Inconsistent Findings on Factually Identical Design Claims Require a New Trial

A. Evidentiary Sufficiency and Weight Are Different Inquiries

BMW challenged both the legal sufficiency and the weight of the evidence. These challenges address different questions. Legal sufficiency asks whether a rational evidentiary route to the verdict exists at all. Weight review asks whether the verdict, although rationally possible, can stand on a fair interpretation of the evidence.

Precedents governing review

Cohen v Hallmark Cards, 45 NY2d 493, 499 [1978]

This decision supplies the controlling legal-sufficiency standard: a verdict may be set aside only when no valid line of reasoning and permissible inferences could lead rational people to the jury’s conclusion. Its influence is fundamental because it prevents an appellate court from substituting its preferred assessment of the proof for a verdict that has a rational evidentiary foundation.

Lu v Gamba, 244 AD3d 1488, 1490 [3d Dept 2025], lv dismissed 46 NY3d 942 [2026]

The court cites this recent Third Department authority alongside Cohen v Hallmark Cards. It confirms the continued application of the demanding legal-insufficiency threshold, rather than a fresh appellate determination of which side presented the more persuasive evidence.

Alaimo v General Motors Corp., 32 AD3d 627, 627-628 [3d Dept 2006]

This precedent supports the separate weight-of-the-evidence inquiry: whether the evidence so favored the unsuccessful party that the verdict could not have been reached on any fair interpretation. It frames BMW’s challenge as more than merely demonstrating that contrary evidence existed.

Warnke v Warner-Lambert Co., 21 AD3d 654, 657 [3d Dept 2005]

Cited with Alaimo v General Motors Corp., this case reinforces the fair-interpretation standard. Together, these authorities explain why disagreement with the jury’s weighing of technical evidence does not itself justify a new trial.

Janof’s proof cleared both thresholds. His expert explained that a tether would limit inflation distance and provide a more stable surface for contact with the driver. The expert also characterized the deflector design as uncommon and not reasonably safe for ordinary use, emphasizing that its maximum inflation distance was approximately six inches greater than that of the proposed tethered alternative.

BMW countered with evidence concerning the restraint system’s development, testing, and performance. That competing proof created an issue for the jury; it did not eliminate the rational basis for Janof’s position.

Precedents supporting deference to the jury

Killon v Parrotta, 28 NY3d 101, 108 [2016]

The court uses this authority’s “utterly irrational” formulation to explain why the verdict survived legal-sufficiency review. Janof’s expert supplied a coherent relationship between the challenged design, a safer alternative, and the alleged injury mechanism.

Streit v Katrine Apts. Assoc., Inc., 212 AD3d 957, 958 [3d Dept 2023]

This case provides additional support for sustaining a verdict that rests on a rational interpretation of the record. Its role is to reinforce the appellate court’s limited authority to reject an adequately supported factual determination.

Veeder v Community Health Plan, 281 AD2d 756, 757 [3d Dept 2001]

Cited in the same discussion, this precedent further supports the conclusion that the liability findings were not irrational. The court invokes it as part of the established framework for respecting a jury’s supported factual conclusions.

Abar v Freightliner Corp., 208 AD2d 999, 1001 [3d Dept 1994]

This authority directly addresses conflicting expert testimony: resolving competing expert opinions is ordinarily the jury’s task. It explains why BMW’s technical evidence, however substantial, did not compel rejection of Janof’s expert.

Fritz v White Consol. Indus., 306 AD2d 896, 897 [4th Dept 2003]

This decision serves two functions. First, it supports upholding a verdict that represents a fair interpretation of competing evidence. Later, it supports the connection between the design evidence and whether the product was reasonably fit for its ordinary purpose under the warranty claim.

The distinction is important to the disposition: the court ordered a new trial because the findings conflicted, not because Janof had failed to present a legally viable case.

B. When Different Answers Become Legally Inconsistent

A verdict is not inconsistent merely because a plaintiff wins one claim and loses another. The decisive question is whether one finding necessarily defeats an element essential to another finding. Here, the court examined the actual evidence and theories, rather than assuming that differently named claims required separate factual answers.

Precedents defining inconsistency and its remedy

Motta v Eldred Cent. Sch. Dist., 172 AD3d 1575, 1578 [3d Dept 2019]

Applying CPLR 4111 (c), this case supplies the remedial framework for inconsistent verdict answers: the court must require further jury consideration or order a new trial when the statutory conditions are met. It establishes that an unresolved contradiction cannot simply be accepted as the basis for judgment.

Barry v Manglass, 55 NY2d 803, 805 [1981]

This decision supplies the substantive test: inconsistency arises only when a verdict on one claim necessarily negates an element of another. It guards against treating every mixed verdict as defective and directs attention to whether the particular findings can logically coexist.

BMW had objected to separate interrogatories and proposed asking whether the airbag was “defectively or negligently designed.” Janof opposed that approach and expressly accepted the risk of an inconsistent verdict. The trial court nevertheless submitted separate questions, producing a finding of defective design but no negligent design.

The appellate court found no evidentiary explanation for that divergence. Janof identified no negligent conduct independent of BMW’s selection of the challenged airbag. The same evidence addressed the design’s risks, utility, and feasible alternative under both theories. Separating the questions therefore divided what was, on this record, essentially one design inquiry.

C. Strict Products Liability and Negligent Design: Distinct Theories, Overlapping Standards

Adams v Genie Indus., Inc., 14 NY3d 535, 543 [2010]

This precedent establishes that when both strict-liability and negligent-design claims are pleaded, the standards for resolving the design-defect question apply to both. It supplied the doctrinal basis for rejecting an artificial separation of Janof’s claims where their factual foundation was identical.

Reis v Volvo Cars of N. Am., 24 NY3d 35, 40 [2014]

This is the closest authority on verdict-sheet structure. It characterizes separate inquiries into whether a design was “negligent” and “not reasonably safe” as essentially redundant in the relevant setting. The court relied on it both to identify the problem and to explain why the separate questions permitted unsupported divergent findings.

Searle v Suburban Propane Div. of Quantum Chem. Corp., 263 AD2d 335, 339 [3d Dept 2000]

The court cites this authority generally when concluding that the findings could not be reconciled on the evidence. Its role is supportive: it reinforces the need for a factual basis that permits apparently divergent liability determinations to coexist.

The holding is expressly limited. The court acknowledged that strict products liability and negligent design remain distinct theories and can yield different outcomes in appropriate circumstances. It did not establish that every defense verdict on negligence defeats strict liability, or that the two causes of action must invariably be merged.

Instead, the governing principle is evidence-sensitive: where the claims present the same design question on the same proof, the verdict sheet must not permit incompatible findings without a discernible factual distinction.

D. Implied Warranty: Substantial Overlap Does Not Eliminate Doctrinal Differences

Denny v Ford Motor Co., 87 NY2d 248, 256, 258-259 [1995]

This precedent explains both the substantial overlap and the difference between strict products liability and breach of implied warranty. It prevents the analysis from collapsing all product claims into a single legal theory merely because they concern the same product.

In conventional terms, design-defect strict liability employs risk-utility analysis, while merchantability concerns the product’s fitness and expected performance in customary, usual, and reasonably foreseeable use. These perspectives may overlap substantially without being identical.

Wojcik v Empire Forklift, Inc., 14 AD3d 63, 66 [3d Dept 2004]

Cited in accord with Denny v Ford Motor Co., this decision reinforces that overlap does not erase the different meanings of defect under the two theories. Its importance is limiting: the court’s ruling cannot properly be read as adopting BMW’s broad suggestion that all three claims are always synonymous.

On this record, however, the same airbag evidence also addressed ordinary fitness. The jury found the design not reasonably safe and not reasonably fit for its ordinary purpose, while rejecting negligent selection of that design. The court found no evidentiary basis reconciling those answers.

The retrial instruction is correspondingly functional rather than mechanical: the verdict sheet must account for the overlap. The court did not prescribe a complete replacement verdict form or direct that every element of all three claims be combined into one question.

E. The Comparative-Fault Contradiction Was Successfully Corrected

A separate issue illustrates the proper use of jury reconsideration. The jury initially assigned Janof 35% of the fault while finding that his negligence was not a substantial factor in causing his injuries. Supreme Court clarified its instructions, revised the verdict sheet, and returned the matter to the jury. The second verdict resolved that inconsistency.

Popolizio v County of Schenectady, 62 AD3d 1181, 1184 [3d Dept 2009]

Cited with CPLR 4111 (c), this precedent supports the corrective procedure used by Supreme Court. It distinguishes an inconsistency that has been cured through reconsideration from the unresolved contradiction among the design findings.

Thus, BMW did not succeed on every inconsistency argument. The appellate court assessed whether each alleged contradiction remained in the operative verdict.

F. Other-Accident Evidence Requires Substantial Similarity

Although the new-trial ruling made the remaining appellate issues academic, the court gave express evidentiary directions for retrial. Before admitting other incidents identified through the accident-sampling system discussed in the opinion, Supreme Court must require a proper foundation.

Daniels v New York City Tr. Auth., 35 NY3d 938, 939 [2020]

This authority supports the threshold requirement that other-incident evidence have an adequate foundation before admission. It anchors the court’s instruction that database identification alone does not establish admissibility.

Hyde v County of Rensselaer, 51 NY2d 927, 929 [1980]

This precedent supplies the governing substantial-similarity principle. Prior accidents may help prove a dangerous condition, causation, or notice, but the proponent must show that they occurred under circumstances substantially similar to the accident being litigated.

Johns v Crown Equip. Corp., 242 AD3d 1398, 1399 [3d Dept 2025]

This recent Third Department decision reinforces the substantial-similarity requirement in the court’s discussion of other incidents. Its relevance is to foundation, not to any suggestion that all accidents involving the same general product category are comparable.

O'Buckley v County of Chemung, 149 AD3d 1232, 1234 [3d Dept 2017]

This case provides further support for requiring meaningful similarity before other accidents can be used as proof. It contributes to the established evidentiary framework the trial court must apply on remand.

Cramer v Kuhns, 213 AD2d 131, 136 n, 137 [3d Dept 1995], lv dismissed 87 NY2d 860 [1995]

The court invokes this authority as part of the same foundational line of cases. Its influence lies in requiring a demonstrated connection between the other occurrences and the particular danger or defect at issue.

Facci v General Elec. Co., 192 AD2d 991, 993 [3d Dept 1993]

This precedent likewise supports substantial similarity as a condition of using prior accidents. Together, these authorities prevent superficially similar events from becoming misleading substitutes for proof about the challenged product.

The opinion makes the required showing concrete: the incidents must involve substantially similar airbag systems, and the injuries must have been caused by the same alleged defect. The relevant comparison is therefore not simply whether another driver suffered an eye injury during airbag deployment. It must address the system and alleged injury-producing mechanism.

The court did not categorically exclude the incident evidence. It required the trial court to determine admissibility on a properly developed foundation.

G. Expert Reliance on Hearsay Requires Professional Reliability

Delosh v Amyot, 186 AD3d 1793, 1796 [3d Dept 2020]

This decision supplies the professional-reliability exception: an expert may base an opinion on otherwise inadmissible hearsay when the material is shown to be of a type commonly relied upon in the profession. The opinion emphasizes that this professional practice must be established, not merely assumed.

Two distinct foundational questions therefore remain important. First, are the other incidents sufficiently similar to be probative? Second, if the expert relies on otherwise inadmissible reports, are those reports the kind of material professionally relied upon? Showing one does not automatically establish the other, and expert reliance does not create a blanket entitlement to admit every underlying report as substantive evidence.