What is substantial evidence?

The Substantial Evidence Standard in Administrative and Arbitration Proceedings

When no-fault insurance disputes proceed to compulsory arbitration under Article 75 of the CPLR, the resulting arbitration awards are subject to limited judicial review. Courts reviewing these awards apply the “substantial evidence” standard, examining whether the arbitrator’s decision has a rational basis in the record rather than conducting independent review of the merits. This deferential standard of review stems from fundamental principles of administrative law: arbitrators and administrative agencies possess specialized expertise in their respective areas, and courts should not substitute their judgment for those specialized decision-makers except in cases of clear error or irrationality.

The substantial evidence test creates significant practical consequences for no-fault litigants. Even when a party believes an arbitration award is wrong—perhaps even clearly wrong—courts will uphold the award if any reasonable basis exists in the record supporting it. This makes the arbitration proceeding itself critically important: parties must develop complete records at arbitration because judicial review provides limited opportunities to correct adverse awards. Understanding what substantial evidence means, and more importantly what it doesn’t mean, is essential for practitioners navigating the no-fault arbitration system.

The Court of Appeals decision in Matter of Marine Holdings, LLC v. New York City Commission on Human Rights provides important clarification about the substantial evidence standard’s application, emphasizing just how limited judicial review really is in this context.

The Legal Framework: Judicial Review Limitations in Administrative Proceedings

Matter of Marine Holdings, LLC v New York City Commn. on Human Rights, 2018 NY Slip Op 03303 (2018)

Every Article 75 compulsory arbitration case involving legal insufficiency or credibility of evidence claims is subjected to a substantial evidence review. The term is a misnomer as we learned this week from the Court of Appeals:

""Quite often there is substantial evidence on both sides” of an issue disputed before an administrative agency (Matter of Ridge Rd. Fire Dist. v Schiano, 16 NY3d 494, 500 ), and the substantial evidence test “demands only that a given inference is reasonable and plausible, not necessarily the most probable” (id. at 499 ). Applying this standard, “ourts may not weigh the evidence or reject determination where the evidence is conflicting and room for choice exists” (Matter of State Div. of Human Rights (Granelle), 70 NY2d 100, 106 ). Instead, “when a rational basis for the conclusion adopted by the is found, the judicial function is exhausted. The question, thus, is not whether find the proof … convincing, but whether the could do so” (id. ).”

The Court’s abdication of the ability to review administrative awards that are inherently questionable but not “off the charts wrong” is improper. We have three levels here. The first is that the award has support. The second is the award smells bad. The third is the award is “off the charts wrong”.

We all agree the Courts have no business addressing level 1 and must address level 3. It is level 2 that forms the basis of so many appeals. I only suggest that a deeper look at level 2 cases is warranted. The dissent (I did not publish it) did just that.

Legal Significance: The Deferential Nature of Substantial Evidence Review

The Court of Appeals’ articulation of the substantial evidence standard reveals just how limited judicial review of arbitration awards really is. Several key principles emerge from the court’s analysis that profoundly impact no-fault arbitration practice.

First, the court acknowledges that “quite often there is substantial evidence on both sides” of disputed issues. This recognition fundamentally shapes the review standard: when conflicting evidence exists, courts cannot choose which evidence they find more persuasive. Instead, courts must uphold the arbitrator’s decision if the evidence the arbitrator credited provides a rational basis for the award, even if contrary evidence might have supported a different conclusion. This means arbitrators possess wide discretion in credibility determinations and factual findings, with courts powerless to intervene unless the arbitrator’s conclusions are completely unsupported.

Second, the standard “demands only that a given inference is reasonable and plausible, not necessarily the most probable.” This language is particularly significant. Even when courts believe a different inference from the evidence would be more probable or more logical, they must uphold arbitration awards based on less probable but still plausible inferences. This gives arbitrators enormous latitude to draw inferences favorable to one party or the other, secure in the knowledge that judicial review will not disturb those inferences unless they are completely implausible.

Third, and most importantly, courts “may not weigh the evidence or reject determinations where the evidence is conflicting and room for choice exists.” This prohibition on evidence weighing eliminates what courts normally do in de novo review: independently assess credibility, weigh conflicting testimony, and reach independent conclusions about disputed facts. Instead, courts must simply determine whether the record contains some evidence—perhaps even minimal evidence—supporting the arbitrator’s findings.

The cumulative effect of these principles is that arbitration awards in no-fault cases are extraordinarily difficult to overturn on substantial evidence grounds. Unless an award is completely divorced from record evidence—what Jason Tenenbaum aptly characterizes as “off the charts wrong”—courts will uphold it even when the losing party believes the arbitrator clearly erred.

The Three-Tier Framework: Support, Suspicious, and Irrational

Jason Tenenbaum’s tripartite framework for understanding arbitration awards provides helpful analytical structure. Level 1 awards—those with clear support in the record—obviously survive substantial evidence review. Level 3 awards—those completely unsupported by record evidence or based on legal errors—are properly vacated. The problematic category is Level 2: awards that “smell bad” but have some minimal record support.

Level 2 awards might include situations where arbitrators credited implausible testimony, drew questionable inferences from ambiguous evidence, or reached conclusions that seem inconsistent with the weight of the evidence. These awards trouble reviewing courts because they appear wrong, yet the substantial evidence standard prevents meaningful judicial intervention. The arbitrator’s specialized role and the deference accorded to fact-finding decisions mean courts must uphold even questionable awards as long as some rational basis exists.

Tenenbaum’s observation that Level 2 cases form the basis of “so many appeals” reflects practitioners’ frustration with the substantial evidence standard. Losing parties in arbitration frequently believe the arbitrator made clearly erroneous factual findings or drew unreasonable inferences. Yet the deferential standard of review means these perceived errors rarely warrant judicial intervention, leading to appeals that seem meritorious to appellants but fail under the restrictive review standard.

The dissent Tenenbaum references (but doesn’t publish) apparently took “a deeper look” at a Level 2 case, suggesting that more rigorous judicial review might have reached a different conclusion. However, the majority’s adherence to the traditional substantial evidence standard reflects prevailing law: courts should not second-guess arbitrator judgments even when those judgments seem questionable, as long as some record support exists.

Practical Implications: The Critical Importance of Arbitration Preparation

For no-fault practitioners, the Marine Holdings decision’s articulation of the substantial evidence standard has crucial practical consequences. Because judicial review provides such limited relief from adverse arbitration awards, practitioners must treat arbitration hearings with the same seriousness and preparation they would give trials. Several key practice points emerge:

First, comprehensive record development at arbitration is essential. Parties cannot rely on judicial review to correct evidentiary deficiencies or incomplete records. All relevant evidence must be submitted to the arbitrator, with careful attention to establishing proper foundations, authenticating documents, and creating clear records of witness testimony. When evidence is excluded or not presented at arbitration, it generally cannot form the basis for overturning awards on appeal.

Second, credibility evidence deserves particular attention. Because arbitrators have wide latitude in credibility determinations that courts won’t disturb, parties should present maximum evidence bearing on witness credibility. This includes impeachment evidence, prior inconsistent statements, bias or interest testimony, and corroborating or contradicting evidence. While arbitrators may still credit testimony despite credibility problems, parties must at least create records showing why testimony should be disbelieved.

Third, parties should request detailed written arbitration awards setting forth findings of fact and conclusions of law. While not always required, detailed awards facilitate judicial review by making clear what factual findings the arbitrator made and what evidence supported those findings. When awards are merely conclusory, courts have even less ability to review them meaningfully, as the record may not clearly establish what the arbitrator found or why.

Fourth, parties facing adverse awards should carefully evaluate whether judicial review is likely to succeed before incurring appellate costs. Under the substantial evidence standard, most factual determinations and credibility findings will survive judicial review. Appeals are most likely to succeed when arbitrators committed clear legal errors, exceeded their authority, or rendered awards completely unsupported by record evidence—relatively rare occurrences.

The substantial evidence standard’s strictures mean that arbitration awards in no-fault cases have substantial finality. Parties dissatisfied with awards have limited recourse, making the arbitration proceeding itself the critical forum for case resolution.

Related Articles

• Understanding Collateral Estoppel in No-Fault Arbitrations: Critical Insights for New York Insurance Claims
• Gary T day
• A default that is more than meets the eyes
• Causation – Be aware of seeking a trial de novo after a master arbitrator affirms an award
• New York No-Fault Insurance Law