A trial lawyer once described Rule 403 as the only rule of evidence with a thumb permanently resting on the scale and a shrug built into its grammar. The rule does not tell the judge what to do. It tells the judge that she may exclude evidence, that the dangers must substantially outweigh the value before she does, and then it steps back and lets her decide. Two careful judges can look at the same photograph, the same revenue chart, the same animated timeline, reach opposite conclusions, and both be affirmed on appeal. No other rule of evidence hands a single human being that much room.

That is exactly why Rule 403 is the most fought-over rule in an American courtroom. Most evidence rules answer a categorical, yes-or-no question: Is this hearsay? Is this witness competent? Is this signature authentic? Rule 403 asks something messier and more human—a question of degree and judgment. It assumes the evidence is relevant and useful, and then asks whether its capacity to inflame, confuse, mislead, or waste time so overwhelms that usefulness that the jury should never lay eyes on it. Rule 403 is, in a phrase, the rule that polices fairness. Learn how it works and you understand why a litigator will spend weeks before trial, and thousands of dollars, fighting over what a jury will be permitted to see.

This guide explains the mechanics of the 403 balance and then puts it to work on the evidence disputes that recur in commercial and especially trademark litigation: demonstrative exhibits and the brand-new Rule 107, evidence of a company's size and revenue, character and other-acts evidence, and a party's history of lawsuits. It is written to be useful to lawyers, to the business owners who pay them, and to anyone who wants to know why the most consequential decisions in a trial are often made in an empty courtroom long before the jury arrives. It takes no side: the analysis is the same whether you are trying to get evidence in or keep it out. And because 403 is one of several pretrial tools for shaping what the jury hears, it connects naturally to the preclusion of undisclosed evidence under Rule 37(c)(1) and to Daubert challenges that exclude unreliable expert testimony.

The Text, Word by Word

Rule 403 is short enough to quote in full, which is fortunate, because almost every word in it does heavy lifting:

"The court may exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence."

Read that sentence slowly, because four of its features structure the entire practice.

First, it applies only to relevant evidence. Relevance is the gate, governed by Rules 401 and 402: evidence is relevant if it has any tendency to make a fact of consequence more or less probable. Rule 403 is the filter that comes after that gate, not a substitute for it. If evidence flunks Rule 402 because it is irrelevant, it is simply out, and the court never reaches the 403 balance at all. A 403 objection is therefore a concession of sorts—it admits the evidence has some legitimate bearing on the case and argues that the bearing is not worth the danger. New litigators sometimes blur "irrelevant" and "prejudicial" into a single grumble; the rules keep them rigorously separate, and so should you.

Second, the balance is deliberately rigged toward admission. The drafters did not write "if the dangers outweigh the probative value." They wrote substantially outweighed. That adverb is the thumb on the scale. Evidence is not excluded because it is merely harmful, merely embarrassing, or merely a close call. It is excluded only when the listed dangers clearly and substantially predominate. Close calls go to the proponent. The Advisory Committee's note to the rule makes the point in the opposite direction too: in weighing probative value against the danger, "consideration should be given to the probable effectiveness or lack of effectiveness of a limiting instruction." Exclusion is the heavy artillery, deployed when lighter measures will not do.

Third, the rule is discretionary, and that single word—may—shapes how the whole game is played. Because the trial judge "may" exclude, appellate courts review 403 rulings only for abuse of discretion, one of the most deferential standards in American law. We will return to what that means for appeals, but the practical consequence starts at trial: outcomes are hard to predict in the abstract, and the framing of the dispute—how each side characterizes the probative value and the danger—often matters more than any abstract rule. The lawyer who can make the evidence sound either indispensable or gratuitous is doing the real work of a 403 motion.

Fourth, the rule lists six distinct dangers, not one. Unfair prejudice gets all the attention, but it shares the list with confusing the issues, misleading the jury, undue delay, wasting time, and needless cumulativeness. The last three are housekeeping concerns about the efficiency of the trial; the first three are concerns about its accuracy and fairness. A 403 motion may rest on any of them, and the strongest motions stack several—arguing, for instance, that a collateral matter is both unfairly prejudicial and a confusing, time-wasting sideshow. A judge who is unsure whether evidence is truly "unfair" may still keep it out because it threatens to consume a day of trial proving a point worth ten minutes.

"Unfair Prejudice" Is Not "Bad for My Case"

The most important—and most consistently misunderstood—idea in all of Rule 403 is hiding in the phrase "unfair prejudice." Clients and even lawyers use "prejudicial" loosely, as a synonym for "this hurts us." But by that measure all good evidence is prejudicial. The other side's best exhibit is supposed to hurt you; that is its job. If Rule 403 excluded everything damaging, trials would be bloodless recitations of stipulated facts. The rule is not bothered by ordinary, legitimate damage. It targets only unfair prejudice, a term of art with a precise meaning.

The Supreme Court supplied the governing definition in Old Chief v. United States, 519 U.S. 172 (1997)—the single most important 403 case, and one every litigator should be able to summarize from memory. Unfair prejudice, the Court explained, means "an undue tendency to suggest decision on an improper basis, commonly, though not necessarily, an emotional one." Id. at 180. Unpack that. Evidence is unfairly prejudicial when it tempts the jury to decide for the wrong reason—out of disgust, sympathy, hostility, or a bare urge to punish a defendant it dislikes—rather than on the legitimate strength of the proof on the actual issues. The line runs between evidence that persuades because it is probative (fair) and evidence that persuades because it is inflammatory (unfair).

A gruesome crime-scene photograph is the textbook illustration. It may genuinely prove a relevant fact—say, the location and nature of a wound—but if its overwhelming capacity to horrify the jury swamps that modest evidentiary contribution, a court may keep it out, not because it damages the defendant but because it threatens a verdict driven by revulsion rather than reason. Commercial cases supply quieter versions of the same dynamic: the email in which an executive boasts about crushing competitors, the reference to a company's foreign ownership, the gratuitous mention of a billionaire founder. Each may carry a sliver of relevance and a torrent of improper emotional appeal.

Old Chief did something subtler and even more useful: it taught us how to value the other side of the scale—the probative value—and the lesson surprises people. Probative value is not measured in a vacuum, as if the evidence were the only way to prove a point. It is measured against the proponent's genuine need for that particular item, "discount[ed]" by the availability of evidentiary alternatives. Id. at 182–83. If the same fact can be established by a less prejudicial route—a stipulation, a redacted version, a different document—then the incremental probative value of the inflammatory version shrinks, because the proponent does not actually need the inflammatory version to make the point. In Old Chief itself, a defendant charged with being a felon in possession of a firearm offered to stipulate that he had a qualifying prior conviction; the Court held it was an abuse of discretion for the government to insist on naming the prior offense (assault causing serious bodily injury) when the sanitized stipulation proved the same element with far less risk that the jury would reason "once a violent criminal, always a criminal."

That holding is the engine behind a move you will see constantly: the party facing prejudicial evidence offers to stipulate to the underlying fact. A well-placed stipulation can drain the dangerous exhibit of its incremental value and tip the 403 balance decisively toward exclusion. (It is not a magic wand—Old Chief expressly noted that a defendant cannot stipulate away the prosecution's right to tell a coherent, evidence-rich story on most issues—but on a discrete, status-type fact, the stipulation is powerful.) For both sides, the takeaway is the same: the 403 question is never just "how prejudicial is this?" It is "how prejudicial is this, and how badly does the proponent really need it given the alternatives?"

The Quieter Dangers: Confusion, Misleading, and Waste

Unfair prejudice hogs the spotlight, but in commercial and trademark litigation the other five dangers often do the actual work—and a motion that leans only on "this is prejudicial" is frequently the weaker motion. The reason is tactical. "Unfair prejudice" requires the judge to make a contestable, slightly philosophical judgment about what counts as an "improper basis." Confusion and waste of time, by contrast, are concrete and easy to credit. A judge protecting her own trial schedule needs no theory of jury psychology to see that a side dispute will eat two days for ten minutes of payoff.

Confusing the issues and misleading the jury bite hardest when evidence threatens to drag the trial into a collateral fight—a "trial within a trial." Evidence that forces the jury to resolve some unrelated question (the merits of a different transaction, the facts and outcome of a prior lawsuit, the validity of a tangential contract) diverts attention from the actual claims and invites a verdict built on a confused or mistaken premise. Complex financial and technical evidence carries its own brand of this danger: a number presented with false precision can mislead a jury into a confidence the data does not earn. Trademark cases are especially fertile ground, because they routinely sprawl into market segments, unrelated product lines, and survey methodology that can baffle as easily as it informs—which is one reason expert survey evidence draws both Daubert reliability challenges and ordinary 403 objections, and why courts pay close attention to consumer-survey expert methodology.

Undue delay, wasting time, and needless cumulativeness are the least glamorous grounds and among the most reliably successful, precisely because courts are fiercely protective of trial time and juror attention. The party determined to introduce its tenth document making the same point, or to spend an afternoon litigating a marginal collateral matter, invites exclusion not because the evidence is unfair but because its dwindling marginal value cannot justify the cost. The disciplined opponent pairs the prejudice argument with these efficiency dangers wherever they fit, handing the judge several mutually reinforcing reasons to exclude. The disciplined proponent does the reverse: streamline the proof, cut the cumulative exhibits, and deny the other side the argument in the first place. Nothing makes evidence look more gratuitous than the proponent's evident unwillingness to part with a single redundant copy of it.

The Motion in Limine: Where the Battle Is Actually Fought

Rule 403 can be invoked the old-fashioned way—an objection shouted from counsel table mid-trial—but that is rarely where the serious fighting happens. The natural home of a 403 dispute is the motion in limine: a pretrial motion (the Latin means "at the threshold") asking the court to rule in advance on the admissibility of evidence the movant expects the other side to offer. According to Practical Law's federal litigation materials, counsel routinely raise in limine the full menu of exclusion grounds—irrelevance under Rule 402, failure of authentication under Rule 901, defective lay or expert opinion under Rules 701–703, character evidence under Rule 404, subsequent remedial measures under Rule 407, compromise offers under Rule 408, hearsay under Rules 801–807, and, threading through nearly all of them, the Rule 403 balance.

Why front-load these fights? Because the danger Rule 403 guards against usually cannot be undone once the jury has heard the evidence. A jury that has seen the inflammatory photograph, or learned of the defendant's vast wealth, cannot truly unhear it, and a limiting instruction afterward can feel like asking the jury not to think about an elephant. Resolving the balance in advance, outside the jury's presence, keeps the bell from being rung. The motion in limine also confers strategic advantages the Practical Law checklists candidly catalog: it lets counsel present researched evidentiary arguments at leisure, narrows and streamlines the case, and can extract a preview of the judge's leanings before opening statements. The same materials flag the costs: a motion in limine tips your hand, signals which evidence you most fear, and gives your opponent advance warning and a chance to shore up the weak proof you hoped to ambush. A barrage of marginal motions can also irritate a judge into giving none of them the attention you wanted. File the motions that matter; resist the urge to file all of them.

A ruling on a motion in limine comes in two flavors, and the difference is not academic—it can decide whether you keep your appeal. A court may rule definitively (the evidence is in, or it is out, full stop) or provisionally (a preliminary ruling, expressly subject to revisiting at trial when the context is clearer, since the 403 balance often depends on how the evidence actually lands and what the proponent genuinely turns out to need). Under Rule 103(b), once the court rules definitively on the record—whether admitting or excluding—a party "need not renew an objection or offer of proof to preserve a claim of error for appeal." But where the ruling is merely provisional, you must raise the issue again at trial, or you forfeit it. The careful litigant pins down, on the record, whether each in limine ruling is definitive or provisional, and renews objections wherever Rule 103(b) leaves any doubt. More appeals die on this rock than on the merits.

The 403 motion in limine sits alongside the other great pretrial exclusion devices—preclusion for nondisclosure under Rule 37(c)(1) and Daubert exclusion of unreliable experts—as one of the principal ways a case is shaped before the first witness takes the stand. Whether that trial will be to a judge or a jury also matters enormously, because 403's force is greatest before a lay jury and considerably weaker in a bench trial, where the judge is presumed able to disregard prejudicial material she herself admits, a dynamic we explore in bench trial versus jury trial issues in trademark litigation.

Demonstrative Exhibits, Summaries, and the New Rule 107

Few categories of evidence generate 403 fights as reliably as demonstrative exhibits—the charts, timelines, summaries, animations, models, and graphics lawyers use to make complicated evidence legible to a jury. They are powerful because they are persuasive, and persuasive is precisely the danger: a demonstrative that distorts, exaggerates, or quietly smuggles in argument can mislead the jury under the innocent banner of merely "illustrating" the testimony. The visual sticks in the mind long after the cross-examination that should have undercut it.

A threshold distinction governs everything that follows. Some exhibits are substantive evidence; others are mere teaching aids, and the rules treat them very differently.

Summaries that ARE evidence (Rule 1006). Under Rule 1006, a party may admit a summary, chart, or calculation to prove the content of voluminous records that cannot conveniently be examined in court, provided the underlying material is itself admissible and is made available to the other side. A Rule 1006 summary is evidence—it goes to the jury room, and the jury may rely on it as proof of what the underlying records show. With that power comes a duty of fidelity: a Rule 1006 summary must accurately and non-argumentatively reflect the records it distills. (A December 1, 2024 amendment to Rule 1006 tidied up its language and confirmed that a proper summary may be admitted "to prove the content of voluminous admissible evidence," reinforcing that the underlying material must be admissible even if not itself admitted.)

Aids that are NOT evidence (the new Rule 107). Other exhibits are pedagogical or illustrative aids—devices used not as proof but to help the jury follow testimony or argument. A timeline drawn on a flip chart during closing, an enlargement of a key contract clause, a diagram a witness annotates while explaining a process: these illustrate, they do not prove, and historically they did not go to the jury room. For decades this category lived in a patchwork of local practice and scattered case law. As of December 1, 2024, it has its own rule. New Federal Rule of Evidence 107, "Illustrative Aids," provides that the court "may allow a party to present an illustrative aid to help the trier of fact understand the evidence or argument if the aid's utility in assisting comprehension is not substantially outweighed by the danger of unfair prejudice, confusing the issues, misleading the jury, undue delay, or wasting time." Read that and you will notice it is Rule 403 wearing a new hat: the same "substantially outweighed" balance, the same enumerated dangers, now aimed specifically at illustrative aids. Rule 107 also confirms that an illustrative aid is not admitted into evidence, that it generally must not accompany the jury into deliberations unless all parties consent or the court for good cause orders otherwise, and—crucially—that it is different from a Rule 1006 summary. The drafters drew the line the courts had blurred.

Under this framework, courts scrutinize demonstratives for accuracy and fairness. A chart that faithfully organizes admitted evidence is helpful and welcome; one that selectively emphasizes, dramatizes, color-codes for emotional effect, or editorializes about the evidence risks misleading the jury and will be excluded or sent back for a neutral rewrite. The law also distinguishes animations from simulations, and the distinction is consequential. An animation merely illustrates a witness's already-admitted testimony ("here is what the witness says happened, drawn in motion") and is treated relatively leniently. A simulation purports to recreate an event using scientific or technical modeling ("here is what the physics shows must have happened"), and because it makes a substantive, reliability-dependent claim, it requires an expert foundation under Rules 702 and 703—the same reliability gate at issue in any Daubert challenge. Courts commonly order pretrial disclosure of demonstratives so the Rule 107/403 balance can be struck calmly in advance rather than in the heat of trial, and exchanging exhibit lists is standard practice in federal civil cases.

The practical lessons fall out cleanly. For the proponent: build demonstratives that are accurate, neutral in tone, and tethered to admitted evidence; disclose them early; and resist the temptation to make them argue the case, because an argumentative aid is an excludable aid. For the opponent: demand disclosure, scrutinize every aid for distortion, selective emphasis, and smuggled argument, and invoke Rule 107 and Rule 403 against any aid that misleads more than it teaches.

Company Size, Wealth, and Revenue: The Deep-Pockets Problem

One of the highest-stakes 403 questions in commercial and trademark litigation is whether the jury may hear that a party is enormous and rich. The fear has a name—"deep-pockets" prejudice—and it is a textbook instance of the Old Chief "improper basis." A jury that learns one side is a multibillion-dollar conglomerate and the other a struggling family business may be tempted to decide liability, or to inflate damages, on the basis of who can better afford the loss, out of sympathy for the underdog or resentment of the giant. Robin Hood is a fine folk hero and a terrible juror. Deciding a case on relative wealth is deciding it for the wrong reason, the precise evil Rule 403 exists to prevent.

So the baseline rule is straightforward: evidence of a party's wealth, offered merely to suggest it should lose or pay more because it can afford to, is excludable under Rule 403, and courts routinely strike gratuitous references to a party's size, riches, or lavish headquarters when they serve no legitimate purpose. Standing alone, "they're a huge company" is an appeal to prejudice dressed up as a fact.

But—and this qualifier is where the real lawyering happens—evidence of size and revenue is often directly relevant to a legitimate issue, and where it is, exclusion is wrong (though a limiting instruction may be warranted to fence in how the jury uses it). Trademark cases are full of such instances:

  • Disgorgement of profits. The plaintiff seeking an accounting of the infringer's profits under 15 U.S.C. § 1117(a) must, as the statute provides, "prove defendant's sales only," whereupon the burden shifts to the defendant to prove deductible costs and apportion profits not attributable to the infringement. The defendant's infringement-related sales are therefore not just relevant—they are the plaintiff's burden and the literal starting point of the remedy. A defendant cannot use Rule 403 to suppress the very sales figures the law commands the plaintiff to prove, a dynamic we unpack in damages apportionment in trademark cases.
  • Reverse confusion. Where a large junior user floods the market and overwhelms a smaller senior user's mark, relative size is a substantive element of the theory—the whole point is that the bigger party's saturation swamped the smaller one. You cannot prove reverse confusion without proving the disparity in market presence.
  • Commercial strength and market power. The strength of a mark, and a party's position in the marketplace, bear on likelihood of confusion under the multifactor tests—the Polaroid factors in the Second Circuit chief among them—so evidence touching size and reach can be genuinely probative there too.
  • Willfulness. A defendant's resources and sophistication can bear on whether infringement was knowing and deliberate, which in turn affects remedies and the availability of enhanced recoveries and attorneys' fees.

One structural feature of trademark law sharpens the deep-pockets analysis in the plaintiff's disfavor. In many tort cases, a defendant's wealth is admissible because punitive damages are on the table, and a punitive award is calibrated in part to the defendant's financial condition—you cannot punish a billionaire and a pauper with the same dollar figure. But the federal Lanham Act does not authorize punitive damages. Its monetary remedies—actual damages, disgorged profits, and in exceptional cases attorneys' fees—are compensatory and equitable, not punitive; indeed § 1117(a) expressly provides that any enhancement of a profits award must constitute "compensation and not a penalty." So in a pure federal trademark case, the usual punitive-damages justification for parading the defendant's wealth before the jury is simply absent, which strengthens a 403 objection to wealth evidence offered for no other legitimate purpose. (Watch for the wrinkle: a plaintiff who joins a state-law unfair-competition or fraud claim that does allow punitive damages may reopen the door—another reason the choice of claims is itself an evidentiary strategy.)

Net it out and the rule is contextual rather than categorical. Wealth or size offered to inflame is excludable. Revenue and size offered to prove profits, market power, reverse confusion, or willfulness are generally admissible for those purposes, very often accompanied by a Rule 105 limiting instruction directing the jury not to decide liability based on the parties' relative wealth. The fight is rarely "in or out." It is "in, for what purpose, and with what instruction."

Character Evidence in Commercial Disputes

Litigants love to tell the jury that the other side is, at bottom, a bad actor: a serial infringer, a habitual corner-cutter, a company with a reputation for sharp dealing and broken promises. The rules of evidence stand squarely in the way, and they interlock with Rule 403 at every turn.

The foundational bar is Rule 404(a), which generally forbids character evidence offered to prove that a person—or a company—acted in conformity with that character on a particular occasion. The law distrusts the propensity inference—"they're the kind of company that infringes, so they probably infringed here"—because it invites exactly the Old Chief decision-on-an-improper-basis, and because it is both unfair and a poor proxy for what actually happened on the day in question. In a commercial dispute, evidence that a party has a bad general reputation, or is "the type" to cut corners, offered to prove it did so in this case, is inadmissible propensity evidence, full stop.

Rule 404(b) extends the principle to other acts—and then carves out the exception that does most of the real work in commercial litigation. Evidence of a party's other wrongs or acts is not admissible to prove propensity, but it is admissible for non-propensity purposes: to prove "motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident." That list is gold in trademark cases, where a party's prior similar conduct may be inadmissible to show bad character yet perfectly admissible to prove intent to infringe, knowledge of the senior mark, or a plan to free-ride. A defendant's pattern of adopting marks confusingly close to competitors' brands might be barred as "they're infringers by nature," yet admitted to prove that this defendant acted knowingly and in bad faith—a permissible purpose that bears directly on willfulness and, in turn, on remedies. (Worth knowing: a 2020 amendment to Rule 404(b) added a notice requirement in criminal cases, obliging the prosecution to articulate the permitted purpose in advance; the civil practice norm of pinning the proponent to a specific non-propensity purpose runs in the same spirit.)

Two further layers complete the picture. Rule 405 governs the methods of proving character on the rare occasions character is itself genuinely in issue. And the impeachment rules—Rule 608(b) (specific instances probative of a witness's truthfulness) and Rule 609 (impeachment by certain criminal convictions)—permit limited character-for-truthfulness attacks on witnesses, each subject to its own constraints. Rule 609 is itself shot through with 403-style balancing: convictions more than ten years old face a reverse-balancing test that presumes exclusion, and even fresh felony convictions of a witness who is a party in a civil case come in only if their probative value outweighs the prejudicial effect. The architecture repeats the master theme—relevance first, then a fairness balance.

And here is the unifying point that ties character evidence back to this entire guide: everything that survives the Rule 404 analysis must still run the Rule 403 gauntlet. Other-acts evidence offered for an impeccable non-propensity purpose can still be excluded if its genuine probative value for that purpose is substantially outweighed by the risk that the jury will quietly misuse it as propensity evidence anyway—reasoning "knowing or not, they're clearly bad people." Courts manage that risk with limiting instructions under Rule 105 confining the evidence to its permissible use, and with the discretionary power to exclude when no instruction can cabin the danger. The recurring "serial infringer" fight in trademark litigation is resolved at exactly this junction: such evidence is barred as propensity, but may come in—with a limiting instruction and subject to 403—where it genuinely proves intent, knowledge, or willfulness rather than mere bad character. The whole sequence is relevance → 404 → 403, and skipping the last step is a common and costly mistake.

Prior Litigation and Other Lawsuits

Close cousin to character evidence is the question of a party's litigation history—prior lawsuits, claims, settlements, and judgments. Litigants are forever eager to paint an opponent as litigious, a serial defendant, a courthouse regular who must be guilty of something. The rules and Rule 403 combine to make that portrait genuinely hard to hang in front of a jury.

Start with the obvious problem: evidence of other lawsuits offered to suggest a party is litigious, or that it must be a wrongdoer because it has been sued (or has sued) before, is classic propensity evidence, inadmissible under Rule 404—and it drags a freight train of Rule 403 dangers behind it. It invites a verdict on an improper basis, and it threatens to spawn mini-trials: satellite litigation over the facts, the merits, and the outcomes of the prior cases, each of which the jury would have to understand and evaluate, none of which is what they were summoned to decide. That is confusion of the issues and waste of time in their purest form. Courts therefore routinely exclude evidence of unrelated prior litigation, especially when it is offered for nothing grander than casting the opponent in a bad light.

As with other-acts evidence generally, though, prior litigation can sometimes squeeze through for a specific, permissible purpose. A prior dispute may show a party's notice or knowledge—for example, that a defendant knew of the plaintiff's mark precisely because the two had tangled before. A genuine pattern of similar adjudicated conduct may bear on intent or willfulness. Prior litigation may also be admissible to rebut a position a party has affirmatively taken ("we've never been accused of this," met with proof that they have). Even then, admission is never automatic: the court must find the evidence genuinely probative of the permissible purpose and then run the 403 balance, weighing that probative value against unfair prejudice, jury confusion, and the time and distraction of litigating the prior matter. In practice the bar is high—a concrete non-propensity purpose, real probative value, and a danger that does not substantially outweigh it—and a limiting instruction will almost always accompany any admission.

A separate rule slams the door on one important subcategory. Rule 408 bars evidence of compromise offers and negotiations—settlement discussions, and the fact or amount of a settlement—when offered to prove or disprove the validity or amount of a disputed claim, or to impeach by prior inconsistent statement. The policy is candor: parties will not negotiate frankly if their concessions can be paraded later as admissions of liability. So even where the existence of prior litigation might clear the 404/403 hurdles, its settlement is independently protected from use to prove liability. (Rule 408 has its own carve-outs—settlement evidence can come in for "another purpose," such as proving a witness's bias—but those are narrow and policed by, you guessed it, Rule 403.) The combined effect is a steep climb: a party's litigation history is barred as propensity, dangerous under 403, and—as to settlements—separately walled off by Rule 408, admissible only for a narrow permissible purpose that survives the balance.

Abuse of Discretion: Why 403 Rulings Almost Never Get Reversed

We promised to return to that little word may, and the appellate consequences it carries. Because Rule 403 commits the balance to the trial judge's discretion, appellate courts review 403 rulings for abuse of discretion—and this is where many a losing party's hopes go to die. Under that standard, the question on appeal is not whether the appellate judges would have struck the balance the same way. It is whether the trial court's ruling fell outside the range of permissible choices—whether it rested on an error of law, a clearly erroneous fact, or a decision so arbitrary that no reasonable judge could have reached it. Two judges can weigh the identical evidence and come out opposite ways, and both rulings can be affirmed, because both lie within the broad zone of reasonable judgment the rule confers.

The practical implications are large and frequently underappreciated. First, 403 disputes are won and lost at trial, not on appeal. If you are betting your case on getting a 403 ruling reversed, you are betting on a long shot; pour your energy into persuading the trial judge, where the real leverage lies. Second, the record is everything. A trial court that conducts the balance explicitly—naming the probative value, naming the danger, explaining why one substantially outweighs the other—builds an affirmance-proof ruling, while a court that excludes or admits with a curt "sustained" gives the appellant more to work with. Smart advocates invite the judge to make findings on the record, especially when they win, precisely to insulate the ruling. Third, even a genuine 403 error may be harmless. An appellate court that finds the trial judge erred will still affirm if the error did not affect substantial rights—if the evidence's admission or exclusion likely made no difference to the verdict. Preserving the issue (recall Rule 103(b) and the definitive/provisional distinction) is necessary but not sufficient; you must also be able to show the error mattered.

This deferential posture is not unique to evidence law, but it shapes appellate strategy in commercial and trademark cases generally. The mix of deferential and de novo review standards across different rulings—facts reviewed for clear error, evidentiary calls for abuse of discretion, legal questions de novo—is itself a strategic terrain we map in Second Circuit appellate standards in trademark cases. The headline for 403 is simply this: the rule's discretionary design means the trial is the main event, and the appellate court is rarely going to rescue you from a balance you lost downstairs.

The 403 Toolkit: Everything Short of Exclusion

A defining and often-missed feature of Rule 403 practice is that exclusion is the last resort, not the first. Good judges—and good advocates—reach for less drastic measures before banishing evidence entirely, and understanding that toolkit frequently determines how a 403 dispute actually resolves. The fight is usually not "in or out" but "in what form and with what safeguards."

The limiting instruction (Rule 105) is the workhorse. The court admits the evidence but instructs the jury on the narrow purpose for which it may be considered and the purposes for which it may not, trusting the jury to follow the instruction—a trust the law generally extends, with the occasional candid acknowledgment that some bells, once rung, are hard to unring. Rule 105 entitles a party, on request, to such an instruction, so the proponent who proposes one and the opponent who demands one are each playing the rule correctly.

Redaction surgically removes the prejudicial portion of an otherwise-admissible document, admitting the probative remainder—the email with the inflammatory boast excised, the financial summary with the gratuitous net-worth line struck. The stipulation, the Old Chief move, lets the opponent concede the underlying fact and thereby drain the prejudicial evidence of its incremental value, tipping the balance toward exclusion of the more inflammatory version. Bifurcation of the trial—separating liability from damages, as discussed in our treatment of bench versus jury trial issues—can keep prejudicial damages evidence, such as a party's wealth, entirely away from the jury during the liability phase, neatly dissolving the deep-pockets problem by sequencing rather than suppression.

For the advocate, the existence of this toolkit dictates strategy. The party seeking to admit evidence over a 403 objection should arrive with the least prejudicial form already prepared and a limiting instruction already drafted, making it easy for the court to admit rather than exclude—because a judge handed a clean, narrow, safeguarded version of the evidence has little reason to keep it out. The party seeking exclusion must do the opposite: anticipate each alternative and explain why it fails—why no instruction can cure this prejudice, why redaction cannot sanitize this exhibit, why the proponent's offered stipulation fully satisfies its legitimate need so the inflammatory version adds nothing but heat. The lawyer who argues only "exclude it" while ignoring the middle-ground options is arguing against the grain of the rule.

A Worked Example (Hypothetical)

Make it concrete with an invented case. This is a hypothetical for illustration only. "Beacon Coffee," a small regional roaster, sues "Brightwater Beverages," a national conglomerate, for trademark infringement, seeking actual damages, disgorgement of Brightwater's profits, and an injunction. As trial nears, the parties file dueling motions in limine, several resting on Rule 403.

Beacon's offerings. Beacon wants the jury to hear that (1) Brightwater earns roughly $20 billion in annual revenue; (2) Brightwater has been sued for trademark infringement three times before; and (3) a former Brightwater executive once called the company "ruthless" in a deposition. Beacon also wants to deploy a slick animated demonstrative dramatizing how Brightwater's product "swallowed" Beacon's market. Brightwater moves to exclude all of it. Run the analysis:

  • Revenue. The bare $20 billion figure, offered to suggest Brightwater should pay because it is rich, is excludable deep-pockets prejudice—gratuitous and aimed at an improper basis. But Brightwater's infringement-related sales are directly relevant to Beacon's profits claim under § 1117(a) and must be admitted for that purpose, almost certainly with a Rule 105 instruction that the jury not decide liability on the parties' relative size. Expect the court to separate the legitimate sales figures from the inflammatory "billions" framing—admitting the former, striking the latter.
  • Prior lawsuits. The three earlier infringement suits, offered to show Brightwater is "the type" to infringe, are barred as propensity under Rule 404 and dangerous under 403 (mini-trials, confusion, waste). But if one of those suits genuinely put Brightwater on notice of Beacon's mark, or shows a pattern probative of willfulness, that narrow slice might come in for that defined purpose, subject to 403 and a limiting instruction—while any settlements of those suits are independently barred by Rule 408.
  • The "ruthless" comment. Pure character/propensity evidence offered to show bad character. Almost certainly excluded under Rules 404 and 403. It proves nothing about whether infringement occurred and does everything to invite a verdict on attitude.
  • The animated demonstrative. Under new Rule 107 and Rule 403, the court asks whether the animation fairly illustrates admitted evidence or instead dramatizes and argues. "Swallowed the market" sounds argumentative and emotionally loaded; expect the court to exclude it or order a neutral, accurate revision tethered to the actual evidence.

Now flip the table. Brightwater wants to prove that Beacon has filed numerous trademark suits against other coffee companies, painting Beacon as a litigious bully who weaponizes the courts. Beacon moves to exclude. The analysis is the mirror image: offered to show Beacon is litigious, the evidence is propensity-based and freighted with 403 danger; it comes in, if at all, only for a specific permissible purpose (perhaps to rebut Beacon's claim that it is a passive, reluctant litigant) and subject to the balance and a limiting instruction. The symmetry is the lesson—the same categories of evidence, the same rules, cut against both sides, and Rule 403 (working in tandem with Rules 404, 408, and 107) is the mechanism that sorts the genuinely probative from the merely prejudicial no matter who is offering it.

Frequently Asked Questions

Is Rule 403 a relevance objection? No, and conflating the two is a classic error. Relevance is decided first, under Rules 401 and 402. Irrelevant evidence is simply inadmissible and never reaches Rule 403. A 403 objection concedes the evidence is relevant and argues that its danger substantially outweighs that relevance. They are different objections aimed at different stages of the inquiry.

What does "unfair prejudice" actually mean? Per Old Chief v. United States, 519 U.S. 172, 180 (1997), it means "an undue tendency to suggest decision on an improper basis, commonly, though not necessarily, an emotional one." It does not mean "harmful to my case." Almost all good evidence harms one side; that is legitimate. Rule 403 targets only the evidence that tempts a verdict for the wrong reason—revulsion, sympathy, hostility, or a bare urge to punish.

Does offering to stipulate really keep evidence out? Sometimes, and powerfully. Old Chief teaches that probative value is discounted by the availability of less prejudicial alternatives, so a stipulation that concedes the underlying fact can drain inflammatory evidence of its incremental value and tip the balance toward exclusion. But it is not absolute—courts will not let a stipulation gut a party's right to tell a coherent, evidence-rich narrative on contested issues. It works best on discrete, status-type facts.

Can a jury ever hear how much money a company makes? Yes, when the figure is genuinely relevant to a real issue—most clearly the defendant's sales as the starting point for disgorgement of profits under 15 U.S.C. § 1117(a), and also reverse confusion, market strength, and willfulness. What gets excluded is gratuitous wealth evidence offered only to suggest a party should pay because it can afford to. Because the Lanham Act authorizes no punitive damages, the usual punitive justification for parading wealth is absent in a pure federal trademark case.

Can I prove the other side is a "serial infringer"? Not as propensity ("they're infringers by nature, so they infringed here")—that is barred by Rule 404(a). But under Rule 404(b) you may offer prior similar acts for a non-propensity purpose such as intent, knowledge, plan, or absence of mistake, which often translates into willfulness in trademark cases. Even then it must clear Rule 403, and a limiting instruction will accompany any admission.

How do the new Rules 107 and 1006 change demonstratives? Rule 1006 summaries are evidence—they prove the content of voluminous admissible records and can go to the jury room, so they must be accurate and non-argumentative. New Rule 107 (effective December 1, 2024) governs illustrative aids—teaching devices that are not evidence, generally stay out of the jury room, and are admitted under a 403-style balance (utility versus the danger of prejudice, confusion, misleading, delay, or waste). The first proves; the second merely helps the jury understand.

Why don't 403 rulings get reversed on appeal more often? Because they are reviewed for abuse of discretion, one of the most deferential standards in the law. The appellate court asks not whether it would have ruled the same way but whether the trial judge's balance fell outside the range of reasonable choices. Win the 403 fight at trial; do not count on the court of appeals to fix it. And remember Rule 103(b): preserve the issue, and be ready to show that any error actually affected the verdict.

Practical Takeaways

For the proponent of evidence, the goal is to make admission the easy choice. Tie the evidence to a concrete, legitimate purpose—an element, a permissible non-propensity use, a genuine need you can articulate in a sentence—because vague, hand-waving relevance invites exclusion. Offer the least prejudicial form: redact what you do not need, draft a limiting instruction and hand it to the court, and resist the temptation to gild probative evidence with inflammatory framing that gift-wraps a 403 argument for your opponent. Anticipate a stipulation offer and ask honestly whether your need survives it. Disclose demonstratives early and build them accurate and neutral so they clear the Rule 107/403 balance. Clean, purpose-tethered evidence with built-in safeguards usually gets in.

For the opponent of evidence, the motion in limine is your instrument and the framing is your craft. Surgically separate the evidence's legitimate probative value from its capacity to inflame, and show that the latter substantially predominates—invoking not just unfair prejudice but, where they fit, confusion, mini-trials, and waste of time, because those concrete dangers are easier for a judge to credit than a contestable theory of jury psychology. Use Old Chief: offer to stipulate to the underlying fact and drain the inflammatory version of its incremental value. Argue affirmatively why each less-drastic alternative fails—why no instruction can unring this particular bell. For wealth, separate gratuitous size references (excludable) from genuinely relevant revenue (admissible for a defined purpose with an instruction). For character and prior litigation, pin the proponent to a real non-propensity purpose and stack Rules 404 and 408 alongside 403. And track whether each ruling is definitive or provisional, renewing objections wherever Rule 103(b) leaves room for doubt.

For both sides, the unifying principle is the one we opened with: Rule 403 is about fairness, not harm. Relevant evidence is not excluded because it hurts—of course it hurts—but because it risks a decision on an improper basis or a trial derailed by confusion and waste, and only when that risk substantially outweighs the evidence's genuine value. The balance is discretionary, weighted toward admission, reviewed deferentially, and almost always litigated through motions in limine using a toolkit of measures short of exclusion. Understand what unfair prejudice really means, weigh probative value against the proponent's true need, reach for limiting instructions, redaction, stipulations, and bifurcation before exclusion, and the most pervasive rule in all of litigation becomes a precision instrument rather than a blunt one.

Related Articles

This article is provided for general informational purposes and does not constitute legal advice. The application of Rule 403 is discretionary and intensely fact-specific; consult qualified litigation counsel about any particular matter.