A good case evaluation answers a deceptively simple question: if you spend the money, what do you actually get? Everything else—the elements, the damages model, the statute of limitations, the collectibility memo, the decision tree—is just a structured way of refusing to answer that question by instinct. Instinct is how lawyers fall in love with bad cases and walk away from good ones.

The lawyers who get this right treat evaluation as a craft with its own tools and its own failure modes. They know the difference between a case that feels strong and one that is strong, between a defendant who owes money and one who can pay it, between a claim worth $2 million on a verdict form and a claim worth $40,000 once you net out fees, delay, and the 35% chance the jury hates your client. They run the same analysis whether they sit on the plaintiff's side or the defense side, because the math does not care which chair you occupy—it only cares about probabilities and dollars.

This guide lays out that craft end to end. We start where every case starts—intake, conflicts, and the engagement—then work through the substantive core: the elements of the claim, the burden of proof, the split between liability and damages, and the unglamorous but decisive question of collectibility. We tackle the threshold killers that end cases before the merits matter—the statute of limitations, personal jurisdiction, and venue. We translate strategy into money through fee arrangements and litigation budgets. And we close the loop by converting all of it into a number, using decision trees and expected-value analysis to find a settlement range you can defend to a client, a partner, or yourself at 11 p.m. the night before a mediation.

A note on perspective. Much of the classic guidance on case assessment is written for corporate defendants—the in-house department staring down a complaint, working backward from the client's goal, calibrating risk tolerance (Marcellus A. McRae & Kahn A. Scolnick, Case Assessment and Evaluation, Practical Law Practice Note 4-525-8907). That perspective is invaluable, and we borrow heavily from it. But the same engine drives plaintiff-side intake, where the lawyer is deciding whether to invest in a case rather than defend against one. We'll flag where the two diverge, because they diverge in instructive ways.

First contact: intake, conflicts, and the engagement

Before a lawyer evaluates anything, the lawyer has to be allowed to take the case at all. That means a conflicts check and an engagement letter, and skipping either is how careful firms generate malpractice claims and disqualification motions.

Run the conflicts check before you fall in love

The conflicts analysis comes first because it can be dispositive and because it gets harder to do honestly once you've invested in the matter. Under the Model Rules of Professional Conduct, a lawyer generally may not represent a client if the representation is "directly adverse to another client" or if there is a "significant risk" that the representation will be "materially limited" by responsibilities to another client, a former client, or a third person (Model Rule 1.7). Duties to former clients run through Model Rule 1.9, which bars representing a new client in the "same or a substantially related matter" adverse to the former client without informed, written consent.

Practically, that means running the proposed parties—and their corporate affiliates, principals, and key witnesses—against the firm's client and adverse-party databases at intake, not after the complaint is drafted. Many conflicts are waivable with informed written consent, but some are not (you cannot, for instance, sue your own current client in an unrelated matter without consent, and some adversities are non-consentable as a matter of state law). The cost of getting this wrong is not abstract: a missed conflict can produce disqualification mid-case, fee disgorgement, and a bar complaint. Build the check into the intake workflow so it cannot be skipped under deadline pressure.

Paper the engagement

Once conflicts clear, the engagement letter (or retainer letter) does the unglamorous work of preventing future fights with your own client. A well-drafted engagement letter defines who the client is (and, just as importantly, who it is not—the individual officer is usually not the client when you represent the corporation), describes the scope of the work, allocates the duties of lawyer and client, and sets out fees and expenses in writing (see Practical Law, Engagement (Retainer) Letter: Contingency Fee Arrangement, Std. Doc. w-000-0000-series). The letter "reduces the risk of misunderstanding" by ensuring a meeting of the minds before any work begins—and in fee-shifting or contingency contexts, written fee terms are frequently required by rule or statute.

Defining the client precisely is not a formality. It controls the conflicts analysis, the privilege analysis, and the question of whom you can later sue for your bill. When in doubt, name the client, name the matter, and disclaim representation of everyone else in writing.

The substantive core: can this claim win?

With the housekeeping done, evaluation turns to the merits. The organizing insight is that "winning" decomposes into several independent questions, each of which can sink the case on its own. A claim that clears the elements can still die on the statute of limitations. A liability slam-dunk can be worthless if the defendant is judgment-proof. Evaluate each gate separately; do not let strength on one dimension blind you to weakness on another.

Start with the elements and the burden of proof

Every cause of action is a checklist. To prevail, the plaintiff must prove each element of the claim by the applicable standard of proof; miss one element and the claim fails regardless of how sympathetic the facts are. The disciplined evaluator writes the elements down—pulled from the controlling statute, the pattern jury instructions, and the verdict form—and asks, element by element, what is the admissible evidence for this, and how good is it?

This is the purpose of the proof matrix, a tool borrowed straight from sophisticated case assessment practice. For each element of every claim, counterclaim, and defense, the matrix identifies (1) the witness who will testify to it, (2) the documentary or other evidence that supports it, and (3) the anticipated evidentiary hurdles or objections and your responses to them (McRae & Scolnick, supra). The matrix forces a brutal honesty that narrative advocacy lets you avoid. It is easy to say "the defendant breached the contract." It is harder to fill in the cell that asks which witness, with personal knowledge, will say so, and which exhibit proves it, and how does it come into evidence over a hearsay objection. Empty cells in the proof matrix are where cases are lost.

Consider a worked example. Hypothetical: Your prospective client, a regional distributor, says a supplier shipped defective components and refuses to make good. A breach-of-contract claim has four classic elements: (1) a valid contract, (2) the plaintiff's performance or excuse, (3) the defendant's breach, and (4) resulting damages. Element one looks easy—there is a signed purchase order. But the proof matrix exposes a problem in element two: the client took delivery, used the components for three months, and never gave the written notice of nonconformity that the contract required as a condition of any warranty claim. Under UCC § 2-607(3)(a), a buyer who accepts goods must notify the seller of breach within a reasonable time after discovery "or be barred from any remedy." Suddenly the "obvious" breach case has a serious performance problem, and the evaluation has to price the risk that the entire claim is forfeited on a notice technicality. That is the matrix earning its keep.

The burden of proof sits underneath all of this and is easy to take for granted. Most civil claims must be proved by a preponderance of the evidence—more likely than not, the "50% plus a feather" standard. But some claims demand clear and convincing evidence, a meaningfully higher bar: fraud in many jurisdictions, punitive damages in many states, the elements of certain equitable claims, and findings like willfulness that drive enhanced remedies. The standard of proof is not a footnote; it changes the value of the case. A fraud claim worth $1 million if proved by a preponderance is worth substantially less in expected value when the jury must be convinced by clear and convincing evidence, because your probability of winning drops. Build the correct standard into the assessment from the start.

Liability and damages are different questions—evaluate them separately

One of the most common evaluation errors is collapsing liability and damages into a single gut sense that a case is "strong" or "weak." They are independent variables, and a case can be strong on one and catastrophic on the other.

Liability asks whether the defendant is legally responsible—did the conduct happen, does it satisfy the elements, do the defenses fail? Damages asks, assuming liability, how much. The two multiply rather than add. A case with a 90% chance of establishing liability but a realistic damages range of $20,000 is a small case. A case with a 30% chance of liability but $10 million in provable damages is a large, risky case. Treat them as one number and you will systematically misprice your portfolio.

Damages analysis deserves its own rigor. For each category—compensatory (economic and non-economic), consequential, statutory, punitive—ask three questions: Is it available as a matter of law? Is it provable on this evidence? And is it capped? Many jurisdictions impose statutory caps on non-economic or punitive damages, and some causes of action carry their own caps or fee-shifting provisions that dramatically change the math (McRae & Scolnick, supra, on identifying punitive exposure, statutory penalties, and caps as part of issue-spotting). Punitive damages also raise constitutional ceilings: the Supreme Court has held that few awards exceeding a single-digit ratio of punitive to compensatory damages will satisfy due process (State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408 (2003)), so a client's dream of a 50-to-1 punitive verdict is usually just that.

Hypothetical: A plaintiff's lawyer evaluates an emotional-distress claim with searing facts and a sympathetic client. Liability looks strong. But the controlling state caps non-economic damages in this category at $350,000, the client has modest provable economic loss, and punitive damages require clear and convincing proof of malice the file does not currently support. The "strong" case has a realistic ceiling far below what the facts emotionally suggest. The caps did not change the wrong; they changed its price—and an honest evaluation says so before the client signs.

Collectibility: the question that turns paper into money

Here is the question that separates litigators from collectors: even if you win, can you actually collect? A judgment is a piece of paper authorizing you to try to take the defendant's assets. If the defendant has no reachable assets, no insurance, and no income, that paper is decoration. Plaintiff-side evaluators who skip the collectibility analysis are the ones who win at trial and then spend two years discovering the defendant was insolvent the whole time.

Collectibility analysis asks: Does the defendant have assets within reach of the forum? Is there applicable insurance coverage, and what are the policy limits? (In many disputes, the available insurance—not the defendant's net worth—is the practical ceiling on recovery.) Is the defendant an individual whose wages and home may be protected by exemptions, or a thinly capitalized LLC behind which the real assets sit? Are there senior creditors or security interests that will be paid first? Is there any prospect the defendant declares bankruptcy, in which case your judgment becomes an unsecured claim worth pennies on the dollar?

For corporate defendants, collectibility folds into a broader question of asset structure—holding companies, operating subsidiaries, and the difficulty of piercing the corporate veil to reach a parent's assets (see our discussion in corporate structuring and running multiple businesses and the offshore and domestic strategies covered in asset protection). A defendant who has deliberately structured assets to be hard to reach is telling you something about how hard collection will be.

The defense side runs the mirror image of this analysis. If your client is the deep pocket, collectibility is the plaintiff's leverage, and the absence of a collection problem means settlement pressure is real. If your client is judgment-proof, that fact is itself a settlement argument—why should a plaintiff spend two more years litigating for a paper judgment? Collectibility is not just a plaintiff's screening tool; it is a bargaining chip for whoever holds the better facts on it.

Threshold defenses: the cases that die before the merits

Some defenses do not contest whether the defendant did the bad thing. They contest whether the plaintiff can be heard to complain about it in this court at this time. Because these defenses can end a case before discovery, the disciplined evaluator screens for them first—a meritorious claim filed one day late is worth nothing.

The statute of limitations

The statute of limitations is the most efficient case-killer in the catalog, and the calendar is unforgiving. Limitations periods vary wildly by claim and jurisdiction—often two or three years for personal injury, four years for many UCC sale-of-goods claims (UCC § 2-725), varying periods for written versus oral contracts, and shorter or specialized periods for statutory claims. The evaluator must identify the correct period for each claim, because a single set of facts can support multiple causes of action with different clocks.

Two doctrines complicate the arithmetic and must be priced into the analysis. Accrual determines when the clock starts. The default is the date of injury, but the discovery rule—recognized for many latent-injury and fraud claims—starts the clock when the plaintiff knew or reasonably should have known of the injury and its cause. Tolling can pause the clock: for fraudulent concealment, for a plaintiff's minority or incapacity, during the pendency of a related proceeding, or by agreement (a tolling agreement, common where parties want to negotiate without forcing a protective filing). On the defense side, pre-suit notice statutes can compress the timeline still further; several federal and state laws require a claimant to give notice 30 days or more before filing, particularly against governmental entities (e.g., 28 U.S.C. § 2675(a) under the Federal Tort Claims Act; consumer-protection notice provisions like California's Consumers Legal Remedies Act, Cal. Civ. Code § 1782(a)). That notice window is a hazard for the plaintiff and an opportunity for the defendant—a chance to cure an alleged violation before suit and cut off damages and fee exposure (McRae & Scolnick, supra).

Hypothetical: A client comes in six weeks before the third anniversary of a contract breach, in a state with a three-year period for written contracts. The case looks fine—until the evaluator notices the agreement contains a contractual limitations clause shortening the period to one year, a clause courts in many jurisdictions enforce. The claim is already two years stale. Nothing about the merits changed; the contract quietly euthanized the case long before anyone called a lawyer. Read the limitations clause before you read the breach.

Because limitations is so often dispositive and so often missed, treat the deadline as the first calendar entry in any new matter, calculate it conservatively, and when in doubt about accrual, file early or get a tolling agreement. The malpractice carrier will thank you.

Personal jurisdiction, subject-matter jurisdiction, and venue

Even a timely, meritorious claim must be brought somewhere the court can hear it. Three related gates apply.

Subject-matter jurisdiction asks whether the court has power over the type of case. Federal courts are courts of limited jurisdiction; a plaintiff generally needs a federal question (28 U.S.C. § 1331) or diversity of citizenship plus more than $75,000 in controversy (28 U.S.C. § 1332). Get this wrong and the case can be dismissed—or remanded—at any time, even years in, because subject-matter jurisdiction cannot be waived. For businesses weighing the federal forum, our guide to federal civil litigation for small businesses walks through the jurisdictional threshold in depth.

Personal jurisdiction asks whether the court has power over the defendant. Since International Shoe Co. v. Washington, 326 U.S. 310 (1945), the constitutional test is "minimum contacts" such that suit does not offend "traditional notions of fair play and substantial justice." Modern doctrine splits this into general jurisdiction—available where the defendant is "essentially at home," typically its state of incorporation and principal place of business (Daimler AG v. Bauman, 571 U.S. 117 (2014))—and specific jurisdiction, which requires that the claim arise out of or relate to the defendant's contacts with the forum (Ford Motor Co. v. Montana Eighth Judicial Dist. Court, 592 U.S. 351 (2021)). For an out-of-state or foreign defendant, personal jurisdiction is frequently the first battleground, and where service must happen abroad it can become a strategy unto itself (see serving defendants in China under the Hague Service Convention).

Venue asks which court among those with jurisdiction is the proper or most convenient one. In federal court, venue is governed by 28 U.S.C. § 1391, and even a properly venued case can be transferred "for the convenience of parties and witnesses, in the interest of justice" under 28 U.S.C. § 1404(a). Courts weigh a familiar cluster of factors—the plaintiff's choice of forum (entitled to a strong presumption), convenience of parties and witnesses, location of documents and sources of proof, and the interest of justice—and the precise factors vary by circuit (see Practical Law, Motion to Transfer Venue Factors by Circuit Chart (Federal); Coady v. Ashcraft & Gerel, 223 F.3d 1, 11 (1st Cir. 2000)). Forum and venue choices are not neutral. They drive the jury pool, the speed of the docket, the substantive law (through choice-of-law rules), and the convenience cost of litigating far from home. Evaluating a case means evaluating where it will be fought, because the same facts can be worth materially different amounts in different courthouses.

The economics: fees, budgets, and the cost of the fight

A case is not just a legal proposition; it is a capital project with an uncertain payoff. Two things drive the economics from the lawyer's and client's side: how the lawyer is paid, and how much the litigation will cost to run.

Fee arrangements shape incentives and risk allocation

Who bears the risk of a long fight depends on the fee structure, and the fee structure changes how each side should evaluate the case.

The billable hour remains the default for much defense and commercial work, but its shortcomings are well documented: unpredictable cost, billing that reflects effort rather than value, and no built-in reward for efficiency (Practical Law, Alternative Fee Arrangements, Practice Note w-003-4295). For a defendant, hourly billing means litigation cost is an open-ended liability that must be modeled stage by stage.

The contingency fee—common on the plaintiff's side—flips the risk to the lawyer, who advances time (and often costs) in exchange for a percentage of recovery, typically in the one-third range, often escalating if the case goes to trial or appeal. A contingency arrangement is itself a market signal: a lawyer willing to take a case on contingency is staking real money on the evaluation, and a lawyer who declines is rendering a verdict. Contingency fees must be in writing under Model Rule 1.5(c), and the agreement must spell out how the fee is calculated and how costs are handled (see Practical Law, Engagement (Retainer) Letter: Contingency Fee Arrangement).

Between these poles sit a growing menu of alternative fee arrangements (AFAs)—flat fees, fixed fees per phase, capped or "collared" hourly fees, holdbacks, and success or bonus fees tied to outcomes (Practical Law, Alternative Fee Arrangements). AFAs exist to give clients predictable cost and to align the lawyer's incentives with the result rather than the hours. The choice among them is itself part of case evaluation: a flat-fee defense of a routine matter prices the risk one way; a partial-contingency defense fee that rewards beating a settlement target prices it another. For a broader map of who does what work and how they charge, see types of lawyers.

Build a stage-by-stage litigation budget

A strategy you cannot afford is not a strategy. The antidote is a realistic budget built phase by phase, because litigation costs are lumpy and back-loaded—discovery and trial dwarf the pleadings. Budget each stage separately: pre-filing investigation; pleadings (including motions to dismiss, removal, and motions to compel arbitration); fact discovery; expert discovery; class certification (if applicable); summary judgment; pre-trial; trial; post-trial motions; and appeal (McRae & Scolnick, supra).

For each stage, estimate the drivers: the number of witnesses to be deposed, the volume of documents to collect, review, and produce, the likelihood of contested motions, and the attorney hours each task requires. Then add the non-attorney costs that clients routinely underestimate—expert witness fees, e-discovery vendors and ESI specialists, jury consultants, mediators, and settlement administrators (id.). In many modern cases, document review and e-discovery are the single most expensive line item, which is why the litigation hold and the early ESI plan are budget items, not just compliance items. (Our [discovery refresher](/documents/a_practical_discovery_refresher---mastering_the_tools_rules_ and_pitfalls_of_federal_civil_litigation) and the broader paper trail guide to federal filings go deeper on what those phases actually entail, and depositions—a major cost center—get full treatment in the art and science of depositions.)

The budget is not a one-time document. It is the denominator in every expected-value calculation that follows, and it should be revisited at each decision point—because the right question is never "what has this cost so far?" but "what will the rest cost, and is the remaining upside worth it?"

Turning analysis into a number: decision trees and expected value

Now the pieces come together. You have probabilities (of clearing each element, of surviving each defense, of winning at trial), you have a damages range, and you have costs. The tool that combines them into a single, defensible number is the decision tree, and its currency is expected value.

How expected value works

Expected value (EV) is the sum of each possible outcome multiplied by its probability. The core insight is that a case is not worth its best-case verdict; it is worth the probability-weighted average of all outcomes, minus the cost of getting there.

Hypothetical (plaintiff side): You evaluate a claim with a realistic trial verdict of $1,000,000 if you win. Your honest assessment of the probabilities:

  • 30% the case is dismissed or lost on a threshold defense (statute of limitations is contestable): recovery $0.
  • Of the remaining 70% that reach a verdict, you estimate a 60% chance of a plaintiff's verdict.
  • If you win, damages realistically range from $600,000 to $1,000,000; use a midpoint of $800,000.

The probability of a winning verdict is 0.70 × 0.60 = 42%. The expected verdict value is 0.42 × $800,000 = $336,000. Now subtract costs: estimated litigation expense to verdict is $150,000, and you discount for collectibility—say the defendant's insurance and assets make recovery 90% collectible. The rough expected value is (0.42 × $800,000 × 0.90) − $150,000 = $302,400 − $150,000 = about $152,000. That number—not the $1,000,000 headline—is what the case is worth today, and it is the floor for any rational settlement discussion. If the defendant offers $200,000, a rational plaintiff strongly considers taking it, because the certain $200,000 beats the risk-laden $152,000 expected value, and certainty has value of its own.

The defense runs the identical tree from the other side. To the defendant, the expected cost of fighting is the probability-weighted exposure (0.42 × $800,000 = $336,000) plus its own defense costs (say $150,000), for a total expected cost approaching $486,000. The gap between the plaintiff's ~$152,000 net expected value and the defendant's ~$486,000 expected cost is the zone of possible agreement—the settlement range where both sides do better than litigating. Decision trees are, in the end, a machine for finding that zone.

Building the tree

A useful decision tree mirrors the litigation's actual structure. Each node is a decision point or an uncertain event—motion to dismiss (granted/denied), summary judgment (granted/denied), trial (win/lose), and damages (high/mid/low)—with a probability on each branch and a dollar value at each endpoint. Multiply along each path, sum across paths, and you get an expected value that updates as the case develops (see Practical Law, Case Assessment Decision Tree and Costs Worksheet, for a worked template). The discipline of assigning explicit probabilities is the point: it converts "I feel good about summary judgment" into "I'd put it at 40%," which is a number a client can weigh and a number you can revisit when the facts change.

Three honesty checks keep decision trees from lying to you. First, calibrate your probabilities—lawyers are chronically overconfident about their own motions, so pressure-test estimates against base rates and a skeptical colleague. Second, price the threshold defenses up front, because a 30% chance of an early dismissal does more to the EV than a 5% swing in the damages estimate. Third, fold in collectibility and time value: a $1,000,000 judgment collectible at 50% over three years on appeal is worth far less than its face, and the tree should say so.

Settlement value is a range, not a point

The output of all this is not a single "right" settlement number but a range, bounded by each side's expected value and shaped by factors the math captures only partially: the client's risk tolerance, the cost and reputational risk of public litigation, the precedent a settlement or verdict might set, and the simple value of certainty and closure. A start-up unaccustomed to litigation may rationally settle below its expected value to avoid distraction and preserve cash; a serial litigant worried about copycat plaintiffs may rationally fight a case it expects to settle, to avoid signaling weakness (McRae & Scolnick, supra, on calibrating risk tolerance). Non-litigation costs of losing—reputational harm, damaged business relationships, business interruption, and collateral consequences like debarment or parallel proceedings—belong in the analysis even though they resist precise quantification.

When the numbers point toward resolution, the next move is usually a well-built demand or a structured negotiation. A demand letter grounded in this analysis—one that signals you have done the math and will collect—carries far more weight than a bluster letter; see writing a demand letter basics for how to translate an evaluation into leverage, and arbitration and AAA-style mediation for the forums where many evaluated cases actually resolve.

Preserving the case while you evaluate it: the litigation hold

One evaluation step cannot wait for the analysis to finish: preserving evidence. The duty to preserve relevant information attaches as soon as litigation is "reasonably anticipated"—which is often before the assessment is complete and almost always before a complaint is filed (McRae & Scolnick, supra). The moment a matter looks litigable, counsel should issue and implement a written litigation hold: identify the custodians and data sources, suspend routine deletion and document-destruction policies, and notify former employees and third parties under the client's control of the preservation obligation.

The stakes are concrete. Federal Rule of Civil Procedure 37(e) governs the loss of electronically stored information that should have been preserved, authorizing curative measures for prejudice and, where a party "acted with the intent to deprive another party of the information's use," severe sanctions up to an adverse-inference instruction or dismissal. A spoliation problem can transform a winnable case into a losing one regardless of the merits, which is exactly why preservation is the one part of evaluation you do first and ask questions about later. Tracking the hold—who was notified, what was preserved, when—is itself a documented process (see Practical Law, Data Collection Tracking a Litigation Hold Template).

Special considerations by case type

The framework is general, but several categories of dispute have evaluation wrinkles worth flagging.

Intellectual property cases layer technical and procedural complexity onto the standard analysis. Patent cases turn on claim construction and often face parallel inter partes review at the PTAB, which can stay or reshape the district-court litigation; recent venue law has concentrated and then partly redistributed where patent suits can be filed. Trademark cases hinge on the strength of the mark and a multi-factor likelihood-of-confusion analysis, frequently aided by survey evidence. Copyright cases bring fair-use defenses and the strategic role of statutory damages, which can dominate the settlement math because they do not require proof of actual loss. Trade secret cases often open with a fight for emergency relief—TROs and preliminary injunctions—and the federal Defend Trade Secrets Act (18 U.S.C. § 1836) adds a federal forum and remedies to what was once purely state law. In each, the proof matrix has to account for expert testimony as a load-bearing element, not an afterthought.

Employment cases frequently carry administrative-exhaustion prerequisites (an EEOC charge before a Title VII suit, for example) that function like statutes of limitations and must be screened at intake, along with fee-shifting provisions that change the economics for both sides.

Class actions add a make-or-break node to the decision tree—class certification—whose outcome can swing the case's value by orders of magnitude and which deserves its own probability estimate, its own budget line, and early attention to whether a pre-certification settlement is achievable (McRae & Scolnick, supra).

Insurance-defense and indemnity matters route part of the collectibility and budget analysis through the policy: coverage position, reservation of rights, policy limits, and the carrier's settlement authority can matter more to the real-world outcome than anything on the merits.

A working checklist

The framework above can be compressed into a sequence the evaluator runs on every new matter. Treat it as gates, in roughly this order, stopping to reassess whenever a gate fails:

  1. Conflicts and engagement. Clear conflicts (Model Rules 1.7, 1.9); paper the engagement and define the client.
  2. Preserve. Issue a litigation hold the moment litigation is reasonably anticipated (Fed. R. Civ. P. 37(e)).
  3. Elements and burden. Build the proof matrix; identify the standard of proof for each claim.
  4. Liability and damages, separately. Estimate each as an independent probability/value; check availability, provability, and caps.
  5. Threshold defenses. Calculate the statute of limitations for each claim; confirm subject-matter and personal jurisdiction and proper venue.
  6. Collectibility. Identify reachable assets, insurance, and limits; price the recovery discount.
  7. Economics. Choose the fee structure; build a stage-by-stage budget including non-attorney costs.
  8. Decision tree and EV. Assign probabilities, compute expected value, and derive a settlement range and the zone of possible agreement.
  9. Risk tolerance and intangibles. Adjust for the client's appetite for risk, reputational stakes, precedent, and the value of certainty.

Frequently asked questions

What is the single most overlooked factor in evaluating a civil case? Collectibility, closely followed by the statute of limitations. Lawyers are trained to analyze the merits, so they reflexively evaluate whether a claim can win. They are less disciplined about whether a win pays—whether the defendant has reachable assets or insurance—and about whether the clock has already run. Both can render a meritorious case worthless, and both should be screened at intake, before anyone invests in the merits.

How do I put a probability on something as uncertain as a jury verdict? You estimate, you write it down, and you pressure-test it. The goal is not false precision; it is replacing an unexamined gut feeling ("I like this case") with an explicit number ("I'd put liability at 60%") that a client can weigh and that you can revise as facts develop. Calibrate against base rates, against outcomes in similar cases, and against a skeptical colleague's view. An honestly wrong number that you can revisit beats a confident feeling you can't examine.

Why evaluate liability and damages separately instead of just asking whether the case is "strong"? Because they multiply, not add, and a case can be lopsided. A 90%-liability case worth $20,000 is small; a 30%-liability case worth $10 million is large and risky. Collapsing them into one "strength" rating systematically misprices cases—you'll overvalue strong-liability/low-damages claims and undervalue weak-liability/high-damages claims. Keep them on separate axes until the final expected-value calculation, where you multiply them together.

What is a "zone of possible agreement," and why does it matter? It is the settlement range where both sides do better than litigating. Run the decision tree from each side: the plaintiff's net expected value sets a floor, and the defendant's expected cost of defending sets a ceiling. When the defendant's expected cost exceeds the plaintiff's expected value—which is common, because both sides pay their own lawyers and bear risk—there is a range in between where a deal beats trial for everyone. Finding that zone, and arguing about where in it to land, is most of what settlement negotiation actually is.

When does the duty to preserve evidence kick in—when I'm served? Earlier. The duty to preserve attaches when litigation is "reasonably anticipated," which is frequently before any complaint is filed and certainly before your evaluation is complete. That is why issuing a litigation hold is the first operational step in any new matter, ahead of the substantive analysis. Wait until you're served and you may already have a spoliation problem under Rule 37(e) that no merits argument can fix.

Should a contingency-fee lawyer's decision to decline a case tell me anything? Yes. A contingency lawyer who declines is, in effect, making a market in your case and refusing to buy. That can reflect liability problems, a damages ceiling, a collectibility problem, a limitations issue, or simply a portfolio decision—but it is data. It does not always mean the case is bad (the lawyer may be conflicted, capacity-constrained, or wrong), but a string of declines from competent contingency firms is a strong signal that the expected value does not justify the investment.

Conclusion: replace the gut with a method

The romance of litigation is the trial—the cross-examination, the closing, the verdict. The reality of litigation is the evaluation, done quietly months earlier, that determined whether the trial was ever worth holding. Cases are won and lost at the assessment stage far more often than in the courtroom, because that is where someone decided which cases to bring, which to fight, and which to settle on what terms.

The method laid out here—conflicts and engagement, then elements and burden, then liability and damages as separate variables, then collectibility, then the threshold defenses that end cases before the merits, then the economics, then the decision tree that turns all of it into a number and a range—is not a guarantee. Uncertainty is permanent; the most rigorous evaluation in the world cannot make a jury predictable. But the method does something more valuable than predict: it forces honesty. It fills in the empty cells of the proof matrix, prices the limitations risk you were tempted to ignore, and converts "I have a good feeling" into a number you can defend.

Do that consistently, on every matter, and you stop being the lawyer who falls in love with bad cases and walks away from good ones. You become the one who knows the difference—and can show the client, the partner, and the skeptical judge exactly why.

Related articles

This article is for general informational purposes only, is not legal advice, and does not create an attorney-client relationship. Statutes of limitations, jurisdictional rules, damages caps, and ethics rules vary by jurisdiction and change over time; consult qualified counsel about your specific situation.