Concept
The Preponderance Standard
Plaintiff's burden of proving negligence
The Preponderance Standard
A negligence plaintiff carries the burden of proving the case, and the earlier sections showed what that burden covers and how it divides into production and persuasion.1 They did not say how convincing the proof must be. A jury told only that the plaintiff must prove the elements cannot tell whether a strong suspicion will do, whether every doubt must be excluded, or whether the answer lies in between. The standard of proof closes that gap, and it decides who loses when the evidence leaves the jury unsure.
In negligence the standard is the preponderance of the evidence, the lowest standard used in civil actions.2 Two questions follow. The first concerns the line itself: what "more probable than not" demands, and what happens to a plaintiff whose proof lands exactly on it. The second concerns the unit to which the line applies, which is each element of the claim and not the claim as a whole.
More Probable Than Not
For each element, the jury must be persuaded that it is more probable than not that the alleged facts are true.3 The question is comparative, and credibility drives much of the answer. Whether the plaintiff meets the standard depends on whether the jury believed the plaintiff's testimony, what the defendant and other witnesses said, and whether the jury believed them.4
Evidence that is evenly balanced on an element, or in "equipoise," leaves the party who bears the burden of persuasion short of the line.5 A plaintiff whose evidence shows the breach of the standard of care to be only as probable as not has therefore failed on breach, while a plaintiff who persuades the jury that the defendant more probably than not breached it has carried that element.6 Any margin in the plaintiff's favor is enough, and no margin is a loss.
Courts sometimes call this the "51%" standard, which captures the idea that the plaintiff need only be past even.7 The label names where the line falls and does not describe how jurors work. In the typical case both sides put on evidence on each disputed issue, so the jury asks which side's contentions are likelier to be true, and it does not calculate probabilities.8
The line sits low because the work it supports is uncertain. A verdict reflects the inherent uncertainty of jury decision, and juries often decide on probabilities that fault really existed. In reconstructing events, a court often cannot be certain what happened or why. It can only assign a probability to each account, and a probability above 50 percent is more probable than not.9
Medical malpractice, where the standard is often stated this way, has also produced a separate theory of recovery. Oregon, for one, has recognized a limited loss-of-chance theory in common-law malpractice cases, in which the plaintiff alleges the loss of a chance at a recovery or better medical outcome.10 That theory describes a different claim from the one studied here, and the ordinary rule just described still governs a plaintiff who must prove that the defendant's conduct caused the harm itself.
Proof that the defendant's conduct could have caused the harm does not clear the line, because the proof must make that explanation the likelier one. Products cases show the difference. When a product malfunctions and injures someone, the doctrine the plaintiff relies on gives no help if the plaintiff leaves a likely alternative cause open: a grinding disc that may have been abused by a prior user, a used pickup truck driven over 50,000 miles by two prior owners, one of whom installed a spotlight on the door frame, or a toaster-oven that may well have lived out its useful life.11 The plaintiff's task is limited, though. A plaintiff must negate only the most likely alternative causes, and only by a preponderance of the evidence, and need not conclusively disprove every conceivable alternative theory.12
The demand for evidence runs against a defendant as well. A defendant who says the plaintiff's smoking and obesity caused difficulty healing a fracture must present adequate evidence that they did, and a mitigation defense needs proof that efforts to mitigate would have reduced the plaintiff's damages.13
Competing causes make the line easy to misjudge, because the largest single share of the evidence can look like a winning share. The test is whether the defendant's conduct is more likely than not the cause, which means more than all the other causes together.
The cyclist's lawyer would invoke Stubbs v. City of Rochester, which the discussion of actual cause covers. A trial court there concluded that no reasonable jury could find for the plaintiff, and the New York Court of Appeals disagreed. The word "reasonable" also appears in a second, unrelated sense: a defendant has actually caused the harm if the jury is "reasonably certain" the defendant caused it, even if the plaintiff has not actively disproved other potential causes.14
That argument is not likely to succeed here. In Stubbs the other potential causes were disparate and impossible to prove, so the court excused the plaintiff from disproving them, since requiring it would have prevented the plaintiff from seeking relief at all. The cyclist is responsible for all five of the other wrist injuries and is uniquely situated to disprove those causes.15 The excuse would not rescue this record in any event, because a jury cannot be reasonably certain of a cause that the plaintiff's own expert rates at 40 percent. A plaintiff needs evidence that makes the defendant's conduct the likelier explanation, and need not rule out every conceivable alternative.
Proof Element by Element
The preponderance standard attaches to each element and not to the claim as a whole. The jury is instructed that the plaintiff has carried the burden of persuasion only if it is persuaded by a preponderance of the evidence that the defendant breached the applicable standard of care, that the plaintiff suffered injury or loss, and that the breach caused it.17 The jury makes each finding on its own evidence, and the plaintiff wins when every answer is yes.
Probability theory raises an objection to that method. When several separate propositions must all be true, the chance that all are true is the product of their separate probabilities, which can be far smaller than any one of them. The source's illustration uses ten propositions: even if each were more than 90 percent likely, the chance that all ten were true would be substantially below 50 percent.18 On a smaller scale, three elements each found 60 percent likely would yield a combined probability of about 22 percent. Taken literally, the theorem says that element-by-element proof lets plaintiffs win claims that are probably false as a whole.
Courts nonetheless keep the theorem out of the burden of persuasion, for reasons that can be taken one at a time.19 In the typical case both sides submit evidence on each disputed issue, so the practical question is which side's contentions are likelier to be true. Once the party with the burden of production has met it, the jury's task is to assess the comparative persuasiveness of the parties' positions, not to calculate probabilities. The theorem also matters only when many separate issues go to the jury, and the source's example needs ten. The percentages in the objection are labels the objection supplies, because jurors do not generate them. An element the jury finds "60 percent likely" means that one side's account was the more convincing.
A student who has absorbed that the jury does not multiply may overcorrect and conclude that strength on one element can make up for weakness on another. The rule gives no such credit, because each element must clear even odds on the strength of its own evidence.
The two results work together. The jury does not combine findings that each lean toward the plaintiff into a total that defeats the claim, and it does not average a strong finding with a tied one to rescue the claim. Every element must independently be more probable than not, and the claim needs nothing more.