Before BAUER, CUDAHY, and TINDER, Circuit Judges.

Donald R. Jackson (argued), Attorney, Peoria, IL, for Plaintiff–Appellant.Mary Ellen Welsh (argued), Attorney, Office of the Attorney General, Chicago, IL, for Defendants–Appellees.

CUDAHY, Circuit Judge.

Mark Rodgers, a longtime employee of the Secretary of State's office in Illinois, was fired from his job but reinstated after arbitration. In this litigation he claims that two white managers targeted him for termination because he is black. The district court granted summary judgment for the defendants. Rodgers's primary argument on appeal is that he and two white coworkers engaged in the same alleged misconduct, yet his white counterparts

were treated less harshly. Rodgers focuses on the coworker with the same job title, but the other white employee, Rodgers's immediate supervisor, is the better comparator. We have observed in many decisions that employees of differing ranks usually make poor comparators, but the rationale behind that general rule does not apply in this case. We conclude, based on evidence that Rodgers engaged in the same conduct as his supervisor but was disciplined more harshly, that a jury could reasonably infer that Rodgers was discriminated against. We thus vacate the judgment and remand for trial.

I. Background

The following account is drawn from the evidence at summary judgment, as viewed in the light most favorable to Rodgers. See Sow v. Fortville Police Dep't, 636 F.3d 293, 299–300 (7th Cir.2011). When Rodgers was fired in 2006, he was the only black employee in a crew of more than 27 lawn-maintenance workers. He had been a grounds worker for more than 20 years, most of that time as a supervisor, and had been disciplined only twice, receiving verbal warnings in 1989 and 1999. The decision to fire Rodgers was made by Donna Fitts, the director of his department, and Stephen Roth, the personnel director. Both are white. Their reasons for discharging Rodgers, which are set out in an August 2006 letter signed by Roth, all stemmed from two incidents.

The first incident involved the alleged misuse of state property. In the Fall of 2005, the Inspector General for the Secretary of State's office issued a report summarizing an investigation into allegations that Rodgers and his immediate supervisor, Dave Rusciolelli, let crew members borrow state equipment for personal use. According to the report, the practice was discovered when a temporary employee injured himself retrieving a state-owned aerator from a crew member's house. The crew member insisted that both Rodgers and Rusciolelli had known he borrowed state equipment and, in fact, embraced an “open door” policy regarding personal use of state equipment. Rodgers and Rusciolelli denied the crew member's allegations, though, according to the Inspector General, Rodgers gave shifting explanations: Rodgers initially said he thought the temporary employee had been injured while getting the aerator from a storage area, but later he told an investigator that he sent the temporary employee and a regular crew member to retrieve the aerator as soon as he heard that an employee had taken it home. The investigator reported that Rodgers had become irritated when asked about this apparent discrepancy, which Rodgers attributed—truthfully, we must assume—to his initial confusion about the investigator's questions. Rodgers also had told the investigator that the temporary employee ascribed his injury to lifting weights, a statement that the young man denied making. At this stage, of course, we must credit Rodgers's version.

In response to the Inspector General's report, the personnel department initiated discipline against the employees involved. The crew member who borrowed the aerator was fired, in part because he borrowed it, but principally because management discovered around this same time that he lacked a valid driver's license, which was a condition of his employment. That employee contested his discharge, but lost. For Rodgers, the personnel department proposed an 18–day suspension, yet for Rusciolelli, who is white, only a 3–day suspension was contemplated. For reasons not disclosed in the record, neither proposed suspension was ever implemented, and, so it appeared at the time, the matter was dropped.

The second incident came to light in early 2006 after Fitts, who was critical of

Rusciolelli's management of Rodgers, was appointed as the acting director of their department. In January 2006, Fitts received from Rodgers and Rusciolelli time slips for the month of December. Rusciolelli alone was responsible for completing these time slips, but he had asked Rodgers to help out, which Rodgers had agreed to do. Fitts thought that the December time slips were missing necessary information, and so she and Roth began to investigate.

What they found is that crew members were requesting leave that was not shown in the payroll system as earned. It turned out that Fitts's predecessor, Cecil Turner, had authorized Rusciolelli to give workers “comp time” for their overtime hours after the Secretary of State imposed a moratorium on overtime pay. Turner had proposed, and gained union approval for, this system because some tasks, such as snow removal, could not always be completed during normal working hours. Under Turner's system the overtime hours were recorded informally “off the books,” and employees used personal time to draw against the banked overtime.

A few days after receiving the December 2005 time slips from Rusciolelli and Rodgers, Fitts met with them to discuss timekeeping. She laid out what she thought was the correct procedure for completing time slips, and she ordered the two men to stop recording overtime “off the books” and to start submitting all overtime requests to her two days in advance. She gave the same order to Robert Deffenbaugh, a white crew supervisor also under Rusciolelli's supervision. But Fitts did not bar overtime work, and from the record it appears that she continued Turner's practice of awarding compensatory time, though with the understanding that formal records would be kept of the overtime hours. Rusciolelli and Rodgers stopped using the “off the books” system immediately, but Fitts still wanted Rodgers to redo his December time slips to conform to her newly implemented procedure. She says that he failed to properly complete the slips even after she returned them to him several times; Rodgers says that he couldn't complete the forms because Fitts took away his computer, where the relevant data was stored.

Rusciolelli and Rodgers also told Fitts that Rodgers had retained handwritten records of his crew's past overtime hours. Fitts asked for copies, and Rodgers provided what he thought was a complete set (though he later found additional overtime records in a filing cabinet outside his office). During February 2006 a liaison from the personnel department tallied the uncompensated overtime for Rodgers's crew and mistakenly concluded that the workers had taken more than that amount of hours as personal time. Rodgers recognized the calculation to be mistaken and refused to sign off on it, as did most of his crew. The next day, March 1, the liaison told Rodgers that Fitts wanted to meet with him and his crew that afternoon, more than an hour after Rodgers's shift was to end. The liaison asked him to notify his crew, and Rodgers contacted every member he could locate on site. But Rodgers himself skipped the meeting because he wasn't told that it was mandatory or whether overtime had been approved for his attendance. Fitts had no authority,

she admits, to order an employee to attend an after-hours meeting without first approving overtime, which she had not done.

Fitts called in Rodgers for another meeting—also after hours but this time with overtime approved—in late March. Roth also was present, and Rodgers brought along a union representative. During the meeting Fitts demanded, and Rodgers relinquished, his original time slips from December, which included the files he had overlooked when he gave copies to Fitts in January. This March meeting was the last time that Fitts asked Rodgers for records. Later, after Rodgers had been fired, an independent auditor concluded that his records of the crew's overtime were accurate.

Although this allegedly “inaccurate reporting of time” was one reason given for Rodgers's discharge, no other employee involved in the incident was punished so severely. Rusciolelli, for his perceived role in the timekeeping problems and equipment misuse, was demoted to yard work. Deffenbaugh, the other crew supervisor, was not disciplined; during discovery Fitts explained the different treatment by saying that Rodgers alone had failed to turn over all of his records when asked, but when pressed she admitted that she never asked Deffenbaugh to turn over his records. Fitts had focused on Rodgers and Rusciolelli, she said, because they sent her incomplete time slips, and she never thought to investigate Deffenbaugh's records. According to the department liaison, Deffenbaugh's crew never protested the department's calculations of their time.

The parties describe one more event that, though not given as a reason for Rodgers's discharge, does show further friction between Fitts and Rodgers. Historically, Rodgers had raised and lowered the flag on top of the capitol building when needed. That task was considered dangerous because of the location of the flagpole, and thus earned him an annual stipend. In April 2006, however, Fitts decided in coordination with the union to rotate flag duty among crew members who volunteered, for a fixed rate per assignment. When the time came to lower the flag, the entire crew, including Rodgers, declined the task. Fitts threatened to discipline Rodgers if he didn't volunteer, but she threatened no one else. Eventually, another employee agreed to lower the flag, and Rodgers avoided discipline for his refusal.

In May 2006 Fitts wrote a memorandum to Roth recommending that Rodgers be fired. As grounds for termination, she asserted that Rodgers had (1) allowed abuse of state equipment and then failed to cooperate with the Inspector General's investigation, and (2) improperly recorded overtime and refused to assist her effort to rectify the problem. Roth then sent Rodgers a letter listing formal charges against him. That communication goes well beyond the content of Fitts's memorandum and accuses Rodgers of (1) lying about his knowledge of the aerator misuse and the temporary employee's injury, (2) allowing the misuse of state equipment, (3) improperly completing the December time slips even after they were returned to him for reworking, (4) being “complicit with” Rusciolelli in failing to “accurately report and record” compensatory time, and (5) skipping the March meeting and not telling his crew to attend. Rodgers's union representative wrote a rebuttal denying the charges, and Fitts responded that the rebuttal lacked evidentiary support. Roth then fired Rodgers.

Rodgers filed a grievance challenging his discharge, and the grievance led to arbitration. The arbitrator concluded that the Secretary of State's office had failed to produce clear and convincing evidence of all but one of the charges against Rodgers.

(The arbitrator faulted Rodgers for not attending the March 2006 meeting and concluded that he should have gone to the meeting and then filed a grievance. Otherwise, though, the arbitrator rejected every charge leveled against Rodgers.) Following arbitration, Rodgers was reinstated with back pay.

Meanwhile, Rodgers brought this suit in the district court against the Secretary of State under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e–2000e–17, and against Fitts and Roth under 42 U.S.C. § 1981 and § 1983. In moving for summary judgment, the defendants argued that Rodgers had no direct evidence of discrimination and no circumstantial evidence of suspicious timing or racially charged commentary surrounding his termination. They also contended that Rodgers could not prove that another similarly situated employee had received more favorable treatment because, in their view, no other employee had faced as many charges of improper behavior.

In opposing the motion, Rodgers argued that indeed he received worse treatment than similarly situated white employees. In particular, he asserted, Rusciolelli and the white crew leader, Deffenbaugh, were both treated more leniently despite engaging in the same alleged misconduct. They, too, had participated in Turner's overtime system, but Rusciolelli was demoted, not fired, and Deffenbaugh wasn't even asked to turn over his records for review. What's more, said Rodgers, Rusciolelli had been accused of adopting an “open door” policy for state equipment, and yet the equipment policies of Deffenbaugh, his white subordinate, were not even examined.

The district court concluded that Rodgers had failed to produce enough evidence to survive summary judgment. According to the court, Rodgers had no direct evidence of discrimination, not even “suspicious timing of events, ambiguous statements made by Roth or Fitts, or behavior toward or comments directed at other employees in the protected group.” Moreover, the court reasoned, Rodgers hadn't proven that white employees received systematically better treatment. In fact, the court said, Rodgers had not shown that any similarly situated white employee received better treatment than him. This failure alone, the court continued, doomed Rodgers's attempt to show discrimination using the indirect method established in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). The court accepted the premise that Rodgers was similarly situated to Rusciolelli and Deffenbaugh in regard to the timekeeping issue. The court reasoned, however, that Rodgers was required to identify a coworker who was “similarly situated regarding both reasons for the termination.” In concluding that Rodgers had not done so, the court overlooked Rodgers's contention that he also was similarly situated to both Rusciolelli and Deffenbaugh concerning the misuse of state property. The court declined to address the other elements of the indirect method.

II. Analysis

In challenging the grant of summary judgment, Rodgers argues that he produced sufficient evidence of discrimination under both the direct and indirect methods. We disagree about the direct method. Rodgers contends that certain “bits and pieces” of circumstantial evidence—including that he was the only black member of his crew, that an arbitrator determined that he was fired without just cause, and that he alone was ordered to lower the flag—raise an inference of discriminatory intent. His argument is unpersuasive, however, because the evidence does not point directly to a discriminatory reason

for his termination, as it must. See Van Antwerp v. City of Peoria, Ill., 627 F.3d 295, 298 (7th Cir.2010); Petts v. Rockledge Furniture LLC, 534 F.3d 715, 720 (7th Cir.2008). He also contends that the defendants systematically treated white employees more favorably. Although, as we discuss later, Rodgers does have evidence that Rusciolelli and Deffenbaugh received better treatment, this evidence falls short of showing that white employees received better treatment on a regular and repeated basis. See Silverman v. Bd. of Educ. of City of Chi., 637 F.3d 729, 734 & n. 2 (7th Cir.2011).

Rodgers's stronger argument relies on the indirect method of proving discrimination, which applies equally to discrimination claims under Title VII, § 1981, and § 1983. See Egonmwan v. Cook Cnty. Sheriff's Dep't, 602 F.3d 845, 850 n. 7 (7th Cir.2010). To survive summary judgment under this method, a plaintiff must produce evidence that he (1) belongs to a protected class, (2) met his employer's legitimate performance expectations, (3) suffered an adverse employment action, and (4) was treated worse than similarly situated employees outside the protected class. McDonnell Douglas, 411 U.S. at 802, 93 S.Ct. 1817; Egonmwan, 602 F.3d at 850. If the plaintiff satisfies these elements, he must then prove that any legitimate, nondiscriminatory reasons offered for the adverse action are pretext. McDonnell Douglas, 411 U.S. at 802, 804, 93 S.Ct. 1817; Egonmwan, 602 F.3d at 850. Because the first and third elements of the prima facie case are uncontested, Rodgers focuses on showing that he performed his job satisfactorily and yet was treated more harshly than his white counterparts.

When a black employee produces evidence that he was disciplined more severely than white employees who shared similar shortcomings, the second and fourth elements of the indirect method merge. See Luster v. Ill. Dep't of Corr., 652 F.3d 726, 730–31 (7th Cir.2011); Weber v. Univs. Research Ass'n, Inc., 621 F.3d 589, 594 (7th Cir.2010); Elkhatib v. Dunkin Donuts, Inc., 493 F.3d 827, 831 (7th Cir.2007); Adams v. Wal–Mart Stores, Inc., 324 F.3d 935, 940 (7th Cir.2003). Employers “cannot intentionally discipline poor employees more severely on the basis of race.” Luster, 652 F.3d at 730. Thus, Rodgers made out a prima facie case if his evidence would establish that the defendants extended leniency to similarly situated white employees who engaged in similar conduct. See Elkhatib, 493 F.3d at 831; Adams, 324 F.3d at 940. The similarly situated analysis requires a flexible, common-sense inquiry that asks “whether the other employees' situations were similar enough to the plaintiff's that it is reasonable to infer, in the absence of some other explanation, that the different treatment was a result of race or some other unlawful basis.” Luster, 652 F.3d at 730; see McGowan v. Deere & Co., 581 F.3d 575, 579–80 (7th Cir.2009); Elkhatib, 493 F.3d at 831.

Throughout this litigation Rodgers has devoted the most attention to Deffenbaugh, since they share the same job title. But Rodgers also made out a strong case that the defendants unfairly focused on his perceived shortcomings and disciplined him more severely than Rusciolelli, and Rusciolelli's status as Rodgers's supervisor does not diminish the strength of this comparison. Many times we have acknowledged that supervisors usually make poor comparators for plaintiffs claiming employment discrimination. E.g., Patterson v. Ind. Newspapers, Inc., 589 F.3d 357, 366 (7th Cir.2009); Burks v. Wis. Dep't of Transp., 464 F.3d 744, 751 (7th Cir.2006); Patterson v. Avery Dennison Corp., 281 F.3d 676, 680 (7th Cir.2002). But usually does not mean always, and we have not

held that a supervisor is never an apt comparator. Supervisors typically make unrealistic comparators because, as relevant to the issues in a particular case, employees of higher rank commonly have different job duties or performance standards. E.g., Burks, 464 F.3d at 751; Keri v. Bd. of Trs. of Purdue Univ., 458 F.3d 620, 626 (7th Cir.2006); Ajayi v. Aramark Bus. Servs., Inc., 336 F.3d 520, 532–33 (7th Cir.2003). And especially in situations where the plaintiff alleges discriminatory promotional practices, it is difficult for the plaintiff to show that he deserved to be promoted over an employee of a higher rank, who usually possesses more experience. E.g., Hudson v. Chi. Transit Auth., 375 F.3d 552, 561 (7th Cir.2004); Patterson, 281 F.3d at 680; Hoffman–Dombrowski v. Arlington Int'l Racecourse, Inc., 254 F.3d 644, 651 (7th Cir.2001). Yet when uneven discipline is the basis for a claim of discrimination, the most-relevant similarities are those between the employees' alleged misconduct, performance standards, and disciplining supervisor. See Weber, 621 F.3d at 594; Amrhein v. Health Care Serv. Corp., 546 F.3d 854, 860 (7th Cir.2008); Humphries v. CBOCS West, Inc., 474 F.3d 387, 404–05 (7th Cir.2007); Adams, 324 F.3d at 940; Snipes v. Ill. Dep't of Corr., 291 F.3d 460, 463 (7th Cir.2002); Radue v. Kimberly–Clark Corp., 219 F.3d 612, 617–18 (7th Cir.2000). Formal job titles and rank are not dispositive; an employer cannot “insulate itself from claims of racial discrimination” by making formalistic distinctions between employees. Johnson v. Zema Sys. Corp., 170 F.3d 734, 743 (7th Cir.1999); see Gorzynski v. Jetblue Airways Corp., 596 F.3d 93, 109 n. 7 (2d Cir.2010); Lee v. Kan. City S. Ry. Co., 574 F.3d 253, 260–61 (5th Cir.2009); Filar v. Bd. of Educ. of City of Chi., 526 F.3d 1054, 1061–62 (7th Cir.2008); Boumehdi v. Plastag Holdings, LLC, 489 F.3d 781, 791 (7th Cir.2007); Bellaver v. Quanex Corp., 200 F.3d 485, 494 (7th Cir.2000). Thus, when a plaintiff and his supervisor were accused of making similar mistakes, were equally responsible for avoiding those mistakes, and were disciplined by the same superior, the plaintiff can make a realistic comparison with his supervisor for purposes of establishing a prima facie case of discrimination. See Filar, 526 F.3d at 1062 (comparing plaintiff to employees with more seniority, when seniority status was discretionary); Lathem v. Dep't of Children & Youth Servs., 172 F.3d 786, 793 (11th Cir.1999) (comparing plaintiff to direct supervisor accused of same misconduct); Johnson, 170 F.3d at 743–44 (comparing two intermediate managerial employees).

The defendants concede that the difference in treatment between Rodgers and Rusciolelli was “pretty big,” and they do not try to distinguish the two men on the basis of rank. Instead, echoing the district court, the defendants assert that Rodgers made more mistakes and engaged in a greater degree of misconduct than Rusciolelli and, on that basis, stands apart. Embracing this premise, however, would require us to disregard the evidence presented at summary judgment. Indeed, just reading the reasons Fitts and Roth gave for firing Rodgers is enough to dispel any notion that Rodgers was accused of more-serious shortcomings than Rusciolelli.

To start, we are at a loss to understand the defendants' insistence on branding Rodgers more blameworthy than Rusciolelli for the aerator incident. What the evidence shows is that Fitts and Roth relied entirely on the Inspector General's report in concluding that Rodgers had allowed state property to be used for private gain and then, when this “open door” policy was questioned, tried to evade responsibility by lying to investigators. But if the defendants' assumption about Rodgers

was sincere, how could they have viewed Rusciolelli as less culpable? The Inspector General's information came from a single employee who accused both Rodgers and Rusciolelli of endorsing private use of state equipment, and both men denied the accusation. The Inspector General did not exonerate Rusciolelli, and since he supervised Rodgers, the sensible conclusion would be that Rusciolelli bore the greater responsibility for adopting a misguided policy of allowing maintenance workers to borrow state-owned equipment. The defendants have never contended that they legitimately held Rodgers to a higher standard than Rusciolelli vis-à-vis enforcing legitimate equipment policies, and yet Rusciolelli, the one with the greater authority to enforce or alter those policies, was demoted, while Rodgers was fired. Moreover, it is specious to rejoin, as the defendants do, that Rodgers was less cooperative with the Inspector General's investigation than Rusciolelli. The defendants demoted Rusciolelli in part because they believed the allegation that the “open door” policy had his blessing, which Rusciolelli flatly denied when interviewed by investigators. So the defendants necessarily concluded that Rusciolelli, like Rodgers, had lied during the Inspector General's probe, and yet they gave no reason for treating Rusciolelli's dishonesty more leniently than Rodgers's.

Even more perplexing is the defendants' contention that Rodgers should receive the lion's share of the blame for the perceived timekeeping problems. The defendants insist that Rodgers was disciplined, not for using the “off the books” method of recording compensatory time, but for keeping time records which Fitts and Roth thought were incomplete and inaccurate. Once again, though, how does that belief, if sincere, differentiate Rodgers from Rusciolelli? If we blind ourselves to their actual job duties, then, at most, Rodgers shared responsibility for the perceived paperwork errors, since he and Rusciolelli jointly submitted the December time slips to Fitts along with Rodgers's supporting documentation. The defendants, though, have never explained how timekeeping errors even concerned Rodgers. As Rusciolelli makes clear in an affidavit, timekeeping and attendance records were solely his responsibility. Rodgers had assisted with timekeeping, a duty outside of his job description, as a favor so that Rusciolelli would have more time to focus on other matters. The defendants, then, are effectively maintaining that they fired Rodgers for poorly performing a task that was not a legitimate employment expectation, while retaining the white employee whose job it was to perform that very task. Rodgers was a volunteer, and the defendants cannot hide behind the pretense that they did not know the duties of a position they created.

What's more, when Deffenbaugh is added to the mix, the explanation that sloppy recordkeeping was a principal reason for firing Rodgers is even more suspect. Deffenbaugh also used Turner's “off the books” timekeeping method, but the only reason Fitts gave for not reviewing his records—even after she personally instructed him to stop using Turner's system—is that Rodgers drew her attention by tendering his records. That explanation is no answer at all; Rodgers tendered his records only because he was directed to do so by Fitts, who did not make the same demand of Deffenbaugh. It is no surprise, then, that mistakes were found only in Rodgers's records, since his were the only records reviewed. We note, however, that, although the defendants' treatment of Deffenbaugh is relevant to our analysis, Rodgers could not have established a prima facie case of discrimination based on a comparison to Deffenbaugh alone. Deffenbaugh, unlike Rusciolelli, was not directly accused of failing to enforce

equipment policies, so we cannot realistically characterize the two as sharing a “comparable set of failings.” See Haywood v. Lucent Techs., Inc., 323 F.3d 524, 530 (7th Cir.2003); accord Lucas v. PyraMax Bank, FSB, 539 F.3d 661, 667 (7th Cir.2008); Faas v. Sears, Roebuck & Co., 532 F.3d 633, 642–43 (7th Cir.2008); Burks, 464 F.3d at 751.

In the end, only Rodgers's absence from the “mandatory” meeting and the accusation that he failed to notify his crew of that meeting arguably distinguish him from Rusciolelli. But, at a minimum, there remains a material question of fact about whether Rodgers's attendance was a legitimate employment expectation. The liaison who contacted him about the same-day, after-hours meeting did not indicate that Fitts had approved overtime, and yet Fitts herself had mandated that all overtime was to be personally approved by her two days in advance. And though her approval of overtime for the meeting might seem implicit in her decision to schedule it, Fitts admitted in her deposition that, in fact, she had no authority to order Rodgers to attend the meeting without agreeing to pay him for attending, which she had not done. Moreover, as for Rodgers's purported failure to inform his crew about the meeting, he testified, and the defendants did not contradict, that he succeeded in telling all but one crew member—who was not at work that day—about the impromptu meeting, even though he was given only a few hours to accomplish the task. Some of the crew members who knew about the meeting did not attend, but, as far as this record shows, the defendants proceeded to fire Rodgers without having any factual support for their accusation that he failed to tell the absent crew members about the meeting.

For these reasons, we are persuaded that Rodgers presented enough evidence that he was similarly situated to Rusciolelli to prove a prima facie case under the indirect method. And on this record, his evidence establishing a prima facie case also defeats the defendants' claim that their reasons for firing him were nonpretextual. The analysis of the prima facie case and pretext often overlap, e.g., Everroad v. Scott Truck Sys., Inc., 604 F.3d 471, 477–78 (7th Cir.2010); Scruggs v. Garst Seed Co., 587 F.3d 832, 838 (7th Cir.2009); Adelman–Reyes v. Saint Xavier Univ., 500 F.3d 662, 665 (7th Cir.2007), as they do here. Rodgers, of course, needed evidence tending to show that the defendants' proffered reasons for his termination were not just erroneous, but were “ ‘factually baseless, were not the actual motivation for the discharge in question, or were insufficient to motivate the discharge.’ ” Gordon v. United Airlines, Inc., 246 F.3d 878, 888–89 (7th Cir.2001) (quoting Adreani v. First Colonial Bankshares Corp., 154 F.3d 389, 395 (7th Cir.1998)); see Humphries, 474 F.3d at 407. Rodgers met that burden by showing that the defendants' proffered reasons for singling him out were all disingenuous. At summary judgment Rodgers produced evidence that, except for missing a meeting, Rusciolelli engaged in the same alleged misconduct and committed the same purported recordkeeping mistakes and yet received only a demotion. The evidence also shows that the defendants hounded Rodgers, the lone black employee, about purported timekeeping errors while ignoring both Deffenbaugh, whose records were not even examined, and Rusciolelli, who admitted that timekeeping was solely his responsibility. And as for the March meeting, Fitts admitted that she had no authority to demand attendance from Rodgers or his subordinates, so that reason for his termination, like the others, is “unworthy of credence.” See

Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 143, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000); Tex. Dep't of Cmty. Affairs v. Burdine, 450 U.S. 248, 256, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). Considering all this evidence together, a jury reasonably could conclude that the actual reason the defendants fired Rodgers was his race. See Hasan v. Foley & Lardner LLP, 552 F.3d 520, 530–31 (7th Cir.2008); Rudin v. Lincoln Land Cmty. Coll., 420 F.3d 712, 726–27 (7th Cir.2005); Gordon, 246 F.3d at 890–92.

Accordingly, the grant of summary judgment in favor of the defendants is vacated, and the case is remanded to the district court for trial.