What Baseball Got Wrong—And What Lawyers Can Learn From It
In the early 2000s, Billy Beane of the Oakland A’s transformed baseball by doing something radical: he questioned the judgment of experts. For decades, Major League Baseball teams relied on scouts who evaluated players based on instinct, experience, and a deeply ingrained lore. They made decisions by watching players swing a bat, run the bases, or throw a pitch—listening to the crack of the ball off the bat, watching how a player “carried himself,” or noting whether he had a “good baseball body.” Scouts favored tools like raw speed, arm strength, and athleticism, often invoking subjective labels such as “projectable,” “gritty,” or “five-tool talent.” These judgments were not systematically validated, and many were driven by bias—favoring players who looked the part, spoke the part, or reminded them of past stars.
But when Beane and his team looked at the data, much of this conventional wisdom didn’t hold up. The attributes scouts prized turned out to be only partially predictive of success—and in some cases, entirely irrelevant. What mattered more was consistent, measurable output: how often a batter got on base, how reliably a pitcher avoided walks, or how many runs a player created over the course of a season. Metrics like on-base percentage (OBP), slugging percentage, and walks—long undervalued by traditional scouts—proved to be more reliable indicators of a player’s contribution to winning games. Beane’s insight wasn’t just that the old system had blind spots, it was that the entire scouting paradigm was anchored in flawed judgment, blind to more predictive, evidence-based measures.
Litigators face a similar challenge. Every day, we’re asked to predict trial outcomes, assess risk, and advise clients—often under pressure, and always facing uncertainty. And yet, despite decades of behavioral science revealing the limits of human judgment, most lawyers never encounter this research. It’s not part of the law school curriculum, and the demands of legal practice—billable hours, client needs, and the urgency of the next deadline—leave little time to explore it. Unless a lawyer goes looking for it, the insights of cognitive psychology, decision science, and behavioral economics remain hidden in academic journals, even as they offer powerful tools for improving legal judgment.
And it’s not just lawyers. Clients, too, are prone to well-documented psychological traps. Litigants routinely fall victim to overconfidence, loss aversion, confirmation bias, and the sunk cost fallacy. They anchor on inflated expectations, discount uncertainty, and interpret fairness through an egocentric lens. These distortions don’t always arise in moments of crisis or conflict—they shape decisions quietly, often without the client realizing it. Even sophisticated business clients, who approach disputes calmly and rationally, are not immune. Bias operates beneath the surface, subtly steering risk assessments, settlement positions, and litigation choices in ways that feel reasoned but are not always sound.
There is another way. Just as Beane used structured analysis to challenge baseball’s gut-driven culture, lawyers can apply decision science to improve clarity and reduce error. One method in particular—decision trees—offer a practical framework for navigating uncertainty. They help lawyers organize assumptions, quantify risks, visualize outcomes, and align strategy with client values.
This article explores how behavioral psychology and decision trees can help lawyers think more clearly in litigation. We begin by surveying the psychological traps that distort legal judgment. We then introduce evidence-based strategies to overcome them, culminating in a detailed look at decision trees as a toolset for making smarter, more transparent decisions in high-stakes legal practice.
What Gets in the Way of Good Judgment: Being Human
Lawyers are trained to analyze facts, apply rules, and make persuasive arguments. But they’re not trained to spot the quiet, internal forces that shape their judgment—or their clients’. Those forces aren’t flaws. They’re just part of being human.
All of us rely on mental shortcuts—known as heuristics—to make decisions under uncertainty. These shortcuts aren’t always bad. Often, they help us make quick calls when time is short or patterns are familiar. But litigation presents a different environment. The stakes are high, the facts are incomplete, emotions run hot, and outcomes are unclear. In that context, those same shortcuts can lead to predictable mistakes.
And they don’t happen at random. In legal practice, certain patterns show up again and again:
- Anchoring: Early numbers—like opening offers or prior verdicts—exert unconscious influence on how parties assess value, even when the numbers are extreme or irrelevant.
- Overconfidence: Lawyers and clients alike routinely overestimate the strength of their own case and underestimate the chances of loss.
- Confirmation bias: We give extra weight to evidence that supports our existing view and downplay facts that challenge it.
- Naïve realism: We believe we’re seeing the facts objectively—while assuming the other side is blinded by bias or self-interest.
- Reactive devaluation: Offers or ideas are reflexively dismissed simply because they come from the opposing party.
- Egocentric bias: Each side tends to interpret fairness in ways that conveniently align with its own interests.
- Availability heuristic: Recent or emotionally vivid outcomes—like a shocking jury verdict—can distort risk perception far out of proportion to their statistical relevance.
- Affect heuristic: Strong emotions, whether about the opposing party, the case, or a prior interaction, unconsciously color judgments of legal strength or value.
- Sunk cost fallacy: The more time, energy, or money has been invested in litigation, the harder it becomes to walk away—even when continuing is no longer wise.
- Framing effects: Logically equivalent information can lead to very different decisions depending on how it’s presented (e.g., “70% chance of success” vs. “30% chance of failure”).
- Probability neglect: When emotionally loaded outcomes are at stake—like public embarrassment or total vindication—people often ignore or misjudge the odds.
- Choice overload: When clients are presented with too many options, they can become paralyzed or make decisions based on irrelevant factors.
- Loss aversion: Plaintiffs often reject settlements that feel like a loss, even if they’re financially sound, and defendants may over-fight to avoid payouts that are smaller than the cost of continued litigation.
- Myside bias (the advocate’s illusion): Even when reviewing the same set of facts from both sides, lawyers still tend to see their own client’s case as stronger.
(Appendix A defines each of these and gives examples specific to litigation, along with references for further reading.)
Take a business dispute heading into mediation. The plaintiff opens with a demand that’s far higher than any likely trial result. That number, though extreme, creates an anchor—not just in the negotiation, but in the plaintiff’s own sense of what the case is worth. Later, the defense makes a creative and potentially fair proposal. But the plaintiff rejects it immediately—not because it’s unfair, but because it came from “them.” That’s reactive devaluation.
Both sides have reviewed the same evidence. Both are experienced. Yet each is convinced it will win at trial. That misplaced confidence reflects both confirmation bias and myside bias: people give more weight to facts that support their view, and they naturally find their own arguments more persuasive. Lawyers aren’t immune. Even while trying to advise objectively, they can fall into the same traps.
As the mediation drags on, the parties dig in. Offers that might have made sense months ago are now off the table—not because the facts changed, but because too much has already been spent. Walking away feels like failure. That’s the sunk cost fallacy: the pressure to continue, simply because of how much has already gone into the case.
And as offers move back and forth, something deeper starts to happen. Parties don’t just weigh numbers—they react to what those numbers mean. For the plaintiff, anything below the original demand feels like backing down. For the defendant, any payout feels like admitting defeat. Even if the proposals are financially reasonable, they feel like losses. That’s loss aversion: the tendency to experience losses more strongly than equivalent gains.
At that point, no one is doing cost-benefit analysis. They’re trying to avoid pain. And in doing so, they may pass on good outcomes—just because those outcomes don’t feel right.
That doesn’t make anyone irrational. These are normal, invisible patterns—especially in stressful situations. The problem isn’t that we rely on shortcuts. The problem is that we don’t notice when those shortcuts start to steer us wrong.
The solution isn’t to get rid of intuition. It’s to build in tools that help us test it—tools that bring structure to complex decisions and make hidden assumptions easier to see. One of the most effective tools for that is the decision tree.
A. Before We Dive into the Antidote, We Should Ask Whether We Mind Being Sick
Bias is part of being human. Our brains are built to take shortcuts—anchoring, availability, overconfidence—because they’re efficient. These heuristics help us get through the day without overanalyzing every decision. If we paused to fully evaluate every choice, we might still be deciding what to wear—or what to eat for lunch.
And in many situations, that works just fine. A little distortion in our thinking may not matter much. The stakes are low, the decisions are routine, and instinct gets us where we need to go. Some clients—especially those with more distance from the case—may feel comfortable making litigation decisions the same way. And that’s their call.
But for me, this is where it starts to matter. Litigation decisions aren’t low-stakes. They’re expensive, disruptive, and often final. They involve real pressure, real money, and real emotional weight. And they’re made in exactly the kind of environment—under stress, with imperfect information, and through the filter of advocacy—where these mental shortcuts are most likely to lead us off course.
The challenge is that bias doesn’t feel like bias. It feels like certainty. It feels like good judgment. It feels like clarity. But those feelings can be misleading. And when a client is facing a decision that could reshape their business or their life, I believe they deserve more than that.
They deserve the chance to make the most educated decision possible. That means understanding how stress, alignment, and story might be shaping their view. It means having space to notice their assumptions—not to erase them, but to weigh them with care. If they still want to trust their instinct, that’s fine. But in my view, it should be a choice—not an accident.
B. Techniques for Reducing the Influence of Bias
Whether or not lawyers know the names of the biases—anchoring, overconfidence, confirmation bias—they’ve seen them in action. They’ve watched clients walk away from reasonable offers because they were stuck on unrealistic numbers. They’ve seen opposing counsel express total confidence in a weak case. They’ve even felt it themselves—moments where something felt off but was hard to explain. That’s bias, whether we call it that or not.
So what can we do about it? Psychologists call the answer debiasing: using strategies to reduce the impact of mental shortcuts on judgment. It’s not about getting rid of intuition or pretending we don’t have emotions. It’s about adding enough structure to slow down automatic thinking—so there’s room for clearer analysis.
Some techniques are simple and easy to use—like checklists or building in a pause before decisions. Others are more involved, like decision trees or formal risk assessments. What follows is a set of tools that can help reduce bias in legal decision-making. Some take only a few minutes. Others take more preparation. But they all work toward the same thing: helping clients make clearer, more informed choices.
- Awareness: Simply recognizing a bias—naming it, understanding how it operates—can be the most effective intervention. Once a client knows about anchoring, they’re more likely to question their opening number. Once they understand loss aversion, they may see why a sure settlement feels worse to give up than it logically should. Nobody likes being played—even by their own brain. Awareness alone often puts people on guard and opens the door to better judgment.
- Delay Snap Judgments: Thinking about a decision—really thinking about it—almost always leads to better outcomes than reacting in the moment. Even sleeping on a question can reveal new considerations or dampen emotional noise. But litigation rarely operates on that kind of timeline. And mediation, in particular, is often designed to produce a final decision in a single day.
That’s not an argument against mediation—it’s an argument for preparation. If we know clients are going to be asked to make serious, often irreversible decisions under time pressure, we should prepare them with tools and frameworks before the day of mediation. Creating space in advance to think through the likely moves, trade-offs, and emotional dynamics helps prevent rushed decisions that feel inevitable but may later be regretted.
- Perspective-Taking: One of the most powerful ways to disrupt biased thinking is to step outside of it. That’s easier said than done—especially in an adversarial system where everyone is aligned with one side. Simply telling a client or lawyer to “be objective” rarely changes anything. But asking them to take the other side’s perspective, or imagine how a neutral third party might view the case, can start to shift the frame.
Structured approaches make this more effective. Having a client identify the opponent’s three strongest arguments—or role-play what the mediator will say to the other side—creates cognitive flexibility. Counterfactual reasoning (“What would I believe if I were representing them?”) and decision journals can also help. The goal isn’t empathy. It’s expanding the frame beyond one’s own narrative, which is often the first step toward clarity.
- Red Teaming and Devil’s Advocacy: When we reason alone—or only with those who agree—we miss things. Having someone explicitly tasked with challenging assumptions can expose blind spots, temper overconfidence, and reveal risks that don’t show up in the primary narrative. It also signals that dissent is not only allowed but expected.
This doesn’t have to be formal. Sometimes it’s as simple as saying, “Take ten minutes and try to talk me out of this strategy.” Other times, it means assigning a colleague or client advisor to argue the other side’s position before committing to a course of action. The goal isn’t just to test the case—it’s to test the thinking behind it.
- Disaggregation: Big-picture questions like “What are our chances of winning?” invite overconfidence and vague reasoning. They feel binary—win or lose, good case or bad—and they often trigger gut-level responses. But litigation outcomes are built from layers: liability, causation, damages, collectability, timing, procedural hurdles.
Informal disaggregation—just walking through those components out loud—can reveal where assumptions differ, where the confidence is warranted, and where it’s not. It helps both lawyers and clients see the actual architecture of the case, rather than reacting to an oversimplified headline. Decision trees, discussed later, are the most structured version of this process. They turn disaggregation into a visual, deliberate map of how the case might unfold. But even without a formal model, breaking complexity into parts is a powerful first step toward clearer thinking.
- Use of Base Rates: Lawyers are trained to see each case as unique—and in many ways, that’s true, but uniqueness doesn’t excuse ignoring history. Referencing base rates—how often certain outcomes occur in similar cases—helps ground expectations in reality rather than in internal narratives or vivid, memorable one-offs.
This is especially effective in countering the availability heuristic—our tendency to judge likelihood by what comes most easily to mind. A client might recall a friend who “won big” in a similar lawsuit or a jury verdict they read about in the news. Lawyers might recall an outlier win—or an outlier loss—and give it too much weight. Base rates anchor judgment in data, not anecdotes.
What percentage of motions to dismiss are granted in this court? How often do cases like this settle before trial? What’s the typical verdict range for this kind of claim? These aren’t predictions—they’re context. And while they won’t dictate the answer, they help frame the conversation with evidence, not just memory.
- Pre-Mortem and “What If” Planning: After something goes wrong, people often conduct a post-mortem—analyzing what failed and why. A pre-mortem flips the timing: you imagine the decision has already failed, before it’s made, and ask, what went wrong? What assumptions collapsed? What blindsided us?
This kind of counterfactual thinking forces people to articulate vulnerabilities they may not have voiced otherwise. It shifts the frame from defending a plan to stress-testing it. “What if the judge doesn’t allow that evidence in?” “What if the key witness underperforms?” These aren’t pessimistic questions—they’re preparation.
“What if” scenarios serve the same purpose. They encourage teams to walk through alternate outcomes—not just what they hope will happen, but what could happen. That mindset creates room for contingency planning while the options are still open.
These tools are especially useful before mediation or other pressure-cooker moments. They allow people to anticipate resistance, explore potential failure points, and clarify fallback positions.
Decision trees, discussed below, take this one step further—by mapping those contingencies visually and explicitly, forcing clarity about paths, probabilities, and consequences.
- Multiple Forecasts: When people make predictions in isolation, they tend to anchor to a single scenario—often one that reflects their hopes, fears, or internal narrative. Asking for multiple forecasts—from different people, or from the same person using different frames—can reveal just how uncertain a situation really is.
For example, a lawyer might give one estimate based on their experience, another based on historical data, and a third based on what opposing counsel likely believes. Or a team might collect independent estimates from several members before group discussion begins. The point isn’t to average them into a perfect number—it’s to see the spread, and ask why it exists.
This approach helps reduce motivated-reasoning and overconfidence. It creates space to question assumptions and forces a more probabilistic mindset. It’s also the kind of thinking that feeds directly into decision trees, where different paths and probabilities must be laid side by side—and where disagreement becomes data, not deadlock.
- Checklists: In medicine, aviation, and engineering, checklists are used to prevent avoidable errors in complex, high-stakes environments. Law should be no different. When decisions carry real consequences, simple checklists can bring consistency, reduce omission, and lower the cognitive load on everyone involved.
They don’t have to be long. A five-question checklist for early case evaluation—about liability, damages, venue, procedural posture, and emotional volatility—can catch issues that might otherwise go unspoken. A mediation prep checklist might include the client’s walkaway point, non-monetary interests, likely anchors, and BATNA analysis.
Checklists don’t replace professional judgment. They support it. They ensure that routine complexity doesn’t become silent risk, especially in moments of stress or time pressure.
These tools don’t replace judgment—they support it. They create space to pause, reflect, and think more clearly under pressure. Used together, they help lawyers and clients make decisions that are more deliberate and less distorted. But even taken together, they remain individual tools. Decision trees pull them into a single, structured system. They don’t replace instinct or experience—they give them form, and help people reason through uncertainty when it matters most
C. Decision Trees: The Most Comprehensive Tool
If the techniques above help slow things down and improve judgment, decision trees pull those ideas into one place. They’re not just about calculating numbers—they help organize thinking, surface assumptions, and make uncertainty easier to see. In high-stakes legal decisions, that kind of structure can be the difference between acting on strategy and reacting on instinct.
1. What Is a Decision Tree?
A decision tree is a visual map of uncertainty. It lays out the key decision points and chance events that define a legal case—from motions and rulings to negotiations, trials, and appeals. Each branch represents a possible path the matter could take, showing not just what might happen, but when and how one outcome leads to another.
At its most basic, a decision tree has three kinds of branches:
- Decision nodes, where a party or client must make a choice (e.g., accept a settlement or continue litigating);
- Chance nodes, where the outcome is uncertain and outside the client’s control (e.g., whether a motion is granted or denied); and
- Outcome nodes, which reflect the result of a particular sequence—often with a financial, emotional, or strategic impact attached.
This simple structure can handle a surprising amount of complexity. Some trees focus on just a few key turning points. Others go much deeper, with multiple layers of branching. Lawyers might use them to outline a big-picture strategy, or to break down narrower questions—like the risk of enforcing a judgment or how parallel proceedings might play out.
Decision trees aren’t just about calculating value. They’re tools for organizing thought. They help clarify issues, expose assumptions, and support more focused conversations about risk. Just as important, they give lawyers and clients a shared way of seeing the problem.
Whether it’s a quick sketch on a whiteboard or a fully built-out model in Excel, a decision tree brings structure to conversations that might otherwise rely on instinct or rough guesses. Some clients use the tree to shift into strategy mode. Others just find it helpful to slow things down when everything feels high-stakes.
2. Two Approaches: Choosing the Right Level of Structure
Decision trees can be used in two ways—qualitatively or quantitatively—depending on the decision-maker, the nature of the case, and the goals of the analysis. The main difference is whether the lawyer and client choose to assign probabilities to uncertain outcomes.
Some people are comfortable with numbers. For them, adding probabilities and values isn’t about chasing perfect accuracy—it’s about bringing structure to a complex decision. Running the math helps them feel like they’re thinking things through logically.
On the other end are clients who want nothing to do with numbers. They’re not interested in breaking down every variable or guessing at odds. To them, trying to quantify the future feels more distracting than helpful. They’d rather talk through the big picture and focus on what feels clear.
Most clients fall somewhere in the middle. The right approach doesn’t depend on job title—it depends on how someone handles uncertainty and how they prefer to make decisions. And either way, the value of a decision tree often comes from building it together. It’s not just about the diagram—it’s about the process. Working through the model forces useful conversations, surfaces assumptions, and helps the client take an active role in the decision—instead of just reacting to the options.
3. Qualitative Decision Trees: Structure Without Quantification
A qualitative decision model can be created ahead of time by the lawyer or built together with the client. Both approaches can work. But when a client has trouble seeing the complexity of the case—or resists thinking in structured terms—building it together is often more effective. It helps them engage with the issues without getting bogged down in abstractions or pressured by numbers.
In a collaborative model, the lawyer might take the lead—sketching out key decisions and uncertainties to get the conversation started. Or they might take a lighter touch, asking questions that help the client talk through the moving parts themselves. Either way, the process gives the client a better sense of how the case might unfold and what’s likely to influence the outcome.
What makes it a qualitative model is that you don’t assign probabilities. It’s a logical structure, not a math problem. Together, the lawyer and client map out key decisions, possible forks in the road, and different outcomes. That might take the shape of a full decision tree—or just a diagram, a list, or a simple chart. The point is to get the case out of their head and onto the page, where the different paths and consequences are easier to see.
You can still include dollar amounts—and often, you should. Seeing that one path could lead to a $2 million exposure while another likely ends with a $150,000 settlement can help put things in perspective, even without knowing the odds. It keeps the conversation grounded in real consequences, without asking the client to start thinking in terms of percentages or probabilities.
This approach is especially valuable when:
- The client is resistant to numerical precision or skeptical of mathematical modeling;
- The case is in an early stage, with too many unknowns to justify quantification; or
- The client is emotionally overwhelmed and needs a visual, intuitive guide to reframe the conversation.
Clients often enter mediation or negotiation with a single question: What are our chances of winning? A qualitative model shows that this is the wrong question—or at least an incomplete one. It makes clear that the case is not a coin toss, but a sequence of decisions and contingencies, many of which are within the client’s control.
It also provides a language for strategy. With a shared model, lawyer and client can talk about options and trade-offs without resorting to generalities. And because the model invites the client to see how the pieces fit together, it can help defuse resistance—especially when a client is struggling to appreciate the true risk or uncertainty in their position.
Ultimately, qualitative decision trees are a way to make the invisible visible. They don’t answer the question for the client—they equip them to ask better questions, and to navigate their case with greater perspective and control.
4. Quantitative Decision Trees: Revealing the Real Odds
Quantitative decision trees build on the same structural tools as qualitative models—but add an essential layer: probabilities and financial values. This approach helps lawyers and clients see how outcomes compound through uncertainty, and how each stage of the case shapes the overall picture.
Its greatest value is in showing how an abstract estimate of success—say, “a better than even chance at trial”—often dissolves when unpacked. If there’s an 80% chance of defeating summary judgment and a 50% chance of winning at trial, the combined probability is only 40%. And that’s just a basic example. The more variables involved, the more diluted those odds become. Quantitative modeling helps clarify what must go right—and how rarely everything does.
Even a simplified litigation path might involve many variables:
- Will a motion to dismiss be granted in full, in part, or denied?
- Will the plaintiff survive summary judgment?
- Will key evidence be admitted or excluded?
- Will certain evidence be well received, neutral, or bad?
- What are the odds of a particular outcome on each potential claim?
- What are the odds of a high, medium, or low recovery on each claim?
- Could an appeal change the result or delay recovery?
- Will the judgment be collectible?
Each of these questions introduces uncertainty—and each probability affects the ones that follow. The further you trace a case’s path, the smaller the combined probability of any specific outcome becomes. That’s not pessimism; it’s just math. And exposing that math can be one of the most useful things a lawyer can do for a client facing a costly decision.
This kind of math can sound intimidating—especially to lawyers who didn’t go to law school because they loved numbers. But at its core, decision trees involve only basic multiplication and addition. The hard part isn’t the arithmetic. If you can multiply percentages—like 80% × 50% = 40%—you can do these trees.
When a lawyer walks a client through the assumptions behind their case—from motions through trial to collection—it doesn’t just produce a number. It produces a conversation. And, it often surfaces places where the client’s expectations are misaligned with the actual risks.
5. Going Deeper: Risk and Rigor – A Lawyer’s Guide to Decision Trees for Assessing Cases and Advising Clients
For lawyers who want to go deeper, Risk and Rigor – A Lawyer’s Guide to Decision Trees for Assessing Cases and Advising Clients by Marjorie Corman Aaron is the most comprehensive treatment of decision trees in the legal context. The book walks through the logic and structure of decision trees with clarity, showing how lawyers can organize uncertainty, quantify risk, and communicate options more effectively. It emphasizes how decision trees support—not replace—judgment, and provide step-by-step examples drawn from real litigation strategy decisions.
The book is thoughtful, methodical, and approachable even for readers without a math background. It avoids jargon, defines its terms carefully, and respects the realities of legal practice. Whether a lawyer wants to sketch a rough model on a whiteboard or build a detailed quantitative analysis for a boardroom client, Risk and Rigor offers practical guidance and intellectual grounding. It also directly addresses the psychological and institutional barriers that often prevent lawyers from adopting structured tools, reinforcing many of the themes explored in this article.
Conclusion: Playing Smarter
The Moneyball revolution didn’t just change baseball—it changed the way we think about decision-making itself. It showed that in any high-stakes environment, intuition alone isn’t enough. Success favors those who impose structure on uncertainty and use tools to sharpen judgment. Law is no exception.
Ours is a profession that values instinct, experience, and advocacy. But even the best lawyers—like the best scouts—are vulnerable to predictable mental shortcuts. Behavioral science has shown that these cognitive biases aren’t signs of incompetence; they’re the cost of being human. That means we can’t outmuscle them with willpower. We need tools.
Debiasing techniques like red teaming, calibration exercises, checklists, and counterfactual thinking help lawyers see problems more clearly and avoid costly errors. Decision trees go a step further. They don’t eliminate uncertainty, but they help us see it plainly—by mapping out risks, testing assumptions, and revealing hidden dependencies.
When used well, decision trees bring structure to strategy and clarity to communication. They foster disciplined thinking, anchor client expectations, and create space for sound judgment. They don’t replace legal instincts—they refine them.
In that sense, they echo what Billy Beane brought to baseball: not a rejection of intuition, but a way to make it better. The lawyers who embrace these tools won’t just think more clearly. They’ll advise more confidently, negotiate more effectively, and litigate with greater precision. And in a world where outcomes are uncertain and clients hire judgment, that edge can make all the difference.