Comparator Evidence After Lewis: Failing the “All Material Respects” Prima Facie Standard Does Not Make Comparators Irrelevant to the “Convincing Mosaic” Inquiry

I. Introduction

Case: Harrius Johnson v. Miami-Dade County
Court: United States Court of Appeals for the Eleventh Circuit
Date: 2026-03-19

This appeal concerns race discrimination and retaliation claims brought by Harrius Johnson, an African American police officer who was disciplined and ultimately terminated by the Miami-Dade County Police Department (“MDPD”). Johnson sued Miami-Dade County (the “County”) under 42 U.S.C. § 1983, Title VII, and the Florida Civil Rights Act of 1992 (“FCRA”), alleging that discipline and termination were motivated by racial discrimination and retaliation for protected activity (including EEOC complaints).

The case returned to the Eleventh Circuit after an earlier remand directing the District Court to reassess comparator evidence under the clarified comparator standard announced in Lewis v. City of Union City, 918 F.3d 1213 (11th Cir. 2019) (en banc). After the District Court again granted summary judgment to the County, the Eleventh Circuit—now guided by intervening clarification in Ismael v. Roundtree, 161 F.4th 752 (11th Cir. 2025)—affirmed, while refining how comparator evidence fits into the overall circumstantial-evidence (“convincing mosaic”) analysis.

II. Summary of the Opinion

The Eleventh Circuit affirmed summary judgment for the County on both discrimination and retaliation.

  • Discrimination: Johnson failed to establish a prima facie case because none of his ten proposed comparators were “similarly situated in all material respects” under Lewis, chiefly due to materially different disciplinary histories and different types of misconduct. The court then assessed whether Johnson nonetheless presented a “convincing mosaic” of circumstantial evidence. It held that, although non-Lewis comparators may still be considered at the mosaic stage, Johnson’s comparator evidence—standing alone—was too weak to permit a reasonable inference of discriminatory intent.
  • Retaliation: Even assuming a prima facie case, the County articulated legitimate nonretaliatory reasons (Johnson’s disciplinary history and conduct). Johnson’s proffer (timing gaps of roughly two months and speculative allegations of falsified disciplinary reports) did not allow a reasonable jury to infer retaliatory intent.
  • Other issues: Challenges to discovery and evidentiary rulings were barred by the law-of-the-case doctrine because they were decided or could have been raised in the first appeal.

III. Analysis

A. Precedents Cited and Their Role

1. Framework and burden-shifting

The court reiterated that discrimination and retaliation claims based on circumstantial evidence are commonly analyzed under McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802-03 (1973). The opinion also relied on the circuit’s repeated holdings that the same analytical structure applies across Johnson’s parallel statutory vehicles: the court earlier noted that the elements/methods are identical when Title VII and § 1983 are used as parallel remedies, citing Cross v. Alabama, 49 F.3d 1490 (11th Cir. 1995) and “Richardson,” and that FCRA is construed consistent with Title VII, citing Harper v. Blockbuster Entm't Corp., 139 F.3d 1385 (11th Cir. 1998).

2. Comparator standard at the prima facie stage

The remand and this second appeal center on Lewis v. City of Union City, 918 F.3d 1213 (11th Cir. 2019) (en banc), which rejected loose formulations such as “nearly identical” or “same or similar” and adopted the “similarly situated in all material respects” standard. Lewis also provided practical guideposts (same basic misconduct, same rule, same supervisor, similar disciplinary history), while emphasizing the ultimate question: whether the employees “cannot reasonably be distinguished,” quoting Young v. United Parcel Serv., Inc., 575 U.S. 206 (2015).

3. What happens if prima facie proof fails or pretext is not shown

The opinion’s most doctrinally significant move is its integration of Ismael v. Roundtree, 161 F.4th 752 (11th Cir. 2025) and Tynes v. Fla. Dep't of Juv. Just., 88 F.4th 939 (11th Cir. 2023). These cases clarify that:

  • A “prima facie case” under McDonnell Douglas is not itself a “substantive standard of liability” (Tynes), but a device creating a rebuttable presumption.
  • A plaintiff’s failure to show pretext does not automatically entitle the defendant to summary judgment (Ismael). The ultimate inquiry is whether the record supports a reasonable inference of discriminatory or retaliatory intent.
  • The “convincing mosaic” formulation from Smith v. Lockheed-Martin Corp., 644 F.3d 1321 (11th Cir. 2011) governs that ultimate inquiry, and pretext evidence is relevant but not strictly necessary (Ismael).

4. The “convincing mosaic” categories

The court drew on Berry v. Crestwood Healthcare LP, 84 F.4th 1300 (11th Cir. 2023) to identify nonexclusive categories of circumstantial evidence: suspicious timing/ambiguous statements, systematically better treatment of similarly situated employees, and evidence of pretext.

5. Summary judgment and evidentiary standards

The opinion recited core summary judgment principles, citing Anthony v. Georgia, 69 F.4th 796 (11th Cir. 2023), Qui v. Thomas Cnty. Sch. Dist., 814 F.3d 1227 (11th Cir. 2016), and Young v. City of Palm Bay, 358 F.3d 859 (11th Cir. 2004). It also invoked Fed. R. Civ. P. 56(c)(4) to note that a declaration must be based on personal knowledge and admissible facts.

6. Direct evidence and timing in retaliation

For the direct-evidence distinction, the court cited Jefferson v. Sewon Am., Inc., 891 F.3d 911 (11th Cir. 2018). For temporal proximity in retaliation, it cited Hurlbert v. St. Mary's Health Care Sys., Inc., 439 F.3d 1286 (11th Cir. 2006), reasoning that a two-month gap was too long to meaningfully support an inference of retaliatory motive on timing alone.

7. Law-of-the-case limits on re-litigation

The court relied on United States v. Jordan, 429 F.3d 1032 (11th Cir. 2005) and United States v. Escobar-Urrego, 110 F.3d 1556 (11th Cir. 1997) to hold that new challenges to pre-remand discovery/evidentiary rulings were barred because they were decided or could have been raised earlier.

B. Legal Reasoning

1. The court corrects the analytical frame post-Ismael

A notable procedural point is the panel’s candid acknowledgment that the District Court “appears” to have treated the absence of pretext as dispositive. Citing Ismael v. Roundtree, the Eleventh Circuit clarified that this is incomplete: the correct question is whether, with all evidence viewed favorably to the plaintiff, a jury could infer illicit intent—discrimination or retaliation—even if pretext is not established.

2. Discrimination: separating (i) prima facie comparator proof from (ii) overall mosaic relevance

The discrimination section draws the opinion’s sharpest doctrinal line:

  • At step one (prima facie): comparator evidence must satisfy Lewis (“similarly situated in all material respects”) to trigger the presumption. The court held Johnson’s comparators failed because they lacked comparable disciplinary histories and did not engage in the same basic misconduct (in particular, Johnson’s record included one Record of Counseling and four Disciplinary Action Reports over about two and a half years, including insubordination/disrespect and other misconduct leading to termination).
  • After step one (ultimate inference): even if comparators fail Lewis, they can still be considered as part of the total evidentiary record. Here the opinion relies on Tynes v. Fla. Dep't of Juv. Just. for the proposition that differences go to “how much weight” the comparator evidence receives, not whether it is categorically irrelevant to discriminatory intent.

Applying that principle, the court affirmed because Johnson’s case effectively rested only on weak comparator proof. With no additional circumstantial indicators (statements, suspicious internal communications, procedural irregularities tied to race, etc.), the comparators—materially distinguishable on discipline and conduct—could not, by themselves, allow a reasonable jury to infer discriminatory intent.

3. Retaliation: assuming prima facie proof, then testing the full record for retaliatory intent

For retaliation, the court assumed (without deciding in detail) that Johnson could make a prima facie case, then held the County rebutted any presumption with legitimate reasons rooted in Johnson’s disciplinary history. It then examined Johnson’s proffered circumstantial evidence:

  • Timing: the shortest gap was “about two months,” which the court deemed too attenuated, by itself, to support an inference of retaliation or to cast doubt on the County’s reasons (citing Hurlbert v. St. Mary's Health Care Sys., Inc. for comparison).
  • Alleged falsification/inconsistent explanations: Johnson argued Captain White’s DARs lacked a factual basis and that later testimony by Director Patterson (in 2019) that he did not trust White implied that Patterson knowingly relied on false reports in 2016. The court found this was speculative and unsupported by evidence that the reports were actually falsified or that Patterson knew of falsity at the relevant time.

Considering the record cumulatively, the court held no reasonable jury could infer retaliatory intent.

C. Impact

Although the panel ultimately affirmed summary judgment, the opinion has meaningful operational impact for Eleventh Circuit employment cases:

  1. Comparator evidence has a dual life. The decision reinforces a two-tier approach: Lewis governs whether comparator evidence is strong enough to create the prima facie presumption, but failing Lewis does not automatically remove comparator evidence from the case. Instead, at the “convincing mosaic” stage, differences affect weight, not categorical admissibility or relevance.
  2. District courts must not treat “no pretext” as the end of the analysis. Post-Ismael, courts should ensure the final step is always addressed: whether the record as a whole supports a reasonable inference of illicit intent.
  3. Plaintiffs face a practical warning: if comparators are the only evidence offered, and they are materially distinguishable, the “mosaic” will likely be too thin to reach a jury—even if comparator evidence is not formally excluded from consideration.
  4. Retaliation timing arguments remain constrained. A gap of roughly two months, without more, will often be insufficient to support an inference of retaliatory intent in this circuit.

IV. Complex Concepts Simplified

1. “Prima facie case” under McDonnell Douglas

In this context, “prima facie case” does not mean the plaintiff has proven discrimination; it means the plaintiff has presented a minimal showing that, if unrebutted, creates a temporary presumption of unlawful intent. The employer can rebut by producing a legitimate reason.

2. “Similarly situated in all material respects” (Lewis)

This is a stricter comparator test used to decide whether the plaintiff gets the presumption at step one. It generally requires comparable misconduct and comparable disciplinary history, among other practical similarities.

3. “Convincing mosaic” (Smith v. Lockheed-Martin Corp.)

Think of all circumstantial evidence—comparators, timing, statements, inconsistent explanations, procedural irregularities—as tiles in a picture. No single tile is required, but the overall picture must permit a reasonable inference that unlawful intent motivated the decision.

4. Pretext

“Pretext” means the employer’s stated reason is not the true reason and is masking discrimination or retaliation. This opinion, following Ismael v. Roundtree, stresses that while pretext evidence helps, a plaintiff’s inability to disprove the employer’s rationale cannot be the only reason to grant summary judgment; the full record must still be assessed for inferable unlawful intent.

5. Law-of-the-case doctrine

If an issue was decided in an earlier appeal—or could have been raised then but was not—it generally cannot be re-litigated in later stages of the same case. That is why Johnson’s later-raised challenges to discovery/evidentiary rulings were not reviewed.

V. Conclusion

Harrius Johnson v. Miami-Dade County affirms summary judgment for the employer but clarifies the post-Ismael, post-Lewis roadmap: (1) Lewis controls whether comparator evidence is sufficient to create the prima facie presumption, yet (2) comparator evidence that fails Lewis may still be considered in the “convincing mosaic” inquiry, with differences affecting weight, and (3) courts must assess the entire record for an inference of illicit intent rather than ending the case solely because pretext was not shown. On these facts, Johnson’s materially distinguishable comparators and speculative retaliation theories did not permit a reasonable inference of discrimination or retaliation.