Mendoza v. United States

• Court: District Court, S.D. New York
• Decided: July 14, 2025
• Docket: 1:16-cv-05414
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT || DOCUMENT SOUTHERN DISTRICT OF NEW YORK 7 —— CNICé.ULY FILED

RUDY MENDOZA, “se Tab | Petitioner, -against- 16 CV 5414 (CM) UNITED STATES OF AMERICA, Respondent.

DECISION AND ORDER DENYING MOTION TO VACATE PURSUANT TO 28 U.S.C. § 2255 McMahon, J.: Rudy Mendoza is currently serving a 20 years sentence in connection with his 2013 convictions for, inter alia, racketeering conspiracy, narcotics conspiracy, robbery conspiracy, and violating federal firearms laws. Before the Court is Mendoza’s pro se motion, filed pursuant 28 U.S.C. § 2255, asking the Court to set aside his convictions and sentence. The motion is denied. Background On November 4, 2013, after a nine-day trial, a jury convicted Rudy Mendoza on five of six counts in Indictment $2 11 Cr. 974 (CM): (a) participating in a racketeering conspiracy in violation of Title 18, United States Code, Section 1962(d) (“Count One”): (b) conspiracy to violate the narcotics laws of the United States, in particular, to distribute and possess with intent to distribute mixtures and substances containing a detectable amount of cocaine, and to distribute and possess with intent to distribute mixtures and substances containing a detectable amount of marijuana, in violation of Title 21, United States Code, Sections 812, 841(a)(1), 841(b)(1)(C), 841(b)(1)(D) and

846 (“Count Two”); (c) conspiracy to commit Hobbs Act robbery, in violation is Title 18, United States Code, Section 1951 (“Count Four’); (d) using and carrying a firearm during and in relation to acrime of violence, specifically, the conspiracy to commit Hobbs Act robbery charged in Count Four, in violation of Title 18, United States Code, Sections 924(c)(1)(A)(i) and 2 (“Count Five”); and (e) conspiracy to violate the narcotics laws of the United States, in particular. to distribute and possess with intent to distribute five kilograms or more of cocaine, in violation of Title 21, United States Code, Sections 812, 841(a)(1), 841(b)(1)(A), and 846 (“Count Six”). Mendoza was acquitted on Count Three, which charged him with using a firearm in connection with the RICO conspiracy charged in Count One.! The defendant’s underlying offense conduct principally involved narcotics trafficking and participating in an armed robbery crew. Just months after having been released from prison on March 15, 2011, the defendant resumed criminal activities in connection with the Los Vagos street gang, which operated in—among other areas in the Northeast United States—Harlem and the Bronx. Mendoza was one of the gang’s cocaine suppliers, PSR §§ 27, 30, and also part of the gangs armed robbery crew. Among other things, on September 19, 2011, the defendant was arrested as part of a robbery sting in which the defendant and others attempted to rob a person that they believed to be a narcotics dealer. At the time of the attempted robbery, the defendant possessed a loaded .380 caliber semi-automatic pistol, which he had obtained from a member of the Los Vagos gang. PSR 36.”

' The count numbers referenced above are those count numbers as presented to the jury and reflected on the judgment, Dkt. 237 (the “Judgment”). The counts listed above correspond to counts one, six, thirteen, fourteen, and fifteen in the Indictment. See Judgment, at 1. See Government Memorandum in Opposition, ECF 378 at 6-21, for a more fulsome summary of the evidence adduced at trial.

On November 5, 2014, the Court sentenced Mendoza to a total term of imprisonment of 300 months, comprising 240 months’ imprisonment on Counts One, Two, Four, and Six to run concurrently, and 60 months’ imprisonment on Count Five (the § 924(c) count) to run consecutively to Counts One, Two, Four, and Six. Dkt. 240 (“Sentencing Transcript” or “Sent. Tr.”), at 22; Dkt. 237, at 3. In connection with the sentencing, the Court adopted the Guidelines calculation in the PSR, which produced a Guidelines range of 235-293 months’ imprisonment, based on a total offense level of 34 and a criminal history category of V. PSR §§ 74, 83, 117; Sent. Tr. at 18-19, 21. By summary order dated October 29, 2018, the Second Circuit affirmed the defendant's conviction. United States v. Climico, 754 F. App’x 25 (2d Cir. 2018). On October 7, 2019, the Supreme Court granted certiorari, vacated the judgment and remanded to the Second Circuit for further consideration in light of United States v. Davis, 139 S. Ct. 2319 (2019), On February 5, 2020, the Second Circuit vacated the defendant’s conviction on Count Five (the § 924(c) count), affirmed the conviction on the remaining counts, vacated the sentence, and remanded for resentencing. Dkt. 312; United States v. Climico, 802 F. App’x 599 (2d Cir. 2020) (summary order). On October 6, 2020, the Court held a hearing and granted the defendant’s request to proceed pro se in connection with his resentencing. On October 14, 2020, in advance of resentencing, the Probation Office prepared a Supplemental Presentence Report recalculating the Guidelines range for the remaining counts of conviction to be 292 to 365 months’ imprisonment, based on an offense level of 36 and a criminal

history category of V. On November 12, 2020, the Court held a resentencing hearing, at which Mendoza appeared pro se. The Court resentenced Mendoza to 240 months’ imprisonment, five years’ supervised release, and a $400 special assessment. (Dkt. 325), The Court emphasized: “I stand by everything I said at the original sentencing about the severity of the conduct of which he stands convicted, about his criminal history, and about the reasons why I agreed with the Probation Department which at the time recommended a 240-month sentence on those counts.” (Resentencing Tr. 19:1- 5). The Second Circuit affirmed the District Court’s amended judgment, and subsequently denied Mendoza’s motion for rehearing. (Second Appeal, Dkt. 83-1, 96). The Circuit’s mandate issued on June 8, 2022. Mendoza is currently serving his sentence in a United States Penitentiary in Colorado, Florence ADMAX. His current release date is January 6, 2030. (Inmate Locator, last checked June 10,2025). On May 4, 2023, Mendoza filed a pro se petition under 28 U.S.C. § 2255 challenging the validity of his conviction. His claims include: (1) ineffective assistance of trial counsel; (2) ineffective assistance of appellate counsel; (3) deprivation of a fair trial based on the “improper joinder of two separate and completely unrelated indictments into a single trial,” and (4) the constitutionality of his conviction. (Mot. 6-21). Although the Court appointed Jeffrey Pittell from the Court’s CJA panel to represent Mendoza in connection with his motion, Mendoza—convinced that he alone would be his best advocate—has rejected any substantive assistance from Mr. Pittell.4

* The Revised Guidelines Range was higher than the original Guidelines range because, in the absence of the Section 924(c) conviction, a five-level firearm enhancement applies to the robbery conspiracy, which raised the total offense level by two levels after the grouping calculation. See SPSR, at 2. ‘In a letter to the Court, Mendoza made clear he did not — an attorney meddling in his case: “[T]his court appointed

On October 9, 2024, the Government filed a Memorandum in Opposition asking that Mendoza’s petition be dismissed as meritless. Relevant Legal Standard Under 28 U.S.C. § 2255, a habeas petitioner can obtain relief from a final judgment only if he can show “a constitutional error, a lack of jurisdiction in the sentencing court, or an error of law or fact that constitutes ‘a fundamental defect which results in a complete miscarriage of justice.”” Graziano v. United States, 83 F.3d 587, 589, 90 (2d Cir. 1996) (quoting United States Bokun, 73 F.3d 8, 12 (2d Cir. 1995)). “The reasons for narrowly limiting the relief permitted under § 2255—a respect for the finality of criminal sentences, the efficient allocation of judicial resources, and an aversion to retrying issues years after the underlying events took place—are ‘well known and basic to our adversary system of justice.”’ Bokun, 73 F.3d at 12 (quoting United States v. Addonizio, 442 U.S. 178, 184 & n.11 (1979)). Although pro se petitioners are entitled to have their pleadings liberally construed, see Green v. United States, 260 F.3d 78, 83 (2d Cir. 2001), they also still bear the burden of proving, by a preponderance of the evidence, that they are entitled to relief, see Triana v, United States, 205 F.3d 36, 40 (2d Cir. 2000). To prevail on an ineffective assistance of counsel claim, a defendant must (1) demonstrate that his “counsel's representation fell below an objective standard of reasonableness” under “prevailing professional norms”; and (2) “affirmatively prove prejudice” from counsel’s allegedly defective performance. Strickland, 466 U.S. at 688, 693; see also Hernandez vy. United States, 202

attorney who I requested that you remove so I could continue pro se status, seems to now want to remain on board, although now as a legal advisor capacity Why? I’m still trying to figure that one out. | even told him that you [the Court] had a history of being extremely conservative with government resources, meaning tax-payer dollars. For Christ sakes you consolidated my two cases into one to not over exhaust government funds, | told him. But he was very insistent, so insistent that I was left speechless. .. . So please feel free to send him elsewhere if you like, no objection from me, unless you and/or the U.S. Attorney see any 7 make use of him. I don’t.” Mendoza letter, ECF 382.

F.3d 486, 488 (2d Cir. 2000). A defendant bears a “heavy burden” under the Strickland test. United States v. Gaskin, 364 F.3d 438, 468 (2d Cir. 2004). Under the first prong of the Strickland analysis, the reviewing court “‘must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance,’ bearing in mind that ‘[t]here are countless ways to provide effective assistance in any given case’ and that ‘[e]ven the best criminal defense attorneys would not defend a particular client in the same way.’” United States v. Aguirre, 912 F.2d 555, 560 (2d Cir. 1990) (quoting Strickland. 466 U.S. at 689). The defendant’s burden is to show “‘that counsel made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment.’” Harrington v. Richter, 562 U.S. 86, 104 (2011) (quoting Strickland, 466 U.S. at 687). A defendant cannot prevail on a claim of ineffective assistance merely because he believes that his counsel’s strategy was inadequate. United States v. Sanchez, 790 F.2d 245, 253 (2d Cir. 1986). Indeed, “[a]ctions and/or omissions taken by counsel for strategic purposes generally do not constitute ineffective counsel.” Gibbons v. Savage, 555 F.3d 112, 122 (2d Cir. 2009) (defendant failed to show that “his counsel’s decisions were anything other than strategic” or that “different choices would have had any likelihood of altering the outcome of the trial”). Counsel’s decisions “to offer or stipulate to certain evidence” and as to “when to object and on what grounds are primarily matters of trial strategy and tactics, and thus are virtually unchallengeable absent exceptional grounds for doing so.” United States v, Cohen, 427 F.3d 164, 170-71 (2d Cir. 2005) (internal quotations and citations omitted). With respect to counsel’s decision to present or forgo a particular defense, “[a] lawyer’s decision not to pursue a defense does not constitute deficient performance if, as is typically the case, the lawyer has a reasonable justification for the decision.” Dupont v. United States, 224 F. App’x 80, 82 (2d Cir. 2007) (quoting

Greiner v. Wells, 417 F.3d 305, 319 (2d Cir. 2005)). Indeed, “[c]ounsel’s election to forgo an unsupported argument plainly falls into th[e] category” of “[a]ctions or omissions by counsel that might be considered sound trial strategy”— not ineffective assistance. United States v. Best, □□□ F.3d 192, 201 (2d Cir. 2000) (internal quotation marks omitted). “Similarly, counsel’s decision as to ‘whether to call specific witnesses—even ones that might offer exculpatory evidence—is ordinarily not viewed as a lapse in professional representation.” Jd. (quoting United States v. Schmidt, 105 F.3d 82, 90 (2d Cir. 1997)). To establish prejudice under the second prong of Strickland, the defendant “must show that there is a reasonable probability that. but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Jd, Thus, a defendant cannot establish prejudice by showing merely that counsel’s errors had “some conceivable effect” on the result, for “not every error that conceivably could have influenced the outcome undermines the reliability of the result of the proceeding.” /d. at 693. “[A]n analysis focusing solely on mere outcome determination, without attention to whether the result of the proceeding was fundamentally unfair or unreliable, is defective.” Lockhart v. Fretwell, 506 U.S. 364, 369 (1993). Thus, “the prejudice component of the Strickland test ... focuses on the question whether counsel’s deficient performance renders the result of the trial unreliable or the proceeding fundamentally unfair.” /d. at 372 (internal quotations omitted), Only where both prongs of the Strickland test are satisfied can it be concluded that “counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment.” Strickland, 466 U.S. at 687. When considering these two prongs, the Supreme Court has provided commonsense instructions, explaining that “[t]he object of an ineffectiveness claim is not to grade

counsel’s performance. If it is easier to dispose of an ineffectiveness claim on the ground of lack of sufficient prejudice . . . that course should be followed.” Strickland. 466 U.S. at 697. Accordingly, the Second Circuit has often rejected ineffectiveness claims by determining that, in view of the strength of the prosecution’s case, the defendant is unable to establish prejudice. See, e.g., Rosario v. Ercole, 601 F.3d 118, 136 (2d Cir. 2010) (Straub, J., concurring) (“Strickland makes clear that ‘a verdict or conclusion only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support.’” (quoting Strickland, 466 U.S. at 696)). Mendoza’s Ineffective Assistance Claims as to His Trial Counsel James Cohen Attorney James Cohen was assigned to represent Mendoza on January 23, 2013. (Dkt. 90). Cohen was an experience criminal defense lawyer. He also served on the faculty of Fordham Law School, where he and another experienced attorney, Michael Martin, ran the Federal Litigation Clinic, which operated under the day to day supervision of a recent Fordham graduate (and a Dean’s Fellow), Giuliana Graham. Students from the clinic participated in the preparation and trial of cases assigned to Cohen or Martin. Cohen served as defense counsel in Mendoza’s case and was the person who was on his feet and who spoke throughout the trial; prep work was performed by Graham and various student interns, acting at all times under Cohen's supervision. (Amirian Aff. § 2; Belgrave Decl. § 3; Hendrix. Aff. at 1). Ordinarily Cohen would have submitted an affidavit concerning his participation in the case and responding to the various allegations made by Mendoza. Unfortunately, Cohen has retired from the practice of law and is ill; his family members have advised the Court that he is medically unable to submit an affidavit. See Government Letters, the first dated 6/1 8/24, ECF 372 (Redacted), and the second dated 8/5/24, ECF 375 (Redacted).

It is not unheard of for defense counsel to die or become ill in the many years that pass between the conduct of a criminal trial and the filing of a motion pursuant to § 2255; when that occurs, the court is forced to rely on its own recollections about the trial, together with any information that can be obtained from others who may have worked on the case. In this case, we are fortunate in that regard. Because of the participation of the Fordham Law School clinic interns, quite a few people are around who worked on the case and who can testify about Cohen’s actions prior to and during Mendoza’s trial. Of the seven members of the clinic staff who worked, in some capacity, on Mendoza’s case—Giuliana Graham, Justin Amirian, Kharis Belgrave, Richard Hendrix, Joseph Jacobs, Anna Makatche, and Jenna Mintzer (Dkt. 154)—six have submitted either an affidavit or declaration that responds to Mendoza’s ineffective assistance of counsel claims against Cohen.* Most importantly, the Court has received affidavits from Graham, Belgrave, and Amirian who were assigned to Mendoza’s core defense team and worked under the direct supervision of lead attorney Cohen. In addition, Mr. Martin, who still runs the clinic, has submitted a declaration attesting to his observation of the role Cohen played in preparing and trying Mendoza’s case. It is on the basis of this record, together with my own recollections of Cohen’s performance, that the court is able to decide Mendoza’s motion.® Mendoza raises numerous complaints about Cohen’s trial performance, but in every instance, he fails to demonstrate that Cohen acted in a constitutionally deficient manner. The record in this case wholly contradicts Mendoza’s claim that Cohen provided him with ineffective assistance of counsel, under the Strickland standard.

> Justin Amirian Affidavit (Dkt. 362); Kharis Belgrave Declaration (Dkt. 374); Giuliana Graham Affidavit (Dkt. 363); Richard Hendrix Affidavit (Dkt. 364); Joseph Jacobs Affidavit (Dkt. 365); Anna Makatche Affidavit (Dkt. 366). The Court has not received an affidavit or declaration from Jenna Mintzer. ° | am grateful to Attorney Martin for his diligence in a record for use on this motion.

Cohen's Involvement and Preparation for Trial Mendoza’s overarching argument is that Cohen was not sufficiently involved in Mendoza’s case, and that Cohen made “a school project out of my life and freedom.” (Mot. 14). That claim is convincingly refuted by Graham, who describes Cohen as retaining “unequivocal control of Mr. Mendoza’s defense at all times,” as being “clearly familiar with Mr. Mendoza’s case,” as “devot[ing] all of his time and energy to Mr. Mendoza’s defense,” and as being “fully engaged in the process of advising the client, preparing the case, and conducting the trial.” (Graham Aff. §§ 9, 11, 14, 22). And Martin recalls: “Mr. Cohen’s strict attention to Mr. Mendoza’s case;”’ that Cohen “decided to stay at a hotel near the Law School during the trial so as to be available for trial preparation with the team and fresh for each day’s trial session; and that Cohen attended “many” trial preparation meetings. (Martin Decl. §§ 12, 14, 18). Mendoza alleges that, “Towards the end of trial[,] Ms. Giuliana Graham informed [him] that Mr. Cohen suffered from some kind of neurological disorder for which he had just underwent surgery,” and speculates that it is “possible that [Cohen’s] ailments prevented him from doing his due diligence.” (Mot. 9). However, Graham adamantly “den[ies]” telling Mendoza any such thing. She agrees that she has “no idea whether Prof. Cohen had been diagnosed with a ‘neurological disorder,” but is “certain that [she] would remember any surgery that could have impacted Prof. Cohen’s professional performance,” and has “no such memory.” (Graham Aff. § 21). Martin— with whom Cohen ran the Law Clinic—likewise “was not aware Mr. Cohen suffered from any medical condition that would have prevented him from performing at trial.” (Martin Decl. 17). Mendoza also attacks Cohen’s age and refers to Cohen as his “old, idiot lawyer.” (Mot.

But Hendrix says that she “do[es] not know what Mendoza is referring to as Cohen’s ‘advanced age and medical conditions,” and Graham stated that “[n]ot once, at any point during [her] service as a student intern or post-graduate fellow at the [Law] Clinic did [she] ever believe that Prof. Cohen’s age negatively impacted his performance as an attorney.” (Hendrix Aff. at 4: Graham Aff. q 20). Mendoza claims that Cohen failed to prepare sufficiently for trial. In particular, Mendoza alleges that Cohen rarely visited him in jail and did so only to “try to talk [Mendoza] into pleading guilty in exchange for a mandatory 15[-]year sentence”;* delegated “every aspect of trial preparations” to Graham, Belgrave, and Amirian; and did not discuss trial strategy with Mendoza or allow him to explain his innocence. (Mot. 6-7). Graham and others on the defense team agree that Cohen spent a significant amount of time talking to Mendoza about the Government’s plea offers. (Graham Aff. § 13; see also Martin Decl. {| 9-10). However, after Mendoza rejected the option to plead (against the advice of counsel) and elected to proceed to trial, Cohen’s focus turned to trial strategy, which “Cohen alone determined.” (Graham Aff. § 15). Hendrix describes Mendoza’s allegations that Cohen never met with him to prepare for trial and delegated to his law students “every aspect of trial preparation due to what they said was [Cohen’s] advanced age and

Cohen was in his early sixties at the time— the same age as the Court at the time of trial. * Mendoza claims that the Government's plea offer carried a 15-year mandatory minimum sentence. However, the credible evidence suggests that at least one plea offer included only a 5-year mandatory minimum sentence. (See Govt. Ex. A at 2). On March 21, 2013, by letter addressed and emailed to Cohen and Martin, the Government extended a plea offer to Mendoza. (Ex. A). The plea offer was not a cooperation agreement. (/d.). Under the terms of the plea agreement, Mendoza could have pled guilty to Counts One, Nine, and Thirteen of the S2 Indictment, which carried a mandatory minimum term of imprisonment of five years, and a maximum term of imprisonment of life. (/d. at 1). The stipulated Guidelines range set forth in the plea agreement was 160 to 185 months’ imprisonment, but the mandatory minimum term was 60 months’ imprisonment. (/d. at 5). To the extent Hendrix (one of the student interns assisting Cohen) recalls a plea offer including a “15-year sentence and testimony by Mendoza against other defendants in the RICO part of his case” (Hendrix Aff. at 2), Hendrix may be equating the offers mandatory minimum with the Sentencing Guidelines, which range was from 13-15 years. 11

medical conditions,” (Mot. 6-7), as “untrue.” (Hendrix Aff. at 4). Cohen tasked his interns with, inter alia: “brainstorming questions to be asked on cross-examination;” meeting with Mendoza “to review discovery material and record Mr. Mendoza’s insight into the facts of the case” (Graham Aff. 11, 16); working on “team coordination and logistics; research[ing] and reviewing] expert- witness reports; prepar[ing] for a competency hearing; review[ing] and summar[izing] .. . discovery and 3500 material; assisting . . . with voir dire:” and other projects (Belgrave Decl. □ 4). According to Graham, whenever the Legal Intern Team met Mendoza without Cohen, Cohen had set the agendas for these meetings in advance, and Graham always “took careful, copious notes at every meeting” and subsequently “shared those notes and impressions” with Cohen. (Graham Aff. 11). “Any suggestion Mr. Mendoza made [to the Legal Intern Team] as to a potential witness, line of inquiry or defense . . . was relayed to Prof. Cohen near verbatim; and thereafter evaluated by him.” (/d.). Cohen “wrote his own opening and closing statements; decided which questions to ask the witnesses; and crafted his own theory of the defense.” (Graham Aff. 7 13). Moreover, Graham says that Cohen personally attended several of the defense team meetings with Mendoza, during which “he took the lead in counseling Mr. Mendoza regarding the key decisions that belonged to the client, i.e., whether to proceed to trial and whether to testify.” (Graham Aff. § 12; see also Amirian Aff. § 4; Belgrave Decl. § 7; Hendrix Aff. at 3-4). Outside of the presence of the Legal Intern Team, Cohen “had separate meetings and conversations with Mendoza,” (Hendrix Aff. at 2), and Cohen personally met with Mendoza every day of the nine-day trial, (Belgrave Decl. § 7; Hendrix Aff. at 3). Cohen's conduct, as described by the people who worked with him on a daily basis, falls well within the range of adequate trial preparation. If anything, his preparation was enhanced, not hindered, by the assistance of law students. See, e.g., United States v. Romero, No. 91 CR. 586 12

(RPP), 1993 WL 485677, at *10 (S.D.N.Y. Nov. 22, 1993), aff'd, 54 F.3d 56 (2d Cir. 1995) (finding no ineffective assistance where counsel had the assistance of law students who, while helpful in researching legal issues, drafting court papers, and relaying communications and documents, were under “close supervision” by the attorney and “were not entrusted with decisions about what motions to file or other strategic issues”). The Plea Offer According to Graham, Martin, and the interns, Cohen provided Mendoza with “extensive counsel” about the Government’s plea offer before trial commenced. (Graham Aff. £ 13). During these meetings, Cohen advised Mendoza of the risks of proceeding to trial (see id.) and ultimately urged Mendoza “to take a plea bargain for his case, given the weight of evidence against him.” (Amirian Aff. { 4). According to Martin, he and Cohen successfully advocated with the Government to re-extend a plea offer that had expired, one that provided for “a 5-year mandatory minimum with a lower-end Sentencing Guideline of 160 months.” The terms of this plea offer were accurately conveyed to Mendoza. (Martin Decl. §§ 9-10). Graham described the amount of time Cohen spent counseling Mendoza regarding the plea offer as “commensurate with the risks Mr. Mendoza faced by proceeding to trial.” (Graham Aff. § 13). In his reply papers, Mendoza rejects the notion that the Government ever offered him a plea deal with anything less than a 15-years mandatory minimum. As Mendoza tells it, Andrew Patel. the attorney assigned to him at arraignment, was “relentless” in his effort to get Mendoza to cooperate with the government— “so incredibly relentless that after several failed attempts where I constantly refused him he even resolved to trickery, disguising a court appearance for a meeting with the Government.” Reply at 4. “I made it clear to Mr. Patel that there will be no cooperation with the Government on my part, that I had every intention of going to trial on the Hobbs Act case 13

but was willing to make a deal on the RICO case.” Jd. “And next thing I know both cases get consolidated into one, Mr. Patel makes me this 15 year mandatory minimum offer, I reject it, then he’s gone,” replaced by James Cohen and his pupils.” Jd. Mendoza says that, once Mr. Cohen was on the case, he and his students tried to “talk me into to pleading to the same 15-years mandatory minimum plea deal mentioned by Mr. Patel.” Jd. at 5. As for Martin’s testimony that he and Cohen successfully advocated with the Government for “a 5-year mandatory minimum with a lower-end Sentencing Guideline of 160 months,” and that the terms of this plea offer were accurately conveyed to Mendoza” (Martin Decl. 9-10), “I [Mendoza] can see a scenario where Mr. Martin is the one trying to pull a fast one on all of us, including the court.” Reply at 6. As for the letter memorializing the 5-year mandatory-minimum plea offer, which the Government sent to Messrs. Cohen and Martin via email on March 21, 2013 (Govt Exh. A), Mendoza speculates that it is a fraud: “But with all due respect, let’s not pretend like the copy of the plea agreement sent to me by the Government has no chance at all of being a fraud.” Reply at 6. Mendoza’s fallback argument is that, even if the Government tendered the lower 5-year mandatory plea offer to Cohen and Martin, neither they nor their students ever communicated that offer to him: “Maybe, like the rest of my case, my defense team completely screwed up the plea bargain process as well. . . . | can totally see a scenario where Mr. Cohen communicated that same

Patel was “gone” because Mendoza wanted Patel gone. See Transcript of Status Conference held on January 23, 2013. Indeed, Mendoza would eventually want all the attorneys assigned to represent him at different stages of his case to be gone--Andrew Patel, James Cohen, Louis Freeman, Glen Garber, Jeremy Schneider, Jeffrey Pittell, every one a distinguished member of the Court’s Criminal Justice Act panel. At some point, Mendoza has found fault with each of these attorneys and, more fundamentally, was suspicious of their allegiance. When the Court attempted to assign counsel to assist Mendoza with the present motion, Mendoza made clear that: “I could care less so long as he [Pittell] knows that he is not to interfere or in any way sabotage my response. So please feel free to send him elsewhere if you like, no objection from me. Unless you and/or the U.S. Attorney’s see any way to make use of him, I don’t.” Mendoza Letter ECF 382 at 2. 14

wrong plea offer to me. That would be classic James Cohen.” Jd. Mendoza insists that it is simply not plausible that he would have rejected such “a sweet plea deal. .. . No reasonable person would have. /d. No matter how innocent or guilty that person is, either his defense team mishandled the plea bargain process, or he [Mendoza] is totally mentally unfit to stand trial.” Jd. Mendoza was not “totally unfit” to stand trial. In fact, not long after the Government sent Cohen and Martin the March 13, 2013 letter with its 5-year mandatory minimum offer, Cohen asked the Court to order a psychiatric examination for his client, to determine whether he was fit to stand trial (quite possibly because, as Mendoza now agrees, no reasonable person would have rejected such a deal). I granted Mr. Cohen’s request. Both the psychiatrist retained by the defense and psychiatrist secured by the Court ultimately agreed that Mendoza was fit to proceed to trial or enter a plea of guilty. As this Court has learned in my 30 years on the bench, being mentally fit to stand trial does not mean a defendant will necessarily make “reasonable” decisions about taking a plea. It appears that one of the sticking points that prevented Mendoza from pleading guilty was that he was adamant—as he told Patel— about “going to trial on the Hobbs Act case.” Reply at 4. But once the RICO and robbery cases were folded into a single indictment, going to trial on the robbery case alone was not an option. Mendoza’s perjured trial testimony supports the notion that he was hellbent on fighting the robbery charges: he admitted to selling cocaine (no more than “2 ounces a week” (Tr. 1067:6-16)). but told the jury an incredible story about getting into his cousin’s car not knowing that his cousin and the others in the car were on the way to robbing a drug dealer (Tr. 1077-89), Mendoza appears to have had other reasons for not wanting to plead guilty. During a hearing to determine Mendoza’s mental competency to stand trial, the Court got a glimpse into the plea 15

negotiation between the defense and Government— the type of negotiation into which the Court seldom has insight, since it is not a party to, and is forbidden from getting involved in plea negotiations. MR. COHEN: He is coping in a significantly improved way with the PTSD symptoms, and he understands much better that going to trial for the purposes of punishing the agents for shooting at him when he was arrested is not the wisest thing that he should do. And we've already begun talking about disposing of this case in a different way, although we need the government's cooperation for that. Transcript from Competency Hearing, ECF 169 at 7. 2 fe 3c 2k fe fe afc ac ake oe akc ake ok ok MR. COHEN: Judge, I will briefly update you with respect to the conversation I had with both Mr. Mendoza and the government attorneys. First of all, Mr. Mendoza appears to me to be the best that I have seen him in many months, and we engaged ina very short discussion about what our options might be beyond going forward with this hearing. So, | engaged in, again, a very short conversation with the assistants to determine-- which I was unsuccessful in doing, but to determine to what extent, if any, a disposition might be offered. The government informed me that it was, I will say unlikely, but I think they used stronger terms, that the same disposition that was offered before would be reoffered. THE COURT: That's usually the case. MR. COHEN: It is usually the case, your Honor, but it's not usually the case that the defendant has suffered from some mental illness. Id. at 2, Contrary to Mendoza’s characterization of Cohen as a passive advocate who delegated his representation of Mendoza to his law students while he suffered from some “neurological condition” that impaired his ability, Cohen was a zealous advocate, intent on getting the best possible outcome for his client. That was certainly the observation of the Court. Cohen, no doubt cognizant of Rule 11, deftly attempted in his colloquy with the Court to leverage the Court’s 16

authority to persuade the prosecutor to put a better plea deal (presumably the 5-year mandatory minimum deal) back on the table. Unfortunately for Cohen and Mendoza, they were negotiating with a tenacious and unyielding prosecutor; once Mendoza tarried on the “sweet offer,” it was withdrawn. The weight of the evidence suggests that the reason Mendoza never pleaded guilty to the terms set forth in the March 13 letter was his own stubbornness and poor judgment— not Cohen’s failure to secure a favorable deal, or because Cohen failed to communicate the good offer once made. When Mendoza elected to reject any subsequent plea offer and proceed to trial, Cohen and his team “doubled [their] efforts towards trial preparation.” (Amirian Aff. 4 5). The Court rejects Mendoza’s claim of ineffective assistance of counsel in connection with the plea negotiations in his case. Cross-Examination of Government Witnesses Mendoza’s complaints about Cohen’s cross-examination decisions are belied by the record. For instance, Mendoza alleges that Cohen “failed to challenge, object, or even address the court about the likely impact” of Krochmal’s expert testimony and translated transcripts of certain audio recordings. (Mot. 15). But Cohen effectively cross-examined Krochmal: Cohen got Krochmal to admit that his familiarity with Mexican Spanish stemmed from a two-week trip to Mexico in the 1980s; that the recordings he translated contained overlapping voices and external noises, which added to the difficulty of transcription; and that certain statements could be translated in multiple, correct ways. (See Tr. 476-477 (Mexico), 480-482 (overlapping voices, outside noise, quality), 482-487 (slang/local dialect), and 494-498 (multiple meanings)). And in his closing statement, Cohen reminded the jury that Krochmal himself found an error in his translation while on the stand, urging the jury to “think carefully about the reliability of these transcripts in context of being Ly

enough reliable credible evidence to prove Mr. Mendoza guilty beyond a reasonable doubt.” (Tr. 1010:8-1011:6). Mendoza also takes issue with Cohen’s cross-examination of Climico and Martinez. As to Climico, Mendoza alleges that Climico “lied under oath and Mr. Cohen failed to pursue an effective line of questioning that would have impeached him.” (Mot. 10). In particular, Mendoza argues that the firearm Climico testified that he loaned Mendoza by way of Quintero was inoperable and contends that Climico “was either lying on the stand, under oath about that gun belonging to him and the gang, or he is by far the stupidest gang member on the face of the earth for getting duped into buying an inoperable gun.” (Mot. 13). But Cohen would have known that cross- examining Climico about the inoperability of the firearm had limited value, since firearm operability is irrelevant to Section 924(c). The Court explained as much to the jury: “In considering the specific element of whether the defendant used or carried or possessed a firearm, it does not matter whether the firearm was loaded or operable at the time of the crime.” (Tr. 1280:18-21). Instead, Cohen cross-examined Climico on a variety of topics bearing on his credibility, including his selective use of a Spanish interpreter when he could understand English, the benefits he expected to receive from his cooperation agreement, and his past violence and drug use history. (See Tr. 805-858). As for Martinez, Mendoza complains that Cohen did not cross-examine Martinez about □□□ fact that Mendoza was purportedly on parole in Chicago in or around the time Martinez claimed to have seen Mendoza at a gang meeting in East Harlem. (Mot. 10). Mendoza claims to have given Cohen “coldblooded evidence of Martinez’s false testimony” in the form of his “discharge papers and . . . rap sheet, which showed that from January 2, 2005 to January 5, 2006[, Mendoza] had been serving a parole violation . . . in upstate New York, and within three days of [his] release[, he] 18

began reporting at the parole office in Chicago until the end of 2007.” (Mot. 10-11). Belgrave recalls that Mendoza told her that he lived in Chicago for a little over a year between 2006 and 2007; Belgrave made no mention of receiving documents from Mendoza to corroborate his statements. (Belgrave Decl. § 10). Contrary to Mendoza’s recollection, Cohen did question Martinez about the date of this meeting, causing Martinez to clarify that the meeting took place in approximately late 2006 going into early 2007. (Tr. 198-199). Cohen did not confront Martinez with the parole paperwork that Mendoza purportedly gave to his defense team. Cohen did, however, press Martinez on his limited association with Mendoza, pointing out that Mendoza’s contact information was not saved in Martinez’s phone, and that there was an approximate five- year time period during which Martinez did not see Mendoza. (See Tr. 156, 198). Cohen emphasized this to the jury during his summation: “If Mendoza is really a member of that gang, is it possible that Martinez sees him once in five years?” (Tr. 1018:2-3). In short, Cohen’s decisions during cross-examination were well within the range of options for constitutionally effective representation. United States v. Luciano, 158 F.3d 655, 660 (2d Cir. 1998) (“the conduct of examination and cross-examination is entrusted tothe judgment of the lawyer”); see also Moreno-Godoy v. United States, No. 13 Civ. 2383 (JSR) (GWG), 2014 WL 1088300, at *73 (S.D.N.Y. Mar. 20, 2014) (noting that “decisions about whether to engage in cross-examination, and if so to what extent and in what manner, are strategic in nature and generally will not support an ineffective assistance claim” and holding that courts “should not second-guess such decisions unless there is no strategic or tactical justification for the course taken”) (internal quotations and alterations omitted), report and recommendation adopted at 2014 U.S. Dist. LEXIS 106715 (S.D.N.Y. Aug. 4, 2014).

Cohen’s Decision to Forgo Defense Case Mendoza complains about numerous decisions Cohen made pertaining to a potential defense case, including Cohen’s purported failure to subpoena and call various alibi and expert witnesses, and pursue an entrapment defense. Mendoza claims that Cohen failed to subpoena Yasmin Osuna, the mother of Climico’s son, as an “alibi witness[].” (Mot. 12). He claims that Osuna would have testified that Climico elected to cooperate with the Government “for payback against [Mendoza] over a previous intimate relationship that [Mendoza] had with [Osuna] that also sparked a rumor around the neighborhood that [Climico’s] son was actually [Mendoza’s].” (Mot. 12-13). Mendoza presents neither an affidavit from Osuna nor any other form of evidence tending to show that Osuna could corroborate his version of events. See Krasniqi v. United States, 195 F. Supp. 3d 621, 634 (S.D.N.Y. 2016) (“Without affidavits from either of these witnesses, [defendants’ □ self-serving, uncorroborated and improbable assertions are insufficient to establish that their attorneys provided ineffective assistance.”). But even assuming Osuna would have testified to this—something that is not at all clear—her testimony would not have given Mendoza an alibi. See ALIBI, Black’s Law Dictionary (12th ed. 2024) (“A defense based on the physical impossibility of a defendant’s guilt by placing the defendant in a location other than the scene of the crime at the relevant time” or “The quality, state, or condition of having been elsewhere when an offense was committed.”). Rather, the proffered testimony goes to impeaching Climico. But Cohen would have been barred from calling Osuna for this purpose. See Fed. R. Evid. 608(b). Moreover, Osuna was Mendoza’s co-defendant. She pled guilty in this case on October 19, 2012 and admitted under oath her involvement in drug trafficking for the Vagos Gang. (Dkt. 71 at 23:9- 24:18). Osuna would have been subject to devastating cross given her association with the Gang 20

and the subject matter about which Mendoza claims she would have testified. Cohen cannot be faulted for not calling her as a witness; he might well have been faulted if he had called her. Mendoza complains about Cohen’s failure to call John Vallejo, Gaspar Vallejo, Jr.. and Gaspar Vallejo, Sr. as witnesses, claiming—again without presenting any affidavits or other evidence—that they were “willing to testify that the only reason why [Mendoza] was at their auto repair shop the morning [of] the robbery . . . was to pay John money [Mendoza] owed him and borrow his car to go see a friend in Queens,” and that “all three guns found in the car at the arrest scene belonged to them.” (Mot. 11, 13). In the first instance, counsel for Mr. Mendoza’s co- defendants would have almost certainly advised their clients against self-incrimination and/or perjury. Moreover, such testimony—to the extent it was actually available—would not have been credible: the evidence at trial made clear that Mendoza’s decision to participate in the robbery with the Vallejos was not a spur of the moment decision. Rather, in the weeks prior to the day of the robbery attempt, Mendoza was communicating with the Vallejos, telling his associates that he was looking for an opportunity to participate in a robbery. (See GX 901-A, 901-B (telephone connection charts); Tr. 1171, GX 7025 (Mendoza commenting to an associate on his Facebook page that he was “chillin, trying to find a come up [robbery]. This job shit ain’t doing it for me.”); Tr. 900, GX 210-HT (Mendoza writing on Facebook: “I need big money[.] I’m tired of this small-time shit.”)). As Graham observes, “It would certainly have been reasonable for Prof. Cohen to conclude that there were more effective and less risky avenues for making use of the information that Mr. Mendoza hoped to put before the jury through the proposed witnesses.” (Graham Aff. § 17); see, e.g., Johnson v. Mann, No. 92 Civ. 1909 (TPO), 1993 WL 127954, at *1 (S.D.N.Y. Apr. 20, 1993) (“As to the alibi issue, counsel made the strategic decision, in the face of a strong prosecution case, to attack the credibility of as who had specifically identified petitioner

rather than to rely on the inherently suspect testimony of family members.”). Mendoza says that Cohen “failed to call his own expert voice analyst to distinguish an exemplar of [Mendoza’s] voice to the voices alleg[ed] by the Government.” (Mot. 8). Cohen’s decision not to call an expert voice analyst was “a tactical decision of the sort engaged in by defense attorneys in almost every trial.” Romero, 1993 WL 485677, at *7. Instead, Cohen decided to attack the Government's witness who identified Mendoza as the speaker in many of the recordings— Checo—presumably because in Cohen’s professional judgment, he believed this to be the more effective path. Indeed, during cross-examination, Cohen effectively pointed out that the voices in the recordings are hard to make out, many parts are unintelligible, and voices closer to the recording device are easier to hear than voices further away. (See Tr. 413-414). Cohen also questioned Checo about his inability to distinguish the voices of Mendoza and another individual, José Ramos. (See Tr. 420-421). Cohen reminded the jury about the unreliability of these recordings during his summation. (See Tr. 1012:10-17). Mendoza complains that Cohen failed to pursue an entrapment defense. (Mot. 8). Mendoza himself acknowledges, though, that, “on several occasions I told Mr. Cohen and his students that although I was not entrapped by SA Checo into committing a robbery directly, I was still entrapped by my cousin John when he lured me into his car without telling me what he was up to.” (Mot. 8). What Mendoza describes is not entrapment. To establish a defense of entrapment, the defendant must show that the government “induced the defendant to commit the crime for which he is accused and that, absent government inducement, the defendant was not predisposed to commit the crime.” United States v. Capozziello, 832 F. Supp. 532, 534 (D. Conn. 1993), aff'd, 28 F.3d 102 (2d Cir. 1994) (citing Mathews v. United States, 485 U.S. 58, 63 (1988)). And an entrapment defense carries substantial risks because it requires that the admit the intent to commit the crime

charged. Id. (citing United States v. Collins, 957 F.2d 72, 76 (2d Cir. 1992). It also opens the defendant up to other evidence demonstrating the defendant’s “propensity to engage in criminal conduct. /d. (citing United States v. Tutino, 883 F.2d 1125, 1139 (2d Cir. 1989)). Given these risks, Mendoza’s continuing insistence on his innocence, and Mendoza’s own admission that he told Cohen he “was not entrapped by SA Checo,” Cohen’s decision to not pursue this argument was objectively reasonable. See, e.g., Borrero v. United States, No. 13 Cr. 0058 (KBF), 2018 WL 358921, at *5 (S.D.N.Y. Jan. 10, 2018) (rejecting ineffective assistance claim based on failure to advise on entrapment defense in Hobbes Act robbery case). In short, Mendoza fails to raise any factual issues about Cohen’s decisions not to investigate or pursue an alibi defense, investigate an entrapment defense, or call certain witnesses, resulted from anything other than reasoned, professional judgment— especially given the futility of presenting such evidence. Mendoza Suffered No Prejudice Assuming arguendo that Cohen’s performance was deficient—and I emphasize that it was not—Mendoza must still “show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694, The Government’s trial evidence was overwhelming. The Government presented recorded phone calls between Mendoza and Climico, the leader of the Gang, during which Mendoza offered and ask for guns, discussed providing cocaine to Climico and other members of the gang, made plans to extort prostitutes, and discussed prior criminal acts he and others—including both Gang and Robbery Crew members—committed. (See, e.g., GX 1002-T, 1006-T, 1115-T, 1116-T, 1117- T, 1118-T, 1119-T, 1120-T, 1121-T, 1122-T (guns); GX 1003-T, 1004-T, 1005- 23

T, 1006-T (extortion/robbery); GX 1101-T, 1102-T, 1103-T, 1005-T, 1106-T, 1007-T, 1108-T, and others (narcotics)). The Government also presented a body wire recording from Checo of the Robbery Crew—including Mendoza—on the day of the robbery. As this Court observed at sentencing, “Mr. Mendoza’s own mouth on a tape recording condemned him.” (Sentencing Tr. 1-12). In addition to the recordings, the Government presented the testimony of several Vagos Gang members, who directly engaged in Vagos-related crimes with Mendoza. Indeed, on appeal. the Second Circuit characterized the Government’s evidence at trial as “more than sufficient to Support the jury’s finding that Mendoza agreed to (and did) participate in the Vagos Gang racketeering conspiracy during the charged time frame,” and noted that “Mendoza appears to acknowledge that the evidence was sufficient to prove his intent to participate in a robbery.” (First Appeal, Dkt. 151-1, at 4-5, 5 n.3). Mendoza has not demonstrated and cannot demonstrate that, had Cohen taken any of the actions Mendoza wishes he had, the outcome of the trial would have been different. See, e.g., Steele v. United States, No. 04 Civ. 6918 (AJP), No. 02 Cr. 629 (JSR), 2005 WL 704868, at *10 (S.D.N.Y. Mar. 29, 2005) (denying ineffective assistance claim where defendant “cannot show that had defense counsel spent more than three and one half hours with him, the outcome at trial would have been different”); Rosario, 2002 WL 31852827, at *31 (noting that in the face of overwhelming evidence, including the defendant’s confession, it was “highly unlikely that additional communication between Rosario and his counsel would have changed the trial outcome”), Additionally, Cohen’s trial tactics resulted in Mendoza being acquitted on Count Nine, one of the two Section 924(c) counts. Cohen’s ability to secure an acquittal for his client in the face of the overwhelming evidence presented at trial speaks to his effectiveness and demonstrates that Mendoza was not prejudiced by his representation. 24

Mendoza’s Ineffective Assistance Claims Regarding His Appellate Counsel Mendoza also accuses his appellate counsel, Glen Garber, of ineffective assistance, for failing to raise certain arguments on appeal. Garber says that he appealed the “issues [that] had the greatest likelihood of appellate success,” (Garber Aff. § 18) and argued the following on appeal: (1) the evidence adduced at trial was insufficient to establish Mendoza’s knowing participation in a RICO conspiracy, (2) the Court’s jury instruction regarding the “pattern of racketeering activity” element of RICO conspiracy was plain error, (3) trial counsel rendered ineffective assistance by not objecting to the erroneous jury instructions regarding the RICO conspiracy, (4) the evidence adduced at trial was insufficient to establish that Mendoza knew that the robbery conspiracy’s objective was to steal cocaine. (5) the cumulative errors at trial resulted in a deprivation of Mendoza’s rights to due process and to a fair trial, and (6) the Court’s application of a two-point enhancement for obstruction of justice was clearly erroneous. (First Appeal, Dkt. 38). First, Mendoza argues that Garber was ineffective for failing to raise on direct appeal the ineffective assistance of counsel provided by Cohen at trial. (Mot. 16). While Garber did not raise all of the ineffective assistance claims against Cohen that Mendoza raises in his present motion, Garber raised the one claim of ineffective assistance as to Cohen— the one Garber “believed had the most merit” (Garber Aff. § 22). This was Cohen’s failure to challenge the racketeering acts jury instructions, (First Appeal, Dkt. 151-1 at 7). However, the Second Circuit found that the “ineffectiveness claim fails because . . . there was no error.” (First Appeal, Dkt. 151-1 at 8). That Garber did not raise other ineffectiveness claims on appeal because he believed them to be less meritorious than the one the Circuit denied does not constitute ineffective assistance of counsel on

his part. See Smith v. Murray, 477 U.S. 527, 536 (1986) (“This process of winnowing out weaker arguments on appeal and focusing on those more likely to prevail, far from being evidence of incompetence, is the hallmark of effective appellate advocacy.”); Mayo, 13 F.3d at 533 (“[I]t is not sufficient for the habeas petitioner to show merely that counsel omitted a nonfrivolous argument, for counsel does not have a duty to advance every nonfrivolous argument that could be made.”); Felder v. United States, No. 20 Civ. 7531 (VEC), 2021 WL 3537164, at *4 (S.D.N.Y. Aug. 10, 2021). And as discussed above, Mendoza’s suggestion of ineffectiveness are all lacking in merit. Second, Mendoza contends that Garber failed to raise on appeal that the jury instructions should have stated that the jury could find Mendoza guilty of a lesser-included drug offense. (Mot. 16-17). Contrary to Mendoza’s assertion, though, the jurors were told that they could find Mendoza guilty of a lesser-included drug offense on Count Six of the trial indictment (Count Fifteen of the S2 Indictment). (See Tr. 1273:14—1274:14). Since there was no error here, there was nothing for Garber to appeal. Third, Mendoza argues that Garber was ineffective for failing to raise on appeal that the jury instructions should have included a multiple conspiracies charge. Multiple conspiracy instructions arise where the evidence suggests that a single charged conspiracy is actually multiple conspiracies. See, e.g., United States v. Berger, 224 F.3d 107, 114 (2d Cir. 2000) (finding the jury was properly instructed where the court “instructed the jurors that, if they found multiple conspiracies, they had to acquit the defendants unless they found unanimously and beyond a reasonable doubt that one of those conspiracies is the single conspiracy charged[.]””); United States v. Eppolito, 543 F.3d 25, 47 (2d Cir. 2008) (“In order to prove a single conspiracy, rather than multiple conspiracies, the government must show that each alleged member agreed to participate in what he knew to be a collective venture directed toward a common goal.”). But in Mendoza’s 26

case, multiple conspiracies were in fact charged: the racketeering conspiracy in Count One, the narcotics conspiracies in Count Six and Count Fifteen, and the robbery conspiracy in Count Thirteen of the S2 Indictment (Counts One, Two, Four, and Six of the trial indictment). (Dkt. 13) The Court made clear to the jury that they must find Mendoza guilty of each conspiracy independent of the others: Four of the six counts charge the defendant with the crime of conspiracy. In each [conspiracy] the government has to prove two things, two elements of conspiracy. First of all, that the charged conspiracy actually existed; second that the defendant, Mr. Mendoza, was a member of the charged conspiracy. The government has to prove those two things with respect to each of Counts One, Two, Four and Six. (Tr. 1252:1-25). Garber’s decision to not raise the issue of a multiple conspiracies jury instruction was well within his discretion as appellate counsel and falls far short of ineffective assistance. See Smith, 477 U.S. at 536; Mayo, 13 F.3d at 533; Felder, 2021 WL 3537164, at *4. Mendoza’s Improper Joinder Claims Mendoza claims that the “consolidation of two unrelated cases into a single trial” was prejudicial. (Mot. 14). Mendoza seemingly raises this issue on the merits, and also raises this issue in the context of alleging ineffective assistance as to Cohen and Garber, for their purported failures to raise this issue. (See Mot. 14-17). Mendoza’s claim on the merits is procedurally barred, and his claim regarding ineffective assistance of counsel is without merit. Mendoza alleges, “Attorney Andrew Patel failed to challenge the government's motion to consolidate the cases before being relieved, but Mr. Cohen failed to file a motion for severance after being appointed to the case.” (Mot. 16). There was no Government motion to consolidate: rather, the Government brought these charges under the same indictment. And Patel did file a motion to sever Counts Thirteen, Fourteen, and Fifteen of the S2 Indictment—the counts related 27

to the September 2011 robbery. (Dkt. 79-80). The Court denied the motion, “given that the charged offenses are of the same or similar character, that the proof of each set of charges will somewhat overlap, and that joinder of the offenses at trial will not substantially prejudice Mendoza.” (Dkt. 89 at 9). Because the court had already disposed of this issue, it would have been futile for Cohen to raise it a second time. See, e.g., Love v. United States, 282 F. Supp. 3d 571, 573 (W.D.N.Y. 2017) (“Since the Court denied the motion for severance directing that the case be tried against both defendants simultaneously, there is no reason to expect that a similar motion for severance by [second counsel] would have been successful either.”). Thus, Cohen’s decision to not re-litigate the already denied motion does not constitute ineffective assistance of counsel. Moreover, Mendoza failed to appeal this decision on direct appeal, so he cannot use a habeas petition to re-litigate it here. A defendant is generally “barred from collaterally challenging a conviction under § 2255 on a ground that he failed to raise on direct appeal.” United States v. Thorn, 659 F.3d 227, 231 (2d Cir. 2011). When a defendant has procedurally defaulted a claim by failing to raise it on direct review, the claim may still be raised in a habeas petition if the defendant can “demonstrate either cause and actual prejudice, or that he is actually innocent.” Gupta v. United States, 913 F.3d 81, 84 (2d Cir. 2019). But Mendoza can do neither. As to Mendoza’s ineffective assistance of counsel claim as to Garber for Garber’s failure to appeal the denial of the joinder motion, such a claim fails where the argument is frivolous and where appellate counsel was within his sound discretion to pursue stronger claims. See, e.g., Felder, 2021 WL 3537164, at *4 (finding that decision to not appeal severance decision was within the “wide range of reasonable professional assistance” mandated by Strickland where appellate counsel raised nonfrivolous issues on appeal and where severance claims were meritless). Garber exercised his professional discretion in deciding not to appeal this decision. Rule 14 permits a trial court to order 28

separate trials of counts if it appears that the defendant is prejudiced by the joinder. Fed. R. Crim. Proc. 14. Thestandard of review for refusing to grant relief under Rule 14 is “abuse of discretion,” however, and “the defendant must show not simply some prejudice but substantial prejudice.” United States v. Werner, 620 F.2d 922, 928 (2d Cir. 1980). Here, there is no evidence that Mendoza was prejudiced by the fact that the racketeering conspiracy and robbery conspiracy tried together; indeed, the jury acquitted of Mendoza as to Count Nine of the S2 Indictment and convicted Mendoza as to Count Fourteen of the $2 Indictment—both Section 924(c) charges relating to separate then-crimes of violence, the racketeering conspiracy and the robbery conspiracy. Mendoza’s Claim that the Section 924(c) Charges Deprived Him of a Fair Trial Finally, Mendoza requests that the Court undertake a “Full Review on the Constitutionality of [his] Conviction,” but the issue he appears to actually raise is an allegation that the inclusion of the now-vacated Section 924(c) charges deprived him of a fair trial. (See Mot. 18-19). Mendoza argues that, post-Davis, without the two Section 924(c) counts, his trial on four counts would have been “significantly different” than his trial on six counts; and the use of the terminology “crimes of violence” in the jury instructions “drastically altered the narrative of the case.” (Mot. 19). Mendoza asks the Court, “[N]ow that the vacated 924(c) is unlawful, do the remaining non-violent counts require a new trial and re-litigation of the facts?” (/d.). The answer is no. Mendoza is procedurally barred from making this argument at this stage. because he failed to raise it either in the district court (see Resentencing Tr.) or on direct appeal (see Second Appeal, Dkt. 43) after the post-Davis vacatur of his Section 924(c) conviction. He cannot demonstrate actual prejudice now. See Thorn, 659 F.3d at 231; Gupta, 913 F.3d at 84. On the merits, Mendoza’s claim amounts to a claim of prejudicial spillover. When considering such claims, the Second Circuit looks to two factors: “(1) whether the evidence on the 29

vacated counts was of such inflammatory nature that it would have tended to incite or arouse the jury into convicting on the remaining counts, and (2) whether evidence and facts pertaining to dismissed counts was similar to or different from those relating to the other counts.” United States v. Davidson, 308 F. Supp. 2d 461, 489 (S.D.N.Y. 2004), aff'd, 220 F. App’x 5 (2d Cir. 2007) (citing United States v. Morales, 185 F.3d 74, 82 (2d Cir. 1999)), The Court must also make a general assessment of the strength of the Government’s case on the remaining counts. /d. Mendoza’s claims fail to meet this standard. Mendoza was originally charged with two Section 924(c) counts, the first relating to the racketeering conspiracy and the second relating to the robbery conspiracy. But the jury acquitted Mendoza of the Section 924(c) charge related to the racketeering conspiracy, demonstrating the jury’s ability to distinguish the charges and the evidence relevant to each charge from each other. Further, the use of a firearm in connection with a conspiracy is not so inherently inflammatory as to prejudice a jury’s ability to cooly assess the evidence when considering other charges. See Morales, 185 F.3d at 83: see also Felder, 2021 WL 3537164, at *S (finding on similar facts that Hobbs Act robbery charge did not cause prejudicial spillover where the evidence presented on that count was the same as the evidence presented on the other counts, was no more inflammatory than evidence presented on other counts, where the government’s case was sufficiently strong, and where the jury demonstrated an ability to distinguish between counts). Additionally, the evidence that was used to convict Mendoza of the second Section 924(c) charge was part and parcel of the robbery conspiracy. The presence of the firearm in the car with Mendoza at the time of his arrest was relevant and admissible evidence as to Mendoza’s intent to commit robbery. Mendoza fails to meet his burden of asserting a successful prejudicial spillover claim, and for the same reasons, cannot demonstrate the actual prejudice that would be required to 30

raise this claim for the first time now. Conclusion Mendoza’s motion to set aside his conviction and sentence is denied— There is no need for a hearing in this case because the existing record conclusively shows that Mendoza is not entitled to the requested relief. The Court declines to issue a certificate of appealability because there has been no “substantial showing of the denial of a constitutional right.” 28 U.S.C. § 2253(c)(2); see United States v. Perez, 129 F.3d 255, 260 (2d Cir. 1997). Further, the Court finds, pursuant to 28 U.S.C. § 1915(a)(3) that any appeal from an order denying Mendoza’s motion would not be taken in good faith. See Feliz v. United States, No. 01-cv-5544, 2002 WL 1964347, at *7 (S.D.N.Y. Aug. 22, 2002). Dated: June 11, 2025 fj ite ye ple United States District Judge

By ECF to all counsel and to Petitioner at: Rudy Mendoza, #90731-054 U.S. Penitentiary ADMAX P.O. Box 8500 Florence, CO 81226-8500

$1