See Aplee.'s Resp. Br. at 5; R., Vol. III, ¶ 130, at 195 (Am. Presentence Investigation Report ("PSR"), filed Feb. 26, 2019) (noting that Mr. Dominguez's "binding Plea Agreement is advantageous to [him], as it is 76 months below the low-end of the advisory guideline range"); R., Vol. III, ¶ 137, at 196 ("The defendant's guideline range is 292 to 365 months imprisonment, plus a 10 year[ ] consecutive sentence, which results in a total sentence of 34 to 40 years. The Plea Agreement provides for a total sentence of 336 months, or 28 years. ... [T]herefore, the Court will need to vary downward 76 months to achieve the agreed upon sentence.").

Beyond the Guidelines, acceptance of the plea agreement was seemingly rational in light of the statutory penalties associated with Mr. Dominguez's charges. Even putting aside for the moment his § 924(c) offenses, if Mr. Dominguez had gone to trial and been convicted of all of the other charges, he faced very lengthy statutory maximum sentences: up to one hundred years on the charges in the New Mexico indictment and up to sixty-five years on the charges in the Wyoming indictment. Moreover, even if he had gone to trial after enactment of the First Step Act, Mr. Dominguez would not have completely escaped the harsh, consecutive-sentence regime of § 924(c). Indeed, if he had been convicted of all three § 924(c) offenses charged in the New Mexico and Wyoming indictments, after the First Step Act's enactment, Mr. Dominguez would have been subject to a mandatory minimum sentence of twenty-seven years, just based on those offenses alone. That is, he would have faced—based solely on the § 924(c) offenses—a mandatory minimum sentence that was only one year shorter than the total sentence that he faced under his plea agreement with the government.

Moreover, recall that the maximum sentence for all § 924(c) offenses is life imprisonment. Therefore, without a plea agreement, Mr. Dominguez had no assurance that he would only be sentenced to a mandatory minimum for any § 924(c) offense for which he was convicted. Furthermore, this tally of penalties that Mr. Dominguez avoided by pleading guilty does not even account for the potential charges the government agreed not to pursue in relation to a bank robbery in Albuquerque, New Mexico. According to Mr. Dominguez's PSR, this offense would have independently yielded upon conviction a Guidelines range of 324 to 405 months' imprisonment. In sum, Mr. Dominguez's plea agreement was very favorable to him, and this militates strongly in favor of a conclusion that it would have been irrational for him to reject it.

To be sure, this favorability calculus turns on the likelihood that Mr. Dominguez would have been convicted if he had proceeded to trial. In other words, Mr. Dominguez's plea agreement was very favorable to him because it protected him from severe penalties that he likely would have faced upon conviction following trial. But, if it was unlikely that he would have been convicted at trial, then, logically, it would have been more rational for him to reject the plea agreement; indeed, it would have been more reasonably probable that he would do so and go to trial. Cf. Heard , 728 F.3d at 1185–86 (concluding that the defendant had demonstrated he was prejudiced by his counsel's ineffective assistance and that it would have been "rational under the circumstances" to reject his plea bargain and proceed to trial where, inter alia , the defendant "could have mounted several potentially meritorious challenges to his conviction on appeal" (quoting Padilla , 559 U.S. at 372, 130 S.Ct. 1473 )); Ostrander v. Green , 46 F.3d 347, 356 (4th Cir. 1995) (concluding that a "reasonable defendant" had demonstrated that he would not have pleaded guilty, but for his counsel's errors, based in part on the "potential strength of the state's case"—"which [was] hardly invincible on its face"), overruled on other grounds by O'Dell v. Netherland , 95 F.3d 1214 (4th Cir. 1996) (en banc).

But Mr. Dominguez does not meaningfully contend that weaknesses in the government's proof put the prejudice question in a different light. In other words, he does not meaningfully argue that such weaknesses would have made it more rational for him to reject the plea agreement—much less more reasonably probable that he would do so and go to trial. Cf. Moreno-Espada v. United States , 666 F.3d 60, 67 n.7 (1st Cir. 2012) (rejecting defendant's argument that, but for his counsel's ineffective assistance, he would have gone to trial instead of pleading guilty based on the weakness of the government's evidence, where the record belied such an assertion of weakness).

As noted, in seeking to withdraw his plea, Mr. Dominguez has never asserted that he is innocent of the charged crimes. Moreover, at no point in his appellate briefing has he questioned the strength of the government's proof as to the New Mexico offenses. Indeed, before the district court, Mr. Dominguez's counsel noted that the evidence against his client in the New Mexico case was "substantial," and that "there was no doubt in [his] mind ... that Mr. Dominguez was likely to be convicted" on the New Mexico charges. R., Vol. V, at 92–93. As to the Wyoming offenses, Mr. Dominguez hardly does more than assert in conclusory fashion that the government possessed "relatively weak evidence"—noting that "there was no eye-witness or DNA evidence connecting Mr. Dominguez to the crimes," whereas there was such evidence with respect to his co-defendants. Aplt.'s Opening Br. at 4, 15.

But, while the government freely acknowledges that there was no such evidence in the Wyoming case that incriminated Mr. Dominguez, it asserts that "the robbery in [New Mexico] was important evidence tying [Mr. Dominguez] to his crimes in Wyoming" and that "the similarities between the two crimes were so striking" that if the Wyoming charges had been tried, "the district court [had] ruled that the government could present evidence of the [New Mexico] robbery in the Wyoming matter to demonstrate identity, intent, preparation, and common plan." Aplee.'s Resp. Br. at 10. Moreover, the government reminds us that it agreed to forgo charges related to the 2015 New Mexico bank robbery, where it claims there was strong forensic evidence linking Mr. Dominguez to the offense. In response to all of these government assertions, Mr. Dominguez offers nothing but deafening silence.

Accordingly, Mr. Dominguez does not come close to showing that any ostensible weaknesses in the government's proof of guilt would have made it rational for him to forgo his very favorable plea agreement. Nor does he point to any other circumstances that might suggest that his rejection of his plea agreement would be rational. In short, in light of the foregoing, Mr. Dominguez has not satisfied Strickland 's reasonable-probability standard under Hill 's rubric: that is, Mr. Dominguez has not shown that there is a reasonable probability that, but for any deficiency in his counsel's representation concerning the First Step Act, he would have rejected the plea agreement and gone to trial. b

As for Mr. Dominguez's focus on the purported denial of a potentially better plea agreement—that is, his "better-plea" theory—to begin, Mr. Dominguez faces two significant threshold hurdles. But, even if he overcomes them, we conclude that Mr. Dominguez's showing on the merits of the prejudice question comes up short. Turning to the threshold issues, first, a cogent argument could be made that Mr. Dominguez has not preserved this theory by failing to make any argument for it in his opening brief. See, e.g. , United States v. Leffler , 942 F.3d 1192, 1197 (10th Cir. 2019) ("In this Circuit, we generally do not consider arguments made for the first time on appeal in an appellant's reply brief and deem those arguments waived."). The government, however, does not make any such preservation argument—at oral argument or otherwise. Cf. United States v. McGehee , 672 F.3d 860, 873 n.5 (10th Cir. 2012) (noting that, "although we do not accord the fact great weight in our analysis, we note that the government did contend at oral argument that [the defendant] failed to preserve the acceptance-of-responsibility argument"). And, even if Mr. Dominguez has failed to preserve this argument, we possess the discretion to consider it. See, e.g. , Abernathy v. Wandes , 713 F.3d 538, 552 (10th Cir. 2013) ("[T]he decision regarding what issues are appropriate to entertain on appeal in instances of lack of preservation is discretionary."). We elect to do so under the circumstances of this criminal case.

The second hurdle is substantive and more serious. At a minimum, it is open to significant doubt whether the better-plea theory is a cognizable theory of Strickland prejudice. See, e.g. , Short v. United States , 471 F.3d 686, 696–97 (6th Cir. 2006) ("The Supreme Court has developed a specific standard that defendants who plead guilty must meet to demonstrate prejudice with respect to their representation by counsel. ... [I]n the present case, the petitioner's claim of prejudice rests upon an assertion that he wound up with a less favorable plea or sentence than he otherwise would have accepted with the advice of competent counsel. Such a claim is insufficient to establish actual prejudice." (citations omitted)); Bethel v. United States , 458 F.3d 711, 720 (7th Cir. 2006) ("To demonstrate prejudice, the defendant must show that he would not have pled guilty and would have insisted on going to trial. Whether he could have negotiated a better plea deal is irrelevant to the issue of prejudice in the ineffective assistance context."); United States v. Horne , 987 F.2d 833, 836 (D.C. Cir. 1993) ("Nothing in the present record suggests that [defendant] had (or even now has) any intention of pleading not guilty and going to trial. ... It is apparent, therefore, that [defendant] wants to withdraw his guilty plea solely in order to strike a better deal with the prosecutor. The constitutional requirement of effective assistance of counsel is intended, however, to guarantee a fair disposition of defendant's case, not to ensure that he is able to drive the hardest possible plea bargain with the Government."); Fields v. Att'y Gen. of Md. , 956 F.2d 1290, 1297 & n.19 (4th Cir. 1992) (concluding that a defendant could not "make the showing of prejudice required under ... Strickland " where the "crux of [his] case [was] that with effective assistance he would have pled to a different plea bargain and received a more favorable sentence," not that he would have "stood trial"); Craker v. McCotter , 805 F.2d 538, 542 (5th Cir. 1986) ("[Defendant's] allegation here that effective counsel would have negotiated a better bargain also fails to satisfy the prejudice prong of Strickland as interpreted in Hill . Under Hill , the question is not whether [defendant] would have received a better deal with effective representation, but whether he would have rejected the deal offered and gone to trial.").

In particular, our research has not uncovered any Tenth Circuit controlling precedent that has expressly considered whether such a prejudice theory is cognizable under Strickland —much less endorsed it. And Mr. Dominguez has not identified any Tenth Circuit authority to this effect. Indeed, Mr. Dominguez completely elides in his briefing this substantive threshold question. Nevertheless, we need not pause to assess the implications of Mr. Dominguez's briefing failure, nor is it necessary for us to consider (unaided) whether this prejudice theory is cognizable under Strickland . That is because, even if Mr. Dominguez's better-plea theory were cognizable, he would not be able to establish prejudice under it here.

More specifically, Mr. Dominguez's better-plea theory is built on a foundation of speculation and conjecture. He cites to nothing in the record indicating that the government would have offered him a plea agreement more favorable than the one he accepted, regardless of the circumstances. Cf. R., Vol. V, at 117 (the government, at the district court's hearing on Mr. Dominguez's motion to withdraw, noting that the plea agreement was "reasonable" because it offered Mr. Dominguez a sentence "76 months below the guideline range"); id. at 118 (the government emphasizing that "the potential sentences" for Mr. Dominguez's § 924(c) violations "were all exactly the same" after the First Step Act, and that if Mr. Dominguez had been "convicted on all of these charges," he would not have been "looking at anywhere close to 28 years or certainly wouldn't be under almost any reasonable analysis").

Without more than sheer speculation and conjecture suggesting that it was reasonably probable that the government would have given him a more favorable plea agreement, Mr. Dominguez's better-plea theory of prejudice must fail. See, e.g. , Hooks v. Workman , 689 F.3d 1148, 1187 (10th Cir. 2012) ("Reasonable probability is more than mere speculation ...."); Byrd , 645 F.3d at 1168 (noting that "mere speculation is not sufficient to satisfy this [ Strickland prejudice] burden"); cf. Frye , 566 U.S. at 148, 132 S.Ct. 1399 ("[D]efendants who have shown a reasonable probability they would have accepted the earlier plea offer must also show that, if the prosecution had the discretion to cancel it or if the trial court had the discretion to refuse to accept it, there is a reasonable probability neither the prosecution nor the trial court would have prevented the offer from being accepted or implemented. This further showing is of particular importance because a defendant has no right to be offered a plea ...." (emphasis added)); Rodriguez-Penton v. United States , 905 F.3d 481, 493 (6th Cir. 2018) (Thapar, J., dissenting) (noting that, even under a more-expansive reading of Supreme Court precedent, "a defendant still must show a ‘reasonable probability’ that, but for counsel's bad advice, he would have received a chance to negotiate a better offer—something much more difficult in the plea-bargaining context" (quoting Lee , 137 S. Ct. at 1969 )).

In short, Mr. Dominguez's suggestion that he could have negotiated a better plea agreement—if he had been equipped by his lawyer with knowledge of the First Step Act—is objectively groundless. Therefore, even if the better-plea theory were a cognizable theory of Strickland prejudice, Mr. Dominguez would not prevail on it here.

c

All that said, we acknowledge Mr. Dominguez's repeated suggestion that the record potently supports his contention that, if properly advised concerning the First Step Act, he would have rejected the plea agreement, because in fact he took affirmative steps to do so. More specifically, Mr. Dominguez emphasizes that he filed his motion to withdraw his plea "immediately after he became aware of the First Step Act's impact on his case." Aplt.'s Opening Br. at 13 (noting that his responsive action in filing the motion "constitutes per se evidence that he would not have pleaded guilty if he had been correctly informed of the law before he had changed his plea" (emphasis added)); see Aplt.'s Reply Br. at 13 (arguing that "the record demonstrates that there is every reason to question whether Mr. Dominguez would have entered into the same plea agreement with the government had he been properly informed of the available alternatives" and pointing out that "[t]he most telling piece of evidence is that when Mr. Dominguez was properly informed of the law, he moved to withdraw his plea" (emphasis added)).

We do not question that, under certain circumstances, a defendant's affirmative efforts to withdraw his guilty plea—upon receiving reasonably competent advice from counsel—may be a relevant factor in assessing whether counsel's allegedly deficient pre-plea advice prejudiced the defendant under Strickland . See, e.g. , Gonzalez v. United States , 722 F.3d 118, 133 (2d Cir. 2013) ("[T]he fact that an attempt was made to withdraw the guilty plea and go to trial may not be dispositive on the issue of [ineffective assistance of counsel] prejudice; however, it is a factor that must be considered by the court in assessing whether there is a reasonable probability that but for substandard performance by counsel, the defendant would have chosen to eschew the plea and go to trial."). However, Mr. Dominguez does not convince us that his actions regarding his motion to withdraw significantly support his contention that there is a reasonable probability that, but for his counsel's allegedly deficient representation concerning the First Step Act, he would not have entered into his plea agreement.

To begin, Mr. Dominguez's actions relating to the filing of his motion to withdraw arguably were not the product of a genuine desire to proceed to trial, upon receiving constitutionally adequate advice concerning the First Step Act, but, rather, the outgrowth of a negotiating strategy aimed at leveraging the First Step Act's stacking amendment to secure a better plea. See, e.g. , Aplt.'s Reply Br. at 15 (noting that the record shows that Mr. Dominguez's "counsel is equivocating [about whether Mr. Dominguez would have pleaded guilty] because ... Mr. Dominguez might still have pleaded guilty, but under a more beneficial plea agreement " (emphasis added)); see also Gonzalez , 722 F.3d at 133 (noting that "a court might reasonably find in some circumstances that a defendant's request to withdraw his plea of guilty did not reflect a genuine desire to go to trial"). Insofar as that is so, we already have explained why, from an objective perspective, the prospect of a better plea here was speculative and conjectural—and, consequently, an inadequate basis for a claim of Strickland prejudice. See, e.g. , Byrd , 645 F.3d at 1168.

Thus, as relevant here, even assuming that Mr. Dominguez's post-plea knowledge of the First Step Act prompted him to file a motion to withdraw with the subjective aim of getting a better plea agreement, that action would not signify that Mr. Dominguez was objectively prejudiced by his counsel's advice to enter the existing plea agreement because Mr. Dominguez's prospect of a better plea agreement was only speculative and conjectural. See, e.g. , Bonney I , 754 F.3d at 884 (noting that the Supreme Court has cited Strickland "in support of an objective prejudice standard"); Hooks , 689 F.3d at 1187 (noting that "[r]easonable probability is more than mere speculation").

Alternatively, let us assume for the present analysis that Mr. Dominguez's actions regarding the motion to withdraw genuinely reflected a subjective willingness to risk a trial. Even so, Mr. Dominguez has not demonstrated that the cause of his newfound willingness to go to trial—which he swore in open court did not exist at the time he pleaded guilty—stemmed from his post-plea awareness of the First Step Act's implications, rather than from some other reason.

In this regard, recall that Strickland 's prejudice analysis involves a "counterfactual" inquiry that hinges on counsel's alleged ineffective representation—that is, the inquiry turns on whether, but for such ineffective representation, there is a reasonable probability that the outcome of the proceeding would have been different. Headley v. United States , 804 F. App'x 973, 979 (10th Cir. 2020) (unpublished). Consequently, factors unrelated to counsel's ineffective representation do not fall within the compass of Strickland prejudice, even if there is a reasonable probability that, but for those factors, the outcome of the proceeding would have been different.

Here, as shown above, whether or not Mr. Dominguez possessed knowledge of the First Step Act, he would have been hard pressed to see his plea agreement—viewed through an objective lens—as anything other than very favorable. And nothing material about the First Step Act changed after Mr. Dominguez pleaded guilty and the statute became law. But there were changes that did not involve the First Step Act around the time and after Mr. Dominguez pleaded guilty relating to the status of his co-defendants; these changes conceivably could have altered his assessment of the risks of going to trial. Specifically, as the government notes, Mr. Dominguez's "motion came ... well after he knew that he would not sit at trial with his two co-defendants and their DNA implicated cases." Aplee.'s Resp. Br. at 35.

In other words, when Mr. Dominguez filed his motion to withdraw, he knew that his co-defendants—as to whom the government's evidence was admittedly stronger, including DNA evidence—were not going be tried with him because they previously had pleaded guilty. And it is at least arguable that this knowledge was the factor that altered his risk calculation regarding going to trial and prompted him to file his motion to withdraw—not the knowledge that he allegedly should have received from his lawyer concerning the First Step Act.

To be clear, we have no need to reach a definitive conclusion on this matter. It suffices for us to conclude that the timing of Mr. Dominguez's motion to withdraw and, relatedly, his actions in filing that motion do not have the clear, unequivocal import that Mr. Dominguez suggests. That is, these circumstances do not "constitute[ ] per se evidence that he would not have pleaded guilty if he had been correctly informed of the law before he had changed his plea." Aplt.'s Opening Br. at 13; cf. United States v. Santiago Miranda , 654 F.3d 130, 139–40 (1st Cir. 2011) (noting that "the timing of [the defendant's] motion to withdraw his guilty plea is suspect" because the motion was filed "well after [the defendant's] attorney received an unfavorable PSR," indicating that "the benefit of pleading guilty and the risk of proceeding to trial dissipated substantially").

Instead, viewed in the context of the foregoing analysis regarding why it was entirely rational for Mr. Dominguez to enter into his plea agreement, his filing of his motion to withdraw and the timing of it do not alter our view that Mr. Dominguez has failed to show that there is a reasonable probability that, but for any deficiency in his counsel's advice concerning the First Step Act, the outcome of his plea proceeding would have been different. In other words, Mr. Dominguez's filing of his motion to withdraw and the timing of it do not—alone or in combination with other factors—change our bottom-line conclusion that he does not establish Strickland prejudice.

* * *

In the end, we are left with little more than Mr. Dominguez's say-so as to whether he would have rejected his plea agreement if his counsel had informed him of the First Step Act's implications. But while such "bald, post hoc and unsupported statements" may be a necessary predicate to Mr. Dominguez's Strickland prejudice showing, they are insufficient, on their own, to carry his burden of proof. Heard , 728 F.3d at 1184.

IV

For the foregoing reasons, we conclude that the district court did not abuse its discretion in denying Mr. Dominguez's motion to withdraw his guilty plea because (1) his plea was knowing and intelligent and (2) he received "close assistance" of counsel. Accordingly, we AFFIRM the district court's judgment.

LUCERO, J., dissenting:

I appreciate the lengthy and thorough review my colleagues provide in the majority opinion, but I remain unpersuaded. Although I agree that "there is no requirement that a defendant be advised of all potential collateral consequences of a guilty plea," (Op. 1105, quoting United States v. Muhammad, 747 F.3d 1234, 1239-40 (10th Cir. 2014) ), the penalties a defendant avoids by pleading guilty are anything but collateral. Avoidance of potential penalties was the basic impetus that motivated Dominguez to enter into his plea agreement with the United States. Unequivocal failure by counsel for Dominguez to advise him of law that dramatically altered those penalties amounts to ineffective assistance of counsel. As a result, I would conclude Dominguez is entitled to withdraw his guilty plea.

I

Plea bargaining is commonly predicated on avoided penalties. A defendant whose counsel fails to provide even a moderately accurate picture of what penalties a guilty plea would avoid cannot make an informed pleading decision. This follows from the nature of plea bargaining. "[P]lea bargains are essentially contracts." Puckett v. United States, 556 U.S. 129, 137, 129 S.Ct. 1423, 173 L.Ed.2d 266 (2009). When a party to a contract is severely misinformed by counsel about the benefits a contract will yield, counsel has undoubtedly failed the client. In the context of entering a plea, a critical stage of the criminal process, such a failure constitutes ineffective assistance of counsel and is grounds for vacating a plea agreement.

The majority consigns counsel's failure to irrelevancy. In its view, Dominguez's expectation that he would understand the benefits of a potential plea bargain is simply too much to ask—a criminal defendant is entitled to understand only "the penalties and direct consequences associated with the charges to which he pleaded guilty," not "penalties or potential consequences associated with charges to which he did not plead guilty." (Op. 1107). No matter that those charges "to which he did not plead guilty" were a primary motivation behind the defendant's guilty plea.

This cabined view of the "knowingly and intelligently" requirement for guilty pleas cannot be right. Demanding only that a criminal defendant understands the penalties to be received, not the penalties to be avoided, is tantamount to requiring that the defendant only understand half the bargain. This falls well short of what the Constitution and our caselaw demand.

II

When a defendant moves to withdraw a guilty plea prior to sentencing, the court must assess whether there is a "fair and just reason for withdrawal" under Federal Rule of Criminal Procedure 11(d)(2)(B) by considering the factors set forth in United States v. Yazzie, 407 F.3d 1139, 1142 (10th Cir. 2005). We review this determination for an abuse of discretion. Id. Dominguez's challenge to the district court's denial of his motion to withdraw his guilty plea turns on two of the Yazzie factors: "whether close assistance of counsel was available to the defendant ... [and] whether the plea was knowing and voluntary." Id. Although "close assistance of counsel" is not further explained in Yazzie, the Supreme Court has stated that "the right to counsel is the right to effective assistance of counsel." Missouri v. Frye, 566 U.S. 134, 138, 132 S.Ct. 1399, 182 L.Ed.2d 379 (2012).

The analysis of the Yazzie factors relevant to Dominguez's motion to withdraw his guilty plea therefore requires a determination of whether his counsel was ineffective under Strickland, a determination that would also render his plea not voluntary or intelligent. Consequently, the analysis outlined by the Supreme Court in Hill v. Lockhart is the relevant inquiry to assess these Yazzie factors in the context of whether Dominguez provided a "fair and just" reason for withdrawal.

In Hill v. Lockhart, the Supreme Court applied the two-part Strickland test to challenges of guilty pleas based on ineffective assistance of counsel. 474 U.S. 52, 58, 106 S.Ct. 366, 88 L.Ed.2d 203 (1985). "To demonstrate that counsel was constitutionally ineffective, a defendant must show that counsel's representation fell below an objective standard of reasonableness and that he was prejudiced as a result." Lee v. United States, ––– U.S. ––––, 137 S. Ct. 1958, 1964, 198 L.Ed.2d 476 (2017) (quotation omitted). As the Supreme Court explained, "the first half of the Strickland v. Washington test is nothing more than a restatement of the standard of attorney competence already set forth in Tollett v. Henderson ... and McMann v. Richardson." Hill, 474 U.S. at 58-59, 106 S.Ct. 366 (citing Tollett v. Henderson , 411 U.S. 258, 93 S.Ct. 1602, 36 L.Ed.2d 235 (1973) and McMann v. Richardson, 397 U.S. 759, 90 S.Ct. 1441, 25 L.Ed.2d 763 (1970) ). Under McMann and Tollett, whether a guilty plea is unintelligent "depends ... on whether that advice was within the range of competence demanded of attorneys in criminal cases." McMann, 397 U.S. at 770-71, 90 S.Ct. 1441 ; see also Tollett, 411 U.S. at 266, 93 S.Ct. 1602. "In order to satisfy the ‘prejudice’ requirement, the defendant must show that there is a reasonable probability that, but for counsel's errors, he would not have pleaded guilty and would have insisted on going to trial." Hill, 474 U.S. at 59, 106 S.Ct. 366.

Unlike my colleagues in the majority, I conclude that Dominguez has met both requirements of the Strickland test. First, his counsel's performance was objectively deficient.

The first prong—constitutional deficiency—is necessarily linked to the practice and expectations of the legal community: The proper measure of attorney performance remains simply reasonableness under prevailing professional norms. We long have recognized that prevailing norms of practice as reflected in American Bar Association standards and the like are guides to determining what is reasonable.

Padilla v. Kentucky, 559 U.S. 356, 366, 130 S.Ct. 1473, 176 L.Ed.2d 284 (2010) (cleaned up). The American Bar Association Criminal Justice Standards for the Defense Function impose "a duty to be well-informed regarding the legal options and developments that can affect a client's interests during a criminal representation." Am. Bar Ass'n Standards for Criminal Justice: The Defense Function, 4-4.1.3 (4th ed. 2017). As trial defense counsel appears to acknowledge, his failure to research or make himself aware of the pending change in federal sentencing law effectuated by the First Step Act did not fulfill this duty, and that failure was objectively unreasonable.

The First Step Act was "heralded as the most far-reaching overhaul of the criminal justice system in a generation," receiving extensive press coverage during its consideration and passage by Congress. Alan Ellis and Mark H. Allenbaugh, The First Step Act of 2018—A Significant First Step in Sentencing Reform, Bloomberg Law, Jan. 10, 2019 (quotation omitted). Given the critical importance of the First Step Act's changes to the stacking provisions under § 924(c) and the widespread publicity concerning the legislation, defense counsel's failure to research or monitor its passage was objectively unreasonable. See Hinton v. Alabama, 571 U.S. 263, 274, 134 S.Ct. 1081, 188 L.Ed.2d 1 (2014) ("An attorney's ignorance of a point of law that is fundamental to his case combined with his failure to perform basic research on that point is a quintessential example of unreasonable performance under Strickland.").

At the hearing on the motion to withdraw his plea, the district court evaluated defense counsel's efforts and determined that Dominguez received close assistance of counsel. For the district court, defense counsel's failure to advise Dominguez on statutory changes was not deficient, but this finding was premised on the court's erroneous determination that the First Step Act was not yet law:

While the timing of the plea negotiations leading up to the change of plea is unusual, it is important to realize that Defendant was engaged in plea negotiations for several weeks prior to the change of plea hearing, when the passage of the First Step Act was uncertain. Failure to fully consider this legislation that was not signed by the President until (short[l]y) after the change of plea hearing does not constitute the lack of close assistance of counsel.

District Court Order at 8 (emphasis added).

Quite to the contrary, the legislation was signed before the change of plea was entered, and was thus the applicable law for analysis of the constitutional effectiveness of Dominguez's defense counsel. Given that the First Step Act was a significant criminal justice reform measure that received national press coverage, the failure of defense counsel to research, understand, and discuss its ramifications if passed with his client is objectively unreasonable. That conclusion is inescapable in light of the First Step Act's becoming law prior to the change of plea.

Although we review the district court's decision on a motion to withdraw a guilty plea deferentially, a district court "by definition abuses its discretion when it makes an error of law." Koon v. United States, 518 U.S. 81, 100, 116 S.Ct. 2035, 135 L.Ed.2d 392 (1996) ; see also United States v. Lopez-Avila, 665 F.3d 1216, 1219 (10th Cir. 2011). The First Step Act was effective for the entirety of December 21, 2018; it was therefore the law prior to the district court's acceptance of Dominguez's change of plea. By basing its denial of Dominguez's motion on its conclusion that the law did not change until after the change of plea hearing, the district court erred on an issue of law and therefore necessarily abused its discretion.

Moreover, incorrect sentencing law was communicated to the defendant as a result of this change of law. That the pertinent statute was signed into law only just prior to the plea hearing is of no moment; the defendant was misinformed and entered a plea based upon that misinformation. Only if we are prepared to say that recent changes in the law are of no consequence could we possibly ignore such a deficiency. I am not prepared to do so.

Dominguez has also shown prejudice arising from his counsel's ineffective assistance. Contrary to the majority's contention, we have much more than Dominguez's assertion that he would not have changed his plea had he known of the changes wrought by the First Step Act to § 924(c)'s stacking provisions. In the motion to withdraw the plea, Dominguez's trial defense counsel (1) described the central importance of § 924(c)'s stacking provisions to Dominguez's plea, (2) explicitly stated at the withdrawal motion hearing that he would not have recommended the plea agreement had he known of the First Step Act changes to the stacking provisions, and (3) explained that Dominguez would not have pled guilty had he been informed of these changes. Counsel's representations to the district court, particularly when coupled with Dominguez's assertions, demonstrate a strong likelihood that effective assistance of counsel would have affected Dominguez's choice to accept the plea agreement.

The majority contends that it would have been irrational for Dominguez to change his plea, describing the plea agreement as "very advantageous to [Dominguez] because it was seventy-six months below the low end of the applicable Guidelines range ...." (Op. 1113). But in the context of the 336 months to which Dominguez was sentenced under the plea agreement, a 76-month discount for agreeing to plead guilty is hardly overwhelmingly advantageous. It would certainly not have been irrational for Dominguez to chance a trial given how lengthy his plea agreement sentence would be if not for the threatened 60-year mandatory minimum that his counsel led him to believe was hanging over his head. And that is all that Dominguez needs to show: that his "decision to reject the plea bargain would have been rational under the circumstances." Padilla, 559 U.S. at 372, 130 S.Ct. 1473. Having surpassed this "objective floor," Heard v. Addison, 728 F.3d 1170, 1184 (10th Cir. 2013) (emphasis omitted), Dominguez's and his trial counsel's statements that Dominguez would not have pled guilty but for his counsel's ineffective assistance warrant giving Dominguez the opportunity to withdraw his guilty plea.

III

A defense attorney must correctly advise his client on the consequences not only of those charges to which the defendant is pleading guilty but also of the benefits accruing from a guilty plea. These benefits are essential, not collateral, to a plea agreement. They fall squarely within the requirements of Strickland—and it cannot be otherwise, given that plea bargaining "is the criminal justice system." Frye, 566 at 144, 132 S.Ct. 1399. "The longstanding test for determining the validity of a guilty plea is whether the plea represents a voluntary and intelligent choice among the alternative courses of action open to the defendant." Hill, 474 U.S. at 56, 106 S.Ct. 366 (quotation omitted). With the ubiquity of plea bargaining in our criminal justice system, "defendants cannot be left to the mercies of incompetent counsel" in assessing the benefits, as well as the costs, of their plea bargain. McMann, 397 U.S. at 771, 90 S.Ct. 1441.

Because the truncated view of the Strickland inquiry adopted by the majority violates basic constitutional principles that a guilty plea must be made knowingly and voluntarily, and because my colleagues trivialize the importance of material elements of the law inherent to the entry of this plea, I respectfully dissent.