Appeal from U.S. District Court, D.N.J. Judge Zahid N. Quraishi, No. 3:24-cv-07125

Before: HARDIMAN, KRAUSE, and MASCOTT, Circuit Judges Argued Mar. 11, 2026; Decided: July 21, 2026

OPINION OF THE COURT

KRAUSE, Circuit Judge. Disciplinary proceedings in the context of higher education are not a zero-sum game. Both sides — students reporting misconduct and students accused of misconduct — have legitimate interests that can coexist when universities employ fair disciplinary procedures to seek truth and accountability. There is a wide range of permissible procedural safeguards that universities can implement to Page 2accommodate both the vital protection of victims’ rights and the essential fairness owed to respondents. Yet in their worthy quest to erase the scourge of assault, particularly sexual assault, from their campuses, many universities have struggled to find the proper balance between these countervailing interests. Princeton University is among them.

We have had occasion to consider concerns about Princeton's disciplinary procedures in the past, see Doe v. Princeton Univ. , 30 F.4th 335, 340-41 (3d Cir. 2022) ( Princeton III ), and Princeton has taken remedial steps in the interim to address some of those issues. But the appeal before us suggests others linger. In the underlying complaint filed by John Doe, a student accused of misconduct by two female students, John alleges that Princeton found him responsible and suspended him for two years as a result of disciplinary proceedings rife with hostility to John, disparate treatment of witnesses, and unsupported credibility assessments.¹ The District Court dismissed the complaint, concluding those allegations were insufficient to support John’s state-law contract claims or his sex discrimination claim under Title IX of the Education Amendments of 1972, 20 U.S.C. § 1681(a). Because we view the allegations as sufficient, we will reverse and remand as to the dismissal of all claims included in this appeal.

Page 3A few weeks after classes began for the fall 2023 semester, John Doe, a sophomore at Princeton, received an unexpected message: A university investigator wanted to interview him as soon as possible regarding unspecified allegations that he had engaged in misconduct. When John met with the investigator a few days later, he learned that two former friends, both female students, had accused him of choking them in separate incidents during the previous semester. He was interviewed for the first and only time about those accusations and denied them. Princeton proceeded to investigate the alleged assaults as violations of its Personal Safety Policy by interviewing student witnesses and compiling an evidence packet that was eventually turned over to the decisionmaker: Princeton's Committee on Discipline. After reviewing the evidence packet and within hours of concluding a late-night disciplinary hearing, the Committee found John responsible for the alleged assaults and suspended him for two years.

We will describe the circumstances of each alleged assault, Princeton’s investigation of those allegations, and its decision to punish John for the charged misconduct. At this stage of the proceedings, there has been no discovery or development of an evidentiary record. We are simply reviewing the District Court's decision to dismiss John's complaint for failure to state a claim, so we must accept as true, and recount below, the factual allegations as described in that complaint. See Princeton III , 30 F.4th at 340. The following narrative, in other Page 4words, “is one -sided because the posture of the case requires it to be, ' Doe v. Purdue Univ. , 928 F.3d 652, 656 (7th Cir. 2019), and “ [d] iscovery might not bear out [John]’s account,” Princeton III , 30 F.4th at 340. Our task at this stage is merely to determine whether, assuming everything John alleges is true, he has stated a claim for relief.

The first alleged assault took place at an on-campus party on the night of March 3, 2023, against Sarah Smith, a female student enrolled at Princeton. There, John “spoke closely and loudly with Sarah, in part out of anger” that she had not upheld her promise to keep a close eye on their mutual friend (Student 3) who had recently been harassed at a different party. App. 56. During that heated conversation, three female bystanders asked Sarah if she was okay, but the nearby security guard did not intervene. John denied making any physical contact with Sarah, and although Sarah “almost immediately complained to her roommates about how John had angrily yelled at her, ” those roommates later testified that Sarah did not say “ the incident ha[d] a physical component. ” App. 56. Sarah likewise did not mention anything about choking when she confronted John the following day. Indeed, it was not until four weeks later, after another friend, Jane Roe, claimed to have been choked by John, that Sarah accused John of doing the same to her.

The second alleged choking incident transpired on the night of April 1, 2023, when John’s friend from high school, Jane Roe, was visiting him at Princeton. The night began with a group of friends — including John, Jane, Sarah, Student 3, and Student 4 — having drinks in John's dorm room before they ventured to an oncampus party, leaving John's roommate (Student 5) behind. The walk to the party was quickly derailed by the fallout from a series of romantic interactions. First, John kissed Jane, upsetting Student 3, who had romantic feelings for John, and causing Student 3 and Student 4 to break off from the group to process Student 3's hurt feelings. Next, Jane kissed Sarah, upsetting John, who had romantic feelings for Jane. John and Jane then got into an argument about the second kiss, while Sarah lay on the ground 'in [a] drunken state.' App. 58. In the middle of that argument, as Student 3 and Student 4 were walking back towards the group, Jane — who was also 'highly intoxicated'abruptly fell backwards into John then dropped to the ground 'screaming and crying loudly.' App. 58. Among Jane's cries were ' words to the effect of, ‘ Z choked me, ’ and, ‘ He choked me ,’” referencing an ex-boyfriend who had choked Jane when they were dating. App. 58.

After the group consoled Jane for one-to-two hours, they made their way back to Sarah’s dorm, where they all agreed Jane would sleep that night. But before parting ways, Jane asked to speak to John alone in a common room and kissed him for the second time that night. Despite the turbulent evening, Page 6Jane stayed on Princeton’s campus for several more nights and spent two of those nights in John’s dorm room.

In a text message to John on the day after the incident, Jane acknowledged that Sarah had not seen the alleged choking. Nonetheless, in the days and months following the incident, Jane attempted to get John to admit that he had choked her. John consistently denied any memory of doing so. The closest Jane got to an admission was a text from John on the night after the incident, delicately declining to accept her recollection of the evening by stating, “it’s probably that it hasn’t fully soaked in yet caus e I haven't processed what I must have done last night, ” App. 60, and a recording of a phone call, which Jane allegedly coerced John to make and send to her —“under threat of reporting him to the University” if he would not comply— in which he told his parents that “on the night in question he was ‘ trying to pull [Jane] in to talk to her and, like, damaged her windpipe ever so slightly ,’” App. 61 (emphasis omitted). Notwithstanding Jane’s accusations and John’s denials , Jane and Sarah continued to spend time with John after the alleged assaults, even vacationing with him at John’s family home for several nights in late August 2023. A few weeks after that trip, however, Jane learned that John had told another male friend that he was “a little scared of her . ” App. 62. As alleged in the complaint, Jane then retaliated by making a complaint to Princeton that John had choked her five months earlier.

Princeton dutifully responded to Jane’s allegation by appointing an internal investigator to obtain Jane’s full account. In their initial interview, Jane told the investigator — despite her previously texting John to the contrary — that Sarah witnessed the alleged choking. She also provided a detailed account of the alleged assault, reporting that John “grabbed her by her throat, and lifted her off the ground ” for “5 -6 seconds,” so she was “straining on her tip toes to get the pressure off of her neck, ” App. 70-71 , and “ having flashbacks and . . . PTSD from her prior rape when she had been choked, ” App. 58. But that account changed when Jane was interviewed again later in the investigation. On that occasion, Jane reported that John had not “ squeezed her neck at all, but rather just pressed from the front as they talked. ” App. 30 (citation modified).

In addition to interviewing Jane, the investigator contacted some of the other students with relevant information about the alleged assault, including Sarah, who said at that point that John likewise had choked her on March 3, 2023. Now responding to a second report of choking, Princeton decided to conduct a joint investigation into the two incidents, eventually culminating in a disciplinary hearing.

As with Jane, Sarah’s narrative evolved over the course of the joint investigation. She initially told the investigator that John put his hand on her throat and used that hand to hold her in place, but in parallel fashion to Jane, later asserted that John had not actually 'meant to choke her' and that he did not Page 8actually 'grasp' or 'squeeze' her neck, but 'was more pushing. ” App. 75-76 (citation modified).

Compounding the inconsistencies in the complainants' accounts was the investigator ’s inconsistent manner of seeking out evidence. She interviewed Jane and Sarah three times each, giving them opportunities to respond to John’s statements . She also, with assistance from Princeton’s Deputy Dean of Undergraduate Students Joyce Chen, interviewed every female and supportive witness identified by the complainants at least once.

As it turned out, however, those witnesses provided little by way of corroboration. Sarah’s roommates stated that Sarah did not mention any physical component to her interaction with John on the night she was allegedly choked; Student 3, who had been walking back towards the group with Student 4 on the night Jane was allegedly choked, confirmed that she did not see what happened between John and Jane; and another female witness asserted that Jane’s reaction on the night of her alleged assault was likely due to Jane feeling triggered when John brought up Jane's ex-boyfriend during their argument. In addition, the investigator asked several female witnesses — but no male witnesses — who had seen Jane in the days following the alleged assault about whether Jane had bruising on her neck and received conflicting answers.

On the other hand, John's initial interview with the investigator, before he had notice of the allegations or formal charges against him, was his only interview that preceded the Page 9formal disciplinary hearing . John's supportive witnesses, all of whom were male, likewise went largely unheard. According to the complaint, the investigator’s treatment of those witnesses did not match the importance of the evidence they had to offer. The investigator did not seek out statements from two individuals with potentially relevant information. The first was John's roommate, Student 5, who, according to the complaint, could have offered relevant information about whether Jane slept in his and John’s shared dorm room during the week after the alleged assault. The second was Student X, a male student at another university who allegedly could have provided evidence showing that Jane had also recently made false choking allegations against him. The investigator did conduct one interview with the sole eyewitness to Jane’s alleged assault, Student 4. Student 4 advised that John did not touch Jane during the interaction, and that, on the night after the alleged assault, Jane told him that “John did not choke her, but he put his hand on her neck and it brought back the memory of her prior assault. ” App. 75 (citation modified).

After several weeks of investigation, Dean Chen scheduled a disciplinary hearing and formally charged John with violating Princeton’s Personal Safety Policy, which prohibits “[a]ny physical assault committed . . . on the premises of the University or in the local vicinity, especially when unprovoked Page 10and/or when injury results.'² App. 36. A Committee on Discipline was then convened to adjudicate the charges through a live hearing.

In advance of the hearing, John requested that Student 4 and Student X be called as witnesses. Denying that request as to Student X, Dean Chen explained that the hearing “would focus only on the incidents at hand,” App. 81, and because Student X 'was not interviewed,' App. 66, “ did not have firsthand knowledge of the incident ,” and “ only knew about the character of Jane ,” App. 81, he would be an impermissible character witness. John then availed himself of alternate avenues to defend against the misconduct allegations. He submitted multiple written statements responding to initial drafts of the evidence packet and detailing Jane's false allegations against Student X, as well as text messages and photographic evidence contradicting Jane and Sarah’s statements about when they cut off contact with John, where Jane slept for the rest of her time on Princeton’s campus, and whether Jane had bruising on her neck following the alleged assault. Among the text messages was an exchange between John and Jane, in which Jane said she would lie under oath Page 11about her false choking allegations against Student X, and that she could get Sarah to lie for her, too.

John’s disciplinary hearing began at 7:27 p.m. on November 6, but one critical party was conspicuously missing — Jane. Dean Chen had informed John the day prior that Jane had declined to attend the hearing and that the proceeding would go forward in any event. At the outset of the hearing, as alleged in the complaint, the Committee questioned John for over ninety minutes. That questioning was “one -sided ” and “ demanding, ” indicating that the Committee members were “hostile to [John’s ] account, ” App. 77-78, and n otwithstanding Dean Chen's prior representation that the hearing “would focus only on the incidents at hand,” much of it “solely focused on whether or not [John] had good character, ” App. 81.

Student 4, whose testimony supported John and who was the only eyewitness, was likewise questioned “in an incredibly hostile manner” for forty minutes. App. 33. The Committee homed in on a portion of the evidentiary packet, which suggested Student 4 had said: “Perhaps it happened if she is saying it and it triggered a memory in [Jane’s] head. I don’t think it's that bad. I don’t think [John] choked her really really hard. I think it is being portrayed a lot worse than it was.' App. 83. According to the complaint, the investigator verified that the quoted statement matched her interview notes. Student 4, however, testified “ unequivocally ” that he did not make that Page 12statement, and a Committee member cut him off as he was trying to explain how he had been misquoted. App. 83.

In contrast, Sarah and the complainant-supportive witnesses were allegedly treated “with kid gloves” and questioned for less than thirty minutes each. App. 77. According to the complaint, one Committee member, Professor Elizabeth Harman, “did not ask Sarah even a single question” and even fell asleep during her testimony. App. 85. The late hour of the hearing and the Committee’s preference to conclude that evening contributed to the brevity of the questioning. For example, the Committee instructed Sarah to “keep her answers as concise as possible” because the hearing was taking longer than expected, and it did not press Sarah on the inconsistencies in her statements. App. 85 (citation modified). As with John, however, the Committee did elicit testimony from her about John’s character.

Although the hearing did not conclude until around 11:30 p.m., the Committee communicated the results of its “extremely limited” deliberations to Dean Chen by 10 a.m. the following morning. App. 87. Dean Chen verbally informed John that the Committee had found him responsible for the charges against him, and the Committee then issued a formal decision letter.

That letter included a statement that the Committee “ had considered all of the information obtained in the investigation Page 13and presented at the hearing ,” but it discussed only two pieces of evidence: (1) what it described as “the women’s continued and consistent descriptions of the incidents in communications and conversations with others” ; and (2) socalled “ admissions and acknowledgements [John] made in text messages.” App. 88 (citation modified). The Committee also wrote that it was not persuaded by John’s assertions that he was blackmailed by Jane into making the purported admissions, though it did not explain why it found John not credible in the absence of any testimony, let alone contradictory testimony, from Jane. Finally, the letter announced the Committee's decision to suspend John for two years — an unprecedented sanction at Princeton for the charges at issue. That significant penalty was supposedly justified by John’s ostensible pattern of misconduct in these two incidents and a prior disciplinary infraction that the Committee characterized — allegedly in error -as an assault.³

Nowhere in the letter did the Committee acknowledge Jane or Sarah ’s inconsistent descriptions of the alleged assaults or specify which of those descriptions it believed were supported Page 14by the evidence. Nor did the Committee identify which text messages it deemed to contain John’s admissions.

At some point following the hearing, John was informed by “ a source involved in the [Committee ’s ] deliberative process ” that “ most of the [Committee] members had decided John was guilty before the hearing even started, ” and Professor Harman, the Committee member who fell asleep during Sarah's testimony, “ even gave an impassioned speech arguing not just that John was guilty, but that it would be a ‘ moral failing ’ to vote for anything other than expulsion .” App. 33 (emphasis omitted).

John subsequently filed an internal appeal of the Committee's decision on the grounds that the disciplinary procedures were not “fair and reasonable” and the “imposed penalty d[id] not fall within the range of penalties imposed for similar misconduct.” App. 37. He raised various arguments, including that Princeton’s investigator treated male and female witnesses differently, that he was not provided the same follow-up interview opportunities as Jane and Sarah, that Princeton failed to obtain testimony from the witnesses he identified, and that the hearing was improperly rushed. Princeton 's appeal panel emailed John a set of questions probing his critiques and, after reviewing John's answers, denied the appeal. Although the appeal panel did not comment on several of John’s arguments, it addressed his core objection by concluding that 'no procedural irregularity existed inPage 15 [John’s] hearing” and that John received adequate opportunities to “submit[] supplemental written materials ” relevant to the charges at issue. App. 92-93.

John then filed the complaint underlying this appeal, claiming sex discrimination in a federally funded education program, in violation of Title IX, 4 as well as breach of contract, breach of the implied covenant of good faith and fair dealing, and gross negligence under state law. The District Court dismissed the complaint, pursuant to Federal Rule of Civil Procedure 12(b)(6), for failure to state a claim, and John filed this timely appeal, challenging all but the dismissal of his gross negligence claim.

The District Court had jurisdiction over John 's Title IX claim pursuant to 28 U.S.C. § 1331, and over his state-law claims pursuant to 28 U.S.C. § 1332. We have jurisdiction under 28 U.S.C. § 1291. We review de novo a district court’s grant of a motion to dismiss. See Princeton III , 30 F.4th at 341. In conducting that review, ' we accept all [the plaintiff's]Page 16 allegations as true and draw all inferences in [his] favor, and we will not dismiss a complaint merely because it appears unlikely that the plaintiff[] can prove those facts or will ultimately prevail on the merits .' Stringer v. County of Bucks , 141 F.4th 76, 84 (3d Cir. 2025) (citation modified). To survive a motion to dismiss, a plaintiff 's complaint must include “ a short and plain statement of the claim showing that [he] is entitled to relief ,' Fed. R. Civ. P. 8(a)(2), which requires “ sufficient factual matter, accepted as true, ” to state a plausible claim -i.e., to 'allow[] the court to draw the reasonable inference that the defendant is liable for the misconduct alleged ,' Ashcroft v. Iqbal , 556 U.S. 662, 678 (2009).

As we consider John 's Title IX and state-law contract claims, we do not write on a blank slate. The legal standards we apply are rooted in two of our prior decisions addressing the sufficiency of a student's complaints that his university conducted biased and unfair disciplinary proceedings. See Princeton III , 30 F.4th at 343; Doe v. Univ. of Scis ., 961 F.3d 203, 209 (3d Cir. 2020) ( USciences ). We discuss those decisions below, distilling their legal precepts before applying those precepts to John’s complaint.

In both USciences and Princeton III , we concluded that the complaint of the male undergraduate challenging the Page 17university's decision to discipline him for alleged sexual misconduct against a female undergraduate was sufficient to state a claim for violation of Title IX, breach of contract, and, in Princeton III , breach of the implied covenant of good faith and fair dealing. Our discussions of the factual allegations in those cases are instructive.

In support of their Title IX claims, both plaintiffs alleged (1) that the university had “yielded to external pressure when implementing and enforcing' its disciplinary policy, and (2) that “ sex was a motivating factor in [the university's] investigation and decision to impose discipline. ” USciences , 961 F.3d at 209; Princeton III , 30 F.4th at 343. Specifically, they pointed to the “more rigorous approach to campus sexual misconduct allegations' ushered in by the United States Department of Education’s 20 11 Dear Colleague Letter (2011 DCL) and universities’ “alleged . . . overreaction” to that Page 18federal pressure.⁵ USciences , 961 F.3d at 209-10; see Princeton III , 30 F.4th at 345. We acknowledged the potential relevance of such historical context and concluded that, even though external pressure 'cannot alone support a plausible claim of Title IX sex discrimination,” USciences , 961 F.3d atPage 19 210, it did “factor [] into the total mix of information supporting a plausible Title IX claim,' Princeton III , 30 F.4th at 345. Based on that total mix, we held that the alleged external pressure, when combined with the plaintiffs ’ allegations of selective enforcement in their individual proceedings, stated a plausible claim for relief.

These cases teach that, in the absence of direct evidence, a Title IX plaintiff can survive a motion to dismiss by alleging both background indicia of sex discrimination (such as external or internal pressure on the university for its handling of complaints of male-against-female violence), and what is sometimes called a “particularized ‘ something more, ’” Doe v. Columbia Coll. Chi. , 933 F.3d 849, 856 (7th Cir. 2019), namely, “ other circumstantial evidence of bias in [his] specific proceeding ,” Doe v. Baum , 903 F.3d 575, 586 (6th Cir. 2018).

Although the circumstantial evidence of bias in both USciences and Princeton III involved assertions that the university asymmetrically enforced its disciplinary policies against male and female students, we also emphasized that plaintiffs are ' free to characterize their claims however they wish ' and are not limited to alleging a particular factual scenario. USciences , 961 F.3d at 209. Instead, we look at the “ total mix of information ” supporting a plaintiff's claim, Princeton III , 30 F.4th at 345, which may include, among other things, allegations that a similarly situated female was treated more favorably by the university or, as here, allegations of procedural irregularities in the university’s disciplinary Page 20proceedings, see Doe v. Stonehill Coll., Inc. , 55 F.4th 302, 334 (1st Cir. 2022) (collecting cases).

As for their state contract claims, the USciences and Princeton III plaintiffs alleged sufficient facts to state plausible claims that the universities breached promises in their disciplinary policies to provide fair and unbiased conduct proceedings. See 30 F.4th at 347; 961 F.3d at 211-12. In Princeton III , as here, we applied New Jersey law to the plaintiff's claims for breach of contract and breach of the implied covenant of good faith and fair dealing. Because New Jersey law required that a university “ follow its own established procedures . . . and that those procedures be fundamentally fair, ” we held that the plaintiff stated a plausible breach-of-contract claim by alleging facts showing the university failed to follow its own requirements that policy violations be established “by a preponderance of the evidence,” and that its decisionmakers be 'impartial and unbiased . ' Princeton III , 30 F.4th at 346-47 (citation modified). And though the plaintiff’s breach -of-implied-covenant claim “share[d] facts with” his breach -of-contract claim, those claims did not arise from an alleged contravention of the same code provisions. Id. at 348 & n.16. Thus, the “factual overlap [was] not fatal” and the plaintiff’s allegations of improper performance and bad faith sufficed to state a claim. Id. at 348.

With those precedents in mind, we now consider the allegations in John’s complaint.

John contends that the investigation and resolution of Jane and Sarah's disciplinary complaints against him violated Title IX by subjecting him to ' the imposition of university discipline when sex [was] a motivating factor in the decision to discipline. ” Princeton III , 30 F.4th at 343. We apply a “straightforward pleading standard” to Title IX claims: “T he alleged facts, if true, must support a plausible inference that a federally-funded college or university discriminated against a person on the basis o f sex.'⁶ USciences , 961 F.3d at 209 (citation modified).

Here, the parties do not dispute that Princeton is a federally funded university and that John's suspension excluded him from participation in its educational programs and activities, so the only question is whether John plausibly alleged that sex motivated Princeton’s decision to suspend him for two years. We address below the sufficiency of John’s allegations as to

1. (1) Page 22background indicia of sex discrimination, consisting in this case of internal and external pressure on Princeton to rigorously pursue allegations of male-against-female assault, and (2) circumstantial evidence of bias, including procedural irregularities, in his disciplinary proceedings.⁷

John argues that the 2011 DCL and related Office for Civil Rights (OCR) investigations into Princeton's handling of sexual assault complaints formed a “backdrop of gender -based pressure” on Princeton “to favor claims of assault brought by women against men. ” Opening Br. 47 (citation modified). But the 2011 DCL was rescinded in 2017 -years before John's disciplinary proceedings. Thus, standing alone, that allegation Page 23carries little weight. See Doe v. Samford Univ. , 29 F.4th 675, 692 (11th Cir. 2022) (stating that allegations about the 2011 DCL, ' a government policy that has been rescinded and replaced, ” do not assist a Title IX plaintiff “ in crossing the line between possibility and plausibility of entitlement to relief ” (citation modified)). As our sister circuits have observed (and even John’s counsel recognized at oral argument ), a plaintiff must allege facts sufficiently close in time to the challenged disciplinary proceedings to show that the pressure on the university to favor certain complainants or respondents based on their gender was active and ongoing.⁸

John’s complaint meets that threshold by alleging specific incidents suggesting the pressure on Princeton, precipitated by the 2011 DCL, continued into the fall of 2023. Upon rescission of the 2011 DCL, for example, “[a]dministrators immediately stated that Princeton would not change its policies and Page 24procedures, insisting that its DCL-era procedures were 'working well' and were 'fair.'' App. 43. Princeton also allegedly submitted comments in 2019 to the United States Department of Education opposing proposed regulations that would increase procedural protections for students accused of sexual misconduct. Then, when those regulations became law in 2020, Princeton announced that “ it would nonetheless strive to adhere as closely as possible to its current system in implementing that final rule ,' App. 46 (citation modified), labeling the new regulations 'problematic in a number of ways,” App. 47 (citation modified).

The complaint describes pressure that was internal, too, including a slew of critical, student-authored articles in the campus newspaper , a protest outside the University president’s building by Princeton Students for Title IX Reform that lasted more than two hundred hours, and objections to on-campus speakers who had made controversial statements about the treatment of campus sexual assault. The newspaper articles specifically criticized Princeton’s investigations and hearings in disciplinary proceedings involving male-against-female assault. In 2022, for instance, a female student authored an op-ed detailing her experience of being “ choked and strangled by a Princeton graduate student ” and criticizing the subsequent disciplinary process because Princeton officials had asked difficult questions that “ made her fe[el] shamed ' and took testimony from witnesses who accused her of lying. App. 50. A nother article, published six months before John’s investigation, lamented that Princeton’s Title IX investigatorsPage 25 “ often asked questions that were ‘ difficult and personal, ’ and that complainants’ character were often questioned during the process. ” App. 52.

Together, these factual allegations suffice to plausibly allege that Princeton was under pressure to crack down on male respondents when enforcing its disciplinary policies against John. See Princeton III , 30 F.4th at 345; USciences , 961 F.3d at 209. But because “ allegations about pressure . . . cannot alone support a plausible claim of Title IX sex discrimination ,” USciences , 961 F.3d at 210, we will next consider whether John ’s complaint adequately alleged 'facts particular to his case' that indicate bias in Princeton's disciplinary process , Columbia Coll. Chi. , 933 F.3d at 855.

To establish the “particularized ‘ something more ’” needed to state a plausible Title IX claim, Columbia Coll. Chi. , 933 F.3d at 856, John points to certain procedural irregularities in his specific processes. Although procedural irregularities can “provide strong support for [a] claim of bias,” Doe v. Oberlin Coll. , 963 F.3d 580, 587 (6th Cir. 2020), and satisfy the “ other circumstantial evidence of bias in [his] specific proceeding ,” Baum , 903 F.3d at 586, they “are not inevitably a sign of sex bias, ” Stonehill Coll., Inc. , 55 F.4th at 334 (emphasis added); see Roe v. St. John ’ s Univ. , 91 F.4th 643, 654 (2d Cir. 2024). The ultimate test is whether the 'total mix of information , ” including background indicia of sex discrimination, plausibly Page 26supports the Title IX claim. Princeton III , 30 F.4th at 345. Those background indicia can provide the necessary “causal connection” between the alleged procedural irregularities and the university's sex -motivated disciplinary outcome. Baum , 903 F.3d at 585 (citation modified).⁹

Such is the case here. John ’s complaint details several procedural irregularities and statements by decision-makers that — in the context of internal and external pressure on Princeton to favor female complainants over male respondents — state a plausible Title IX claim.

Page 27First , John alleges that Princeton conducted its investigation and hearing in an imbalanced manner, prioritizing Jane and Sarah's accounts and witnesses over John's. According to the complaint, Princeton interviewed Jane and Sarah three times each and afforded them an opportunity to respond to John’s statements about the alleged assaults, but it interviewed John only once — and that was without advance notice of the complainants’ allegations or the charges against him. Princeton also gathered testimony from at least five witnesses supportive of the complainants but only one witness, Student 4, supportive of John. The investigator did not interview two male students, Student 5 and Student X, whom John had identified as having relevant information, and when John asked to call Student X to testify at the hearing, Princeton refused. Yet Student X allegedly would have relayed that Jane had a history of making false choking allegations, and Student 5 would have recounted that Jane slept in John’s dorm room during the week after the alleged assault. And while Princeton asked female witnesses whether Jane had any bruising on her neck in the days after the assault, it did not pose that question to any male witnesses.

According to the complaint, the slanted nature of the investigation carried over to the hearing. The Committee questioned John and his sole supporting, male witness for over two hours combined and in a hostile manner, while it questioned Sarah and the complainant-supportive witnesses for less than thirty minutes each and “with kid gloves.” App. 77. Though Princeton justified its refusal to hear from Student XPage 28 on the ground that “ character witnesses were not permitted at the hearing because it would focus only on the incidents at hand” and Student X’s only relevant information was “ about the character of Jane, ” the Committee asked questions of John and other witnesses that “ were solely focused on whether or not he had good character .” App. 81 .

We agree with our sister circuits that those types of irregularities, at least where a university is subject to external or internal pressure, support an inference of sex discrimination. See, e.g. , Doe v. Columbia Univ. , 831 F.3d 46, 56-57 (2d Cir. 2016) (holding that allegations that the university “declined to seek out potential witnesses [the respondent] had identified as sources of information favorable to him” supported, in part, an inference that it was “ motivated to favor the accusing female over the accused male' ); Vengalattore v. Cornell Univ. , 36 F.4th 87, 107 (2d Cir. 2022) (explaining that the university’s failure to “interview certain witnesses or ask certain questions that could have produced information favorable to [the respondent]” contributed to an inference of sex discrimination); Schwake v. Ariz. Bd. of Regents , 967 F.3d 940, 951 (9th Cir. 2020) (noting that “allegations of the [u] niversity’s onesided investigation,” including that the university 'failed to . . . follow up with the witnesses and evidence [the respondent] offered in his defense,” contributed to an inference of sex discrimination).

Second , John contends that Princeton made credibility determinations in a gendered manner. The complaint alleges Page 29that Princeton credited the female complainants' testimony over his and Student 4's testimony despite significant and evolving discrepancies in the complainants' narratives and Jane’s total absence from the hearing. Cf. Purdue Univ. , 928 F.3d at 669 (observing that the strongest fact demonstrating sex discrimination was the decisionmaker’s choice “ to credit [the complainant’s] account without hearing directly from her ”) . He also notes that, while Princeton pressed him and Student 4 on inconsistencies in their statements, it 'did not probe inconsistencies in female witnesses’ stories , ” App. 70, did not allow Student 4 to explain what he insisted was a misquoted statement in the investigator's interview notes, and rejected John's argument that Jane had blackmailed him into making an incriminating voice recording without even questioning Jane.

Similar allegations supported an inference of gender bias in Baum , where the university ' credited exclusively female testimony (from [the complainant] and her witnesses) and rejected all of the male testimony (from [the respondent] and his witnesses) .” 903 F.3d at 586. There, the Sixth Circuit reasoned that , “[w] hen viewed against the backdrop of external pressure, ” gender bias was “ one plausible explanation ” for the university's decision to ' discredit[] all males, including [the respondent], and credit[] all females, including [the complainant]. ' Id. at 586-87. The same inference can be drawn here: Princeton’s credibility determinations in favor of the female complainants and their witnesses over the male respondent and the sole eyewitness — notwithstanding the unresolved inconsistencies in the complainants' accounts or Page 30the Committee’s inability to question Jane at a hearing where credibility was critical — provide circumstantial evidence of gender bias in John’s disciplinary proceedings.

Third , recognizing that “statements by pertinent university officials, . . . [including] decision[-]makers, can support an inference of gender bias,” Doe v. Regents of Univ. of Cal. , 23 F.4th 930, 939 (9th Cir. 2022) (citation modified), the complaint alleges statements of Committee members indicating they “held biased assumptions against male respondents,” id. , or “presumed [ John] to be guilty before the investigation had concluded,' Schiebel v. Schoharie Cent. Sch. Dist. , 120 F.4th 1082, 1100 (2d Cir. 2024). It describes, for example, how “ a source involved in the [Committee] ’s deliberative process ” reported that “ most of the [Committee] members had decided John was guilty before the hearing even started, and one of them — Professor Elizabeth Harman — even gave an impassioned speech arguing not just that John was guilty, but that it would be a ‘ moral failing ’ to vote for anything other than expulsion .” App. 33. This prejudgment, according to the complaint, explains why Professor Harman fell asleep during Sarah's hearing testimony and why the Committee's post-hearing deliberations were “extremely limited.” App. 87.

Fourth , John alleges the outcome of the proceedings is so lacking in factual support that “ the merits of the decision itself, as a matter of common sense . . . support an inference of sex bias. ” Oberlin Coll. , 963 F.3d at 588; see Doe v. Univ. of Ark.-Fayetteville , 974 F.3d 858, 865 (8th Cir. 2020) (“ APage 31 decision that is against the substantial weight of the evidence . . . may give rise to an inference of bias . . . .”); Columbia Univ. , 831 F.3d at 57 (“W hen the evidence substantially favors one party’s version of a disputed matter, ” but the decisionmaker “ conclu[des] in favor of the other side (without an apparent reason based in the evidence), it is plausible to infer . . . bias .”).¹⁰ Here, the Committee found John responsible largely based on (1) “the women’s continued and consistent descriptions of the incidents in communications and conversations with others, ' and (2) 'the admissions and acknowledgements [John] made in text messages.” App. 88 (citation modified). Neither withstands scrutiny.

The complainants ’ descriptions were far from “consistent.” Jane’s account evolved dramatically through her three interviews with Princeton's investigator ; starting with her statement that John “grabbed her by her throat, and lifted her off the ground ” for “5 -6 seconds,” so she was “straining on her tip toes to get the pressure off of her neck, ” App. 70-71, and ending with her assertion that John did not “ squeeze[] her neck at all, but rather just pressed from the front as they talked, ” App. 30 (citation modified). And per Student 4 's testimony, Jane told him, the night after the alleged assault, that “John ‘ did Page 32not choke her, but he put his hand on her neck and it brought back the memory ’ of her prior assault. ' App. 75. Sarah's recollection was also inconsistent; she did not mention any physical component when she debriefed her roommates the night of her alleged assault. Nor did she reference any choking when she confronted John the following day and 'only discussed the fact he had been screaming at her.' App. 57 (citation modified). And when she spoke to the investigator months later, Sarah first stated that John “ had his hand on her throat and was using that hand to hold her in place ,” App. 75 (citation modified), but later recounted that 'John hadn't actually meant to choke her at all” and “ did not actually grasp or squeeze her neck, but was more pushing, ” App. 75-76 (citation modified). In explicitly relying on these statements, the Committee did not simply fail to address the Page 33inconsistencies, 11 but mischaracterized them as “ continued and consistent ,” App. 88 (citation modified), against the substantial weight of the evidence.

The disciplinary record, as described in the complaint, also belies the Committee’s assertion that John made “ admissions and acknowledgements . . . in text messages” about his responsibility for the alleged assaults. App. 88. Instead, “ John repeatedly refused to agree that he had choked Jane,” App. 89, and the Committee did not identify “any actual texts”Page 34 containing admissions 'because it couldn't,' App. 96. The most compromising text that John sent to Jane, the day after the alleged assault, stated 'it's probably that it hasn't fully soaked in yet cause I haven’t processed what I must have done last night. ” App. 60. But that is plausibly read as a placation, not as an admission of guilt. It references what “must have” happened, accepting Jane’s account while making plain that it did not accord with John’s. App. 60. And a decisionmaker’s choice to treat a respondent's statement that he 'may have' done something 'as an admission of guilt' supports the inference that the ' objective was to reach a finding of responsibility rather than to determine what actually happened. ” Schiebel , 120 F.4th at 1102. Princeton’s alleged reliance on unidentified text messages as admissions, “ without an apparent reason based in the evidence ,” Columbia Univ. , 831 F.3d at 57, thus also supports John’s assertion that “ sex was a motivating factor in [Princeton’s] investigation and decision to impose discipline, ” USciences , 961 F.3d at 209.¹²

In sum, the complaint alleges a series of procedural irregularities — including an imbalanced investigation and hearing, gender-based credibility determinations, and a decision against the weight of the evidence — that are commonly recognized in case law as indicative of gender bias, Page 35as well as statements evincing prejudgment by the decision-makers. In view of those allegations and the internal and external pressure on Princeton to rigorously pursue allegations of male-against-female assault, we reject Princeton’s argument that the complaint should be dismissed because “[a]t most, [John] has suggested that Princeton had a bias in favor of the complaining parties, ” without “ specifically tying that favoritism to gender. ” Answering Br. 49-50. At the motion to dismiss stage, even if “ anti-male bias is not the only plausible explanation for the university ’ s conduct, or even the most plausible, alternative explanations are not fatal. ” Princeton III , 30 F.4th at 344 (citation modified). Assuming that inferences of both anti-male bias and anti-respondent bias are supported by the allegations in the complaint, we still must draw “ all reasonable inferences in the light most favorable to [the plaintiff],” USciences , 961 F.3d at 210 — not give primacy to the explanation that would relieve the university of liability. Nor must the plaintiff refute alternative explanations for the Page 36alleged procedural irregularities to satisfy the plausibility standard.¹³

Here, t he 'total mix of information' supporting John's claim, Princeton III , 30 F.4th at 345 — namely, allegations of procedural irregularities and statements suggesting prejudgment by decision-makers, combined with internal and external pressure on Princeton to favor female complainants over male respondents — makes it reasonable to infer that “sex was a motivating factor in [Princeton's] investigation and decision to expel [John], ” and thus suffices to plausibly state a Page 37claim of sex discrimination in violation of Title IX, USciences , 961 F.3d at 210.

In addition to his Title IX claim, John alleges that Princeton's conduct throughout his disciplinary proceedings breached the terms of the Rights, Rules, and Responsibilities (RRR) policy that governed their contractual relationship. Under New Jersey law, “the relationship between [a] university and its students should not be analyzed in purely contractual terms,” Mittra v. Univ. of Med. & Dentistry of N.J. , 719 A.2d 693, 694 (N.J. Super. Ct. App. Div. 1998), so we should not “rigid[ly] appl[y] . . . the law of contracts to students ’ disciplinary proceedings, ” Napolitano v. Trs. of Princeton Univ. , 453 A.2d 263, 272 (N.J. Super. Ct. App. Div. 1982). Instead, when a student disciplined for misconduct brings a breach-of-contract claim under New Jersey law, we must assess whether “ the school ‘ follow[ed] its own established procedures, ’” and whether “ those procedures [are] ‘ fundamentally fair .’” Princeton III , 30 F.4th at 346 (quoting Hernandez v. Don Bosco Preparatory High , 730 A.2d 365, 367, 376 (N.J. Super. Ct. App. Div. 1999)).

John alleges that (1) Princeton failed to follow its established procedure under the RRR of requiring his guilt be proven by “clear and persuasive” evidence , and (2) Princeton denied his appeal without arranging for a “fair and reasonable” investigation and resolution process. We agree on both points.

The RRR requires that, before deciding a student has violated the Personal Safety Policy, a majority of the Committee members must “conclude that the evidence presented constitutes a clear and persuasive case in support of the charges.' App. 160. In Princeton III , we considered whether Princeton satisfied its own standard of proof and concluded it did not meet even the lower “preponderance of the evidence ' standard in the relevant RRR provisions when it (1) “disregard [ed] evidence that tended to . . . exculpate ” the respondent; (2) “failed to consider the entirety of the evidence with a neutral gaze'; and (3) 'rendered inconsistent and skewed credibility determinations.” Princeton III , 30 F. 4th at 347 (citation modified).¹⁴ To be clear, we did not state that these three categories of evidence are necessary to establish that a university failed to apply its own standard of proof; we simply acknowledged that they are sufficient.

Page 39In John’s case, the allegations in the complaint easily clear that “preponderance” bar, let alone the “clear and persuasive” standard applicable here, for the same three reasons.

First , John adequately pleads facts suggesting Princeton “disregard [ed] evidence that tended to . . . exculpate [him], ” Princeton III , 30 F. 4th at 347, when it brushed aside evidence related to Student 4, Student 5, and Student X. According to the complaint, Student 4 — the only eyewitness to the interaction between John and Jane — testified that John did not touch Jane during the alleged assault and that Jane told Student 4 the following night that John “‘ did not choke her, but he put his hand on her neck and it brought back the memory ’ of her prior assault .' App. 75. The Committee allegedly overlooked this testimony when finding John responsible for the alleged assaults.¹⁵ John also contends that Princeton disregarded testimony from Student 5 and Student X by failing to seek it out, ignored evidence showing Jane had falsely accused Student X of choking her, and failed to account for Jane's statement to John that she would lie about those Page 40accusations if Student X sought to hold her accountable. Such evidence suggests Jane had a “motivation[] to lie,” Princeton III , 30 F.4th 347 (citation modified), and her unwillingness to appear and be examined at the hearing precluded any credibility assessment by the Committee. Yet the Committee nonetheless credited her account.

Second , John sufficiently alleges that Princeton “failed to consider the entirety of the evidence with a neutral gaze .” Id. (citation modified). Notably, John pleads that “ a source who was involved in the [Committee’s] deliberative process said the [Committee] had prejudged guilt ” and that the Committee’s deliberation period was “extremely limited”— occurring in the few waking hours between 11:30 p.m. on the night of the hearing and 10:00 a.m. the next morning. App. 87. In addition, allegations about the Committee’s hostile and extended questioning of John and Student 4, while treating “witnesses supporting the complainants with kid gloves, if not total disinterest,” App. 77, suggest Princeton was not neutral about the outcome of the proceedings.

Third , the Committee's failures to challenge Jane and Sarah's inconsistent narratives of the alleged assaults, their reports to friends in the aftermath, and their time spent with John after the alleged assaults — followed by its explicit statement in its decision letter that it relied on “the women’s continued and consistent descriptions of the incidents in communications and conversations with others ,' App. 88 (citation modified) — indicate that the Committee madePage 41 “skewed credibility determinations,” Princeton III , 30 F.4th at 347 (citation modified). No doubt, John had credibility issues of his own, particularly related to the voice recording in which he said he ' damaged [ Jane’s ] windpipe ever so slightly. ” App. 61 (emphasis omitted). But without an opportunity to question Jane about whether she did indeed coerce John into making that recording, as John alleges, that recording is hardly “clear and persuasive” evidence of John’s culpability.

John also asserts that Princeton breached its contractual relationship with him by upholding the Committee’s decision on appeal even though the disciplinary procedures were not “fair and reasonable.” App. 97. The appellate panel allegedly concluded “that no procedural irregularity existed in [John’s] hearing ,” App. 92, but as discussed above, that is not so. As described in the complaint, John's disciplinary procedures were replete with asymmetries in credibility determinations and the treatment of complainant-supportive versus respondent-supportive witnesses. Just as those procedural errors demonstrate that Princeton did not resolve the disciplinary charges by “ clear and persuasive ” evidence at the Committee level, by the same token, they show that it was error Page 42to deny John’s appeal on the ground that those Committeelevel procedures were “fair and reasonable.” 16

John's final claim is that Princeton breached the implied covenant of good faith and fair dealing, which exists in every contract governed by New Jersey law, by failing to conduct a fair investigation or hearing, and by neglecting to provide a meaningful rationale for its decision. Opening Br. 35 (citing Wade v. Kessler Inst. , 798 A.2d 1251, 1259 (N.J. 2002)). Before addressing the sufficiency of John's allegations, we consider Princeton’s argument that this claim should have been dismissed because it is duplicative of John’s breach -of-contract claim.

Princeton asserts that New Jersey law supports dismissal of redundant breach-of-implied-covenant claims, see Answering Br. 38-39 (citing Wade , 798 A.2d at 1261), but John’s c omplaint lays out 'several additional ways' that Princeton Page 43breached the implied covenant beyond failing to apply the RRR’s “ clear and persuasive ” evidence standard and appeal provisions, App. 98. Princeton III teaches that 'factual overlap” between these breeds of state-law contract claims “is not fatal” at this early stage of litigation. 30 F.4th at 348. So , although the allegations supporting John’s breach-of-implied-covenant claim — i.e., facts related to the Committee's unfair proceedings and insufficient rationaleoverlap with those indicating Princeton breached the standard of proof and appeal provisions of the RRR, dismissal on that basis is not warranted. Cf. id. at 348 n.15 (noting that “ New Jersey courts permit the pleading and pursuit of alternative . . . theories ,” so long as they do not result in “double recovery” (citation modified)).

We therefore consider the sufficiency of John’s allegations, namely, whether he has stated a claim that Princeton’s conduct had “ the effect of destroying or injuring the right of the other party to receive the benefits of the contract ,' regardless of whether Princeton violated an express term of the RRR. Brunswick Hills Racquet Club, Inc. v. Route 18 Shopping Ctr. Assocs. , 864 A.2d 387, 396 (N.J. 2005) (citation modified). Proof of the breaching party’s “ bad motive or intention is vital to an action for breach of the covenan t,” id. (citation modified), but bad faith does not require a blatant desire to harm the other contracting party. Even “without regard to the harm caused to [the] plaintiff,” id. at 397, New Jersey courts have found bad faith in at least two circumstances: (1) where an actor employs “s ubterfuges and evasions in the performance of a Page 44contract . . . even [when] the actor believes his conduct to be justified, ” id. at 396 (quoting Restatement (Second) of Contracts § 205 cmt. d (A.L.I. 1981)); and (2) where a party is “ vested with the exercise of discretion under a contract ” and “ exercises its discretionary authority arbitrarily, unreasonably, or capriciously, with the objective of preventing the other party from receiving its reasonably expected fruits under the contract, ” Wilson v. Amerada Hess Corp. , 773 A.2d 1121, 1128, 1130 (N.J. 2001).

Both pertain here. Princeton may have acted without desire to harm John and believed its conduct was justified by the need to balance complainants’ and respondents’ interests . Nonetheless, its actions, as alleged in the complaint, plausibly show bad faith under New Jersey law 's definition , and that definition controls our analysis.

First , a s to “subterfuges and evasions,” John contends that Princeton sidestepped deciding “ which version of Jane's or Sarah's story' of the alleged assaults it believed to be true, App. 88 (emphasis omitted), and speciously labeled the complainants “consistent” despite their varying accounts of the incidents and the contradictory testimony from an eyewitness to Jane's alleged assault, App. 88. Princeton also allegedly told John that Student X could not testify at the hearing because “character witnesses w ere not permitted at the hearing, ” which “ would focus only on the inciden ts at hand,” then switched bait by asking John and Sarah questions “solely focused on whether or not he had good character.” App. 81.

Page 45Second , the complaint ’s allegations suggest Princeton exercised its discretion to determine whether John violated the RRR “ arbitrarily, unreasonably, or capriciously. ” Wilson , 773 A.2d at 1130. Indeed, its allegations mirror the three we deemed “sufficient, at this stage, to allege improper performance” in Princeton III : that Princeton (a) “ subjected [the respondent] to a discriminatory disciplinary process, ” (b) “ disregarded exculpatory evidence ,” and (c) “ construed all discrepancies and inconsistencies in [the complainant s’ ] favor. ” 30 F.4th at 348 (citation modified).

Here, John likewise asserts Princeton subjected him to a 'discriminatory disciplinary process,' id. , by '[giving] Jane and Sarah multiple interview opportunities to respond to testimony before the hearing without doing the same for John ,” “[failing] to press Jane and Sarah as firmly as it did John ,” “[evincing] gender bias through the witnesses who were pursued and what they were (and were not) asked ,' and '[engaging] in unbalanced questioning of the witnesses,' App. 98-99. He contends that Princeton “disr egarded exculpatory evidence, ” Princeton III , 30 F.4th at 348 (citation modified), by “[fail ing] to interview' and '[r efusing] to call relevant witnesses,” “[refus ing] to accept relevant text message evidence,' and '[c onducting] a rushed investigation . . . that devoted more resources to obtaining evidence for Jane and Sarah than for John. ” App. 98-99. Finally, the allegation that Princeton “ based its decision on a rationale that was almost entirely conclusory, which falsely stated that Jane's and Sarah’s stories had remained consistent , [and] which refused Page 46to identify (because it couldn't) which of Jane's stories was more credible ,” App. 100, signals that Princeton “ construed all discrepancies and inconsistencies in [ the complainants’ ] favor ,' Princeton III , 30 F.4th at 348 (citation modified). It did so, according to the complaint, because, rather than using the disciplinary process for its intended truth-seeking purpose, it “ [p] rejudge[d] the outcome of [John’s] hearing before it even started .” App. 99.

In sum, by alleging both evasive and arbitrary conduct that deprived him of a fundamentally fair disciplinary process under the RRR, John has stated a plausible claim for breach of the implied covenant of good faith and fair dealing.

The disruption of a student’s education and the psychological scars of on-campus assault, particularly sexual assault, are considerable. Universities understandably wish to punish such misconduct when it occurs and to preempt further incidents through penalties that result in general deterrence, but those aims are not well served when disciplinary proceedings are conducted in a manner that neither fairly airs both parties’ evidence nor weighs that evidence using the university’s own standards. Because Joh n’s factual allegations raise a plausible inference that Princeton, however well-intentioned, conducted gender-biased and unfair disciplinary proceedings, we will reverse and remand the District Court’s dismissal of his Title IX, breach-of-contract, and breach-of-implied-covenant claims.

Page 47Counsel for Appellant Kimberly Blasey Justin Dillon Christopher C. Muha [ Argued] DILLON PLLC

Jamie Hoxie Solano DYNAMIS LLP

Counsel for Appellee James A. Keller [Argued] Patrick F. Nugent Amy L. Piccola SAUL EWING LLP