The Restaurant That Wouldn’t Let Go — Staying Too Long Doesn’t Undo What You Already Spent

Kevin and the Restaurant That Wouldn’t Let Go

Picture a man — call him Kevin — who spent four years trying to save a restaurant that was losing twelve thousand dollars a month. He opened it in 2017 with two hundred eighty thousand dollars of his own savings and a two-hundred-thousand-dollar loan against his house. By 2018, the numbers were clearly wrong. The location was underperforming. The concept was oversaturated in that market. The labor costs made the unit economics unfixable without a total model change. His accountant told him this in July 2018. His business partner told him the same thing in September. Three independent restaurant consultants told him the same thing again between 2018 and 2020. Kevin kept going anyway. In his own head, the reasoning looped the same way every time. He’d put in almost half a million dollars. He couldn’t accept that it was just gone. Every time he thought about walking away, he thought about every dollar sitting in that building, and he couldn’t do it. He finally closed in March 2020. The pandemic made the decision for him, when he couldn’t make it for himself. Total loss by then: six hundred eighty thousand dollars, and four years he can’t get back.

You need to see this clearly before we go any further. The money Kevin had already spent pulled him forward into a situation that was destroying him. The rational question — what is the future value of continuing versus stopping — was never the question he was actually asking himself. He was asking a completely different one: how do I avoid accepting that the past investment is already lost? Those two questions have entirely different answers. The wrong one cost Kevin an additional two hundred thousand dollars and forty-eight months of his life. And if you’re honest with yourself right now, you’ve probably asked the wrong one too, at least once, somewhere in your own life.

This episode is what I’m calling the Exit Calculation Protocol. It’s a systematic framework for evaluating your own sunk costs clearly. It’s built to help you understand why this bias is so psychologically powerful in you. And it hands you the tools you need to catch yourself before you replicate Kevin’s outcome. We’ll walk through the research behind it. Hal Arkes and Catherine Blumer’s foundational 1985 experiments. Richard Thaler’s loss aversion framework. Barry Staw’s work on escalation of commitment. Kahneman’s prospect theory. Annie Duke’s practical framework from her book Quit. And we’ll walk through four composite cases of how this bias operates in your life: business, relationships, career, and financial investment.

No platitudes here, and nothing for you to just nod along to. No “it’s okay to start over” without the reasoning behind it, laid out for you plainly. Just the mechanics of why this bias exists in you, why it’s so durable in you specifically, and how you systematically overcome it in your own life.

The Academic Foundation: What the Evidence Reveals

The sunk cost fallacy entered behavioral economics through a series of landmark experiments in the 1980s. Start with the foundational paper: Hal Arkes and Catherine Blumer’s 1985 study, “The Psychology of Sunk Cost.” Their design was simple and elegant. Give subjects a hypothetical scenario where they’ve paid for a ski trip. Then give them a conflicting free ski trip invitation to a better resort on the same day. The subjects who had paid more for the original trip were significantly more likely to go to the worse resort. Simply because they had invested more money in it. The investment in the original trip was already spent. It could not be recovered, by anyone, under any circumstance. But it changed behavior anyway.

Arkes and Blumer ran multiple variations of this experiment and found the same result every time. People weighted irretrievable past investments when making future decisions, contrary to what rational choice theory predicted they should do. The sunk cost is, by definition, gone. It cannot affect the future value of either option in front of you. Economically, it’s irrelevant. Psychologically, it’s enormously powerful, and it’s operating in you right now, in whatever you’re currently weighing.

Richard Thaler won the Nobel Prize in Economics in 2017, and he gave you the theoretical foundation for why this happens, through his work on mental accounting and loss aversion. Building on Kahneman and Tversky’s Prospect Theory, which showed that losses feel roughly twice as powerful to you as equivalent gains, Thaler showed you maintain “mental accounts” for your investments. When you’ve invested five hundred thousand dollars in a restaurant, you have a mental account for that money. Closing the restaurant forces you to close that account at a loss. The psychological pain of that closure is roughly twice as large as the pleasure you’d have felt had it been equally profitable.

“Losses loom larger than gains. The aggravation that one experiences in losing a sum of money appears to be greater than the pleasure associated with gaining the same amount.” — Daniel Kahneman and Amos Tversky, Prospect Theory, 1979.

Sit with that result for a second. You will make economically irrational decisions to avoid the psychological event of closing a mental account at a loss. You’ll keep investing in losing propositions. You’ll refuse to sell depreciating assets. You’ll stay in bad situations. Your avoidance of psychological pain overrides your calculation of actual future benefit, and it happens quietly enough that you usually don’t notice it happening.

Barry Staw at Berkeley extended this into organizational behavior with his concept of “escalation of commitment.” That’s the tendency of decision-makers to increase their investment in a failing course of action in response to negative feedback. Staw’s 1976 paper found that people who had personally made the original decision to invest in a project kept investing in it, even after poor results, more than people who’d simply inherited someone else’s decision. Personal agency in the original decision amplified the sunk cost effect. The more you owned the original choice, the harder it became for you to admit the choice was wrong.

Staw’s organizational research showed this same escalation pattern in corporate settings, government policy, and military strategy. Think of the Vietnam War as the canonical large-scale example. Troop commitments and expenditures kept escalating in response to deteriorating outcomes. Part of the driver was the psychological impossibility, for the decision-makers who’d authorized the original intervention, of accepting that the previous investment was already lost.

Prospect Theory: The Architecture of the Bias

You need Kahneman and Tversky’s Prospect Theory to fully understand what’s operating in you here. It’s the 1979 framework that’s probably the single most important contribution to behavioral economics in the twentieth century, and it’s the foundation of Kahneman’s Nobel Prize in 2002.

Classical economic theory assumes you evaluate outcomes in terms of absolute wealth states — that you care about how wealthy you are, not about gains and losses relative to some reference point. Kahneman and Tversky showed this was empirically wrong. You evaluate outcomes in terms of gains and losses relative to a reference point, usually your current state or some expectation you’ve already formed. And you weigh those gains and losses asymmetrically. A loss of a thousand dollars feels roughly twice as bad to you as a gain of a thousand dollars feels good.

That asymmetry, loss aversion, is the engine running your sunk cost bias right now. When you’ve invested two hundred eighty thousand dollars in a restaurant and it’s failing, stopping feels like a loss of that whole amount to you. The psychological weight is enormous. Continuing feels like maintaining the possibility of recovery — not gaining, but not yet fully losing either. Your brain isn’t calculating the real expected value of continuing versus stopping. It’s calculating how to avoid locking in this loss. Those are different problems, and the second one is what produces your sunk cost behavior.

Kahneman also documented the “endowment effect”: your tendency to overvalue things you already own relative to their market value. It’s a related manifestation of the same asymmetry. An investment you’ve made becomes yours in a way that makes losing it feel more costly to you than an equivalent gain feels beneficial. This applies to your business, your relationships, your career, and your financial assets, all equally.

Here’s the practical implication for you. Your feelings about a past investment are not a reliable guide to the future value of continuing versus stopping. Your feelings are calibrated around avoiding the psychological event of loss realization. The rational calculation is about future value only. These two things will systematically give you different answers, and the rational one is the one you should trust.

Annie Duke: Quitting Is a Skill

Annie Duke’s 2022 book, Quit: The Power of Knowing When to Walk Away, is the most practically useful treatment of this problem you’ll find. Duke spent years as a professional poker player before becoming a behavioral economist and decision consultant. Her poker background gives her a unique angle on this. In poker, you are constantly making sunk cost decisions. Every chip you’ve put into a pot is a sunk cost. The question you’re always facing is: given the cards you have and the cards that are out, what’s the expected value of continuing versus folding right now?

Duke’s central argument is that your culture systematically celebrates persistence and stigmatizes quitting, in ways that aren’t calibrated to actual outcomes. “Winners never quit and quitters never win” is a survivorship bias artifact. You only hear the stories of people who persisted and eventually succeeded. You don’t hear the much larger population who persisted and eventually failed even bigger. That selection bias makes you dramatically overestimate the value of staying, and dramatically underestimate the value of a strategic exit.

Duke introduces the concept of “quitting criteria”: specific, predetermined metrics that, if triggered, signal your exit regardless of what you’ve already invested. The key is timing. You have to set these criteria before the investment begins, or at least before the weight of the sunk cost makes objective evaluation impossible for you. You set your quitting criteria when you’re not yet in the grip of loss aversion. You execute them when you are.

Duke also names what she calls the “identity problem” with quitting. For a lot of people, the project, the relationship, or the career becomes entangled with their identity. Quitting the thing gets experienced as quitting on yourself — abandoning a self-concept, not just a strategy. This is why the sunk cost effect hits hardest where your identity is most invested. Kevin’s restaurant wasn’t just a business. It was the fulfillment of a twenty-year dream of owning his own place. Walking away felt, to him, like walking away from a version of himself. That architecture made the correct decision almost psychologically impossible for him.

Escalation of Commitment: The Organizational Dimension

Barry Staw’s research shows you that the sunk cost fallacy operates at the organizational level through mechanisms that amplify the individual bias further.

In organizations, the decision-makers who authorized the original investment have a public commitment riding on its success. Reversing course means admitting publicly that the original decision was wrong. That triggers not just psychological loss aversion but reputational loss aversion too. The cost of being wrong becomes both financial and social. The pressure to stay the course is proportional to how loudly and publicly the original commitment was made.

Picture a woman — call her Maria — who was SVP of Operations at a mid-sized software company. Her team launched a major initiative to build an internal project management platform. She’d championed the initiative to the board, secured a three-million-dollar budget, and publicly committed to a twelve-month delivery timeline. By month eight, it was clear the build approach was wrong. The timeline was unfixable. The right move was to evaluate commercial alternatives that had improved since her original decision. She kept building anyway. Not because she didn’t see the signals — she did. She’d championed this publicly, and reversing it meant admitting, in front of everyone, that her judgment had been wrong. The platform shipped sixteen months late. It was functionally inferior to the commercial alternatives. It was replaced two years after launch. Later, she’d put it simply: she knew at month eight. She knew. She just couldn’t make herself say it out loud.

Staw’s research suggests your best protection against escalation of commitment is structural. Build decision review processes that evaluate current investments from the perspective of a new decision-maker who didn’t make the original choice. Ask yourself this: if you had no existing investment in this project, and someone proposed starting it today with its current specifications and prospects, would you fund it? If the answer is no, the sunk cost is driving your continuation, not a rational calculation.

The Four Domains: How Sunk Cost Operates Across Your Life

The sunk cost fallacy usually gets illustrated with financial examples. But it operates with equal power across every domain of your life. Here’s how it shows up in each one.

• Business and financial investment. Kevin’s story is the archetype. The question that cuts through it: if you woke up tomorrow with no history in this investment, and someone showed you its current state and asked you to invest fresh, would you? If no, you’re standing in sunk cost territory right now.
• Relationships. “I’ve been in this relationship for six years” is not a reason to stay in one that’s making you miserable. The six years are gone. The relevant question is: given everything you now know, would you choose to start this relationship today? If the honest answer is no, sunk cost is sustaining it, not genuine present-value judgment. This isn’t a call for casual exit from imperfect relationships — every long-term relationship goes through hard periods. It’s a call for honesty about the difference between a hard period and a structural incompatibility you’re refusing to name.
• Career and education. The degree you’re three years into but hate. The certification track you’ve invested eighteen months in that doesn’t lead where you want to go. The career path your family encouraged and you committed to publicly. All of these can become sunk cost traps for you. The mechanism is identical every time. The past investment creates resistance to a change that future value calculation clearly supports.
• Non-financial time investments. The book you’re reading but not enjoying — do you finish it because you’ve already read two hundred pages? The project going nowhere — do you keep at it because of the weeks already spent? The habit you’ve tried to build eleven times without success — do you try a twelfth time the same way, because you can’t accept the cost of the first eleven failures? Time is a sunk cost resource, exactly like money. The hours you’ve already spent cannot be recovered. The only rational question is what the best use of your future time is, not what it says about your past time if you change direction now.

Notice something about your own life right now as you hear these four. You are probably running at least one of them at this exact moment, and you probably already know which one it is. You knew before I finished the list. That flash of recognition you just had is worth trusting. It is your own honest signal, arriving faster than your rationalizations can catch up to it and talk you out of it. Sit with whichever domain lit up for you. Don’t rush past it. You’ll need it again by the end of this episode, when you run your own numbers through the protocol.

Notice, too, what your mind just did in that half-second of recognition, because it’s worth naming before we move on. You probably felt the flash, and then you probably felt the pull to explain it away. That’s fine, this is different, my situation isn’t like theirs. You have a reason ready, and the reason feels solid to you, because it was built specifically to hold up under your own scrutiny. It’s had years of practice. Every sunk cost trap you’ve ever seen in someone else came wrapped in a reason that felt solid to them too. Kevin had a reason. James had a reason. The reason is not evidence that you’re the exception. It’s the standard feature of the trap itself, present in every single case, including yours. Hold the explanation you just generated a little more loosely than you’re inclined to. You’ll get the chance to actually test it later in this episode, against real numbers instead of against your own certainty.

Here’s a case worth sitting with, because it touches most of these domains at once. Call him James. He spent seven years in a job that was, in his own words, slowly eating him alive. The work was wrong for him. The culture was wrong. His boss was a textbook case of several patterns we covered in the dark psychology episode. The role wasn’t building toward anything he actually wanted. He stayed because he’d already put seven years in. He stayed because he was two years from vesting his last tranche of equity. He stayed because leaving felt like admitting the seven years had been wrong. He finally left in year nine, at forty-one. The first month at his new company felt to him like recovering from something he hadn’t realized he’d been carrying. Two years of extra equity was worth far less than the two extra years of compounding damage it cost him.

The Exit Calculation Protocol

Here’s what you actually do about it. The Exit Calculation Protocol is a six-step framework for evaluating whether you’re in a sunk cost trap, and what to do about it once you know.

1. Zero-base the decision. Imagine you have no prior investment here — no money, no time, no public commitment, no emotional history. You’re evaluating it fresh. Given everything you now know, would you choose to start this investment today? Write the answer down before you continue. This is the most important step you’ll take. It strips away the sunk cost and gives you a pure future-value reading.
2. Separate what’s recoverable from what isn’t. Not everything is a sunk cost. Some parts of your investment — skills you’ve built, relationships you’ve made, knowledge you’ve gained — are recoverable assets that transfer regardless of whether you exit. Inventory those. The sunk cost is only the part that’s truly non-recoverable. For Kevin, the restaurant knowledge and the operational skills he’d built were recoverable. The six hundred eighty thousand dollars was not. Focusing on what’s recoverable changes your emotional calculus.
3. Quantify the future cost of continuing. What will you actually spend — money, time, health, relationships — if you continue for the next twelve months? The bias makes your past losses feel enormous and your future costs feel abstract. Make the future cost concrete. Kevin could have calculated, back in July 2018, exactly how much more he’d lose at his current burn rate. He knew the number. He just never wrote it down and made it real.
4. Set your quitting criteria before you need them. If you haven’t exited yet but you’re weighing it, set specific, measurable criteria that trigger your exit regardless of what you’ve invested. If revenue hasn’t hit a certain number by a certain date, you close. If you’re not genuinely enjoying a new path within twelve months, you leave. Make these specific enough that you can’t rationalize around them once loss aversion has its grip on you. Write them down. Share them with someone who’ll hold you to them.
5. Find an external perspective. Staw’s research shows you that people who made the original decision are the worst evaluators of whether to continue it. Find someone with no stake in the original decision, and ask them to evaluate your situation fresh. Listen to what they say. They aren’t subject to your sunk cost distortion. Their read is cleaner than yours.
6. Distinguish the story from the numbers. “I’ve put so much in, I can’t walk away now” is a story you tell yourself. It is not an analysis. Separate the story from the numbers by writing out the calculation explicitly. What’s the expected value of continuing twelve more months versus exiting now? The numbers don’t need to be precise. Making them explicit puts analysis in the driver’s seat instead of the emotional narrative.

The Psychological Work: Reframing the Loss

The protocol solves the analytical problem for you. But the sunk cost fallacy is primarily a psychological problem. The tools only work if you use them honestly, and that requires reframing the emotional landscape of exit in your own mind first.

Here’s the key reframe, and Duke states it clearly in Quit: staying in a bad situation is not preserving your investment. It is continuing to invest in a negative-return asset. Every day you stay is a new investment decision. If your honest answer to the zero-base question is “I wouldn’t start this today,” then every day you stay is a new bad decision, stacked on top of the original one. You are not protecting your past investment by continuing. You are compounding it.

The second reframe: exit is not failure. Exit is redeployment. Kevin’s six hundred eighty thousand dollars was his entire net worth. If he’d closed at the eighteen-month mark, the roughly two hundred thousand dollars remaining could have gone into a different business, or an investment, or a more stable financial footing. Staying until the end meant he redeployed nothing at all. His real choice was never between losing two hundred eighty thousand and possibly winning. It was between losing two hundred eighty thousand and redeploying two hundred thousand, or losing six hundred eighty thousand and redeploying nothing. Framed accurately, the exit was clearly superior. Kevin never framed it that way.

The third reframe connects to Duke’s identity point directly. The thing you invested in does not define you. Your choices going forward define you. A career path you chose in your twenties doesn’t have to be the story of your whole life if you change it in your thirties. A relationship that was wrong doesn’t become right by your continuing it. A business that failed taught you things you can deploy in the next one. Your identity gets built in how you respond to the outcome, not in whether the original investment succeeded.

The Opportunity Cost Nobody Counts

One of the most invisible costs of your sunk cost thinking is opportunity cost: the value of the alternatives you’re not pursuing because you’re continuing the current investment. Kevin’s six hundred eighty thousand dollars and four years were the direct costs. The businesses he might have started, the experiences he might have had, the relationships he might have built — those are uncountable, but just as real.

Kahneman notes that loss aversion and opportunity cost blindness are related. You’re calibrated to feel losses acutely and to feel forgone gains much less intensely. The alternative you’re not pursuing doesn’t feel like a loss to you. It feels like an absence. But economically, it’s a cost. The year you spent in the wrong job is a year you didn’t spend building the right one. The money you poured into a failing business is money that didn’t compound somewhere else. The forgone gain is a real subtraction from your lifetime wealth, even though it never shows up as a line item you can see.

Building opportunity cost into your exit calculation is one of the most useful adjustments you can make right now. Step three of the protocol, quantifying the future cost of continuing, should include opportunity cost explicitly. What would you do with the money and time if you exited? What’s the expected value of that alternative? Now compare the two paths. If the alternative is higher, continuing is costing you the difference. That isn’t a sunk cost. It’s an active cost that compounds every day you don’t exit.

You will not want to do this math. Nobody wants to do this math, because doing it honestly means admitting that a situation you’ve defended to yourself, maybe out loud to other people too, has been quietly draining you of a better life you could already be living. Do it anyway. Write down what you’d do with your money and your time if you were free of this thing tonight. Write down what that alternative is actually worth to you, in plain terms you’d say out loud to someone you trust. Then hold the two numbers side by side and let yourself see them clearly, without flinching away from either one.

When Persistence Is Actually Right

I want to be clear about something important here. Persistence is sometimes correct. The sunk cost framing can be misapplied as a justification for giving up on things genuinely worth continuing.

The key distinction is between sunk cost persistence, continuing because of what you’ve already invested, and rational persistence, continuing because the forward-looking value is genuinely positive. The protocol is built to separate these two for you. If you run the zero-base test and your honest answer is yes, then your persistence is rationally justified. The past investment isn’t why you’re continuing. The future value is.

Many legitimate long-term investments look like losses in the short term. Building a skill, a business, a meaningful career — all of these require extended periods of negative return before the compounding begins for you. The sunk cost analysis doesn’t preclude long-term thinking. It just requires that your long-term investment be justified by actual future value, not by the psychological impossibility of admitting a past mistake.

Duke’s quitting criteria concept is calibrated for exactly this distinction. The criteria should be set on whether the forward-looking value stays positive, not on whether exit is emotionally painful for you. If your business is losing money but you have specific, time-bound reasons to believe the inflection point is near, and those reasons are grounded in evidence, continuation may genuinely be rational. If the only reason you’re continuing is that you’ve already invested too much to stop, the sunk cost is driving your decision instead.

The difference is ultimately testable. Can you articulate a specific, evidence-based case for positive future return? If yes, you’re reasoning about the future. If your case reduces to “I can’t accept what I’ve already invested is lost,” you’re reasoning about the past. One of these produces good decisions for you. The other produces more Kevins.

For more on the cognitive mechanisms that amplify this effect in you, go back to EP67 on cognitive biases, specifically loss aversion, status quo bias, and escalation patterns. The full toolkit for why your brain makes these errors lives in that episode.

The first-principles framework from EP68 gives you the cleanest tool for the zero-base test: strip away every assumption from the history of the investment and reason from bedrock. What is actually true about this situation’s future? Not what you’ve been telling yourself. Not what you’ve committed to publicly. What is actually true about the forward value?

And EP69 on dark psychology connects here in a way you should know about. Sunk cost thinking is something manipulators deliberately create and exploit in you. The dependency phase of the grooming process we covered there is designed to maximize your psychological investment, so exit becomes psychologically prohibitive for you. Sunk cost gives you another angle on why manipulation dynamics are so hard to walk away from.

The complete decision-making architecture these episodes build toward lives in the Mindset Tools library. And if you want the applied resilience framework that connects these tools to real-world pressure, start with the Resilience Principles section.

Why Exit Feels Like Death

The protocol gives you the analytical tools for a clear evaluation. But here’s why you need them in the first place: exit doesn’t feel like a rational choice to you. It feels like a form of death. That’s why Kevin kept going for twenty-four months after his accountant told him to stop. Understand the mechanics of that feeling, and you can work with it instead of being controlled by it.

You’ve probably noticed the specific shape this takes in your own body when you get close to your own exit decision, even if you’ve never described it out loud before. A tightness that shows up before you’ve consciously thought anything at all. A reflexive reach for one more piece of information you tell yourself you need before you can possibly decide, even though you’ve already gathered more information than any decision like this actually requires. A sudden, urgent interest in every reason this time might be different. None of that is you thinking clearly. That’s your body registering the approach of a loss before your mind has caught up to name it. Recognizing the physical signature for what it is, rather than mistaking it for genuine new uncertainty about the facts, is itself one of the more useful skills you can build from everything in this episode.

Kahneman’s loss aversion research gives you the quantitative foundation: losses feel roughly twice as bad to you as equivalent gains feel good. But the sunk cost loss is more complex than a simple multiplier. What you’re losing when you exit a major investment is not just money or time. You’re losing the future self you expected to have. The restaurant Kevin was going to own. The career Maria was going to build. The relationship James was going to repair. These anticipated futures were real parts of how these people understood their own lives. Abandoning them means abandoning a self-concept, not just an asset.

Roy Baumeister and colleagues studied what they call “identity fusion” — how deeply a goal becomes integrated into your sense of self. The more identity-fused you are with a goal, the more its failure feels like personal annihilation to you, rather than mere disappointment. Kevin’s restaurant wasn’t just a business. It was a twenty-year dream. It was the story he told himself about who he was. Closing it meant revising that story fundamentally, which is far more demanding, emotionally, than simply accepting a financial loss.

This is why outside perspectives matter so much for your sunk cost decisions. Someone who doesn’t share your identity fusion sees clearly what you can’t. The accountant who told Kevin to close in July 2018 didn’t have twenty years of identity invested in that restaurant. He could see the numbers clearly, precisely because his own identity wasn’t at stake. His recommendation was correct, and Kevin couldn’t hear it, precisely because of the identity dimension sitting underneath everything. That’s not a failing specific to Kevin. It’s a predictable feature of how identity-fused investments work in anyone, including you. Seek out people outside your own identity investment. Weight what they tell you heavily.

How Sunk Cost Interacts With Social Accountability

Staw’s research shows you the sunk cost effect gets amplified by public commitment. When you’ve announced a direction and recruited support for it, the psychological cost of reversal includes more than the loss of the investment. It includes the loss of credibility with everyone who backed you. This social dimension is one of the primary reasons large-scale sunk cost catastrophes happen inside organizations. The person with the authority to call off a failing project is usually the person who most publicly committed to it — and so the person facing the highest social cost for doing so.

Maria’s situation is the clearest example here. The board presentation where she championed the initiative was the public commitment that made reversing course so costly for her. Every subsequent update, every project review, every team meeting where she described progress, added another layer of social accountability. Each layer made honest reassessment more threatening to her identity.

Here’s your practical defense: structure your own accountability differently from the start. Instead of committing to a specific solution publicly, commit to a specific outcome, and make the success criteria explicit before you begin. Commit to a functional platform by a set date, with an open evaluation of build versus buy versus partner at each decision gate. That’s a structurally different commitment than announcing you’re building an internal platform. The first creates accountability for the outcome, with flexibility about the path. The second creates accountability for the path itself, which makes any change look like failure regardless of whether it serves the outcome better.

This reframe, committing to outcomes rather than solutions, is one of the most durable defenses you can build against escalation of commitment. It gives you permission to update your approach when new information arrives, without that update reading as an admission of failure. Updating becomes just following the evidence toward the outcome you already committed to. The social cost of course correction drops dramatically once you’ve committed to a destination rather than one specific route.

The Timing Problem: When Is It Actually Too Late

Here’s a question worth asking honestly. Are there situations where it’s actually too late to exit, where continuation is genuinely the only rational option left? The honest answer is yes, occasionally. But far less often than the sunk cost fallacy makes it feel to you.

The genuine cases exist where the cost of exit exceeds the cost of continuation in pure prospective terms. Where the contractual penalties, the reputational consequences, or the relationship damage of stopping now genuinely outweigh the future cost of continuing. A business owner who’s personally guaranteed debt has a different exit calculation than one who hasn’t — the guarantee triggers an immediate financial event rather than a slower ongoing loss. A professional who’s made public commitments that can’t be reversed without severe damage faces a genuinely different cost structure than one who can exit quietly.

Here’s the critical diagnostic question for you: is your continuation justified by the prospective cost structure, or by the weight of past investment alone? These two things can look identical from the inside, which is exactly why the protocol requires explicit prospective analysis. If the analysis shows exit costs genuinely exceed continuation costs, continuing is correct — but for prospective reasons, not because of what you’ve already spent. If the analysis shows the opposite, and you’re still choosing to continue, you’ve found a clear sunk cost trap, and you know exactly what it’s costing you.

The more common error is the reverse. It’s continuing when the prospective cost structure clearly favors exit, but the psychological weight of the sunk cost makes exit feel catastrophic anyway. Kevin’s July 2018 analysis was clear. Continuing would cost another two to four hundred thousand dollars, with no realistic path to profitability. Exit would cost the roughly four hundred eighty thousand already spent, plus whatever couldn’t be recovered from the assets. The calculation strongly favored exit. He continued for twenty-four more months regardless. The sunk cost was never a rational argument for staying. It was a psychological obstruction sitting on top of a clear calculation.

Designing Systems That Make Exit Easier

Your individual psychological tools — the protocol, the zero-base test, the quitting criteria — are powerful. But they work against the grain of how your brain naturally processes ongoing investments. Your most durable protection is systemic. Design the structures around your major decisions to make rational exit easier and more socially acceptable, before the investment even begins.

Duke’s quitting criteria concept is the single most useful tool for this. Before any major investment, define explicitly what would need to be true, by what date, for the investment to have proven its viability. What specific metrics, if unmet, would trigger a formal reassessment for you? Write these down. Share them with at least one other person. Revisit them at the stated date, no matter how you feel about the investment when that date arrives.

The discipline of pre-committing to quitting criteria separates rational persistence from sunk cost persistence in your own decisions. Set the criteria before you’re emotionally invested, and you’re reasoning prospectively. Evaluate against them at the agreed date, and you’re still reasoning prospectively, but now with real data in front of you. The criteria exist to stop your future self, who’ll be more emotionally invested than your past self was, from rationalizing what your earlier self would have called irrational immediately.

Thaler and Sunstein’s “nudge” framework offers you one more design principle: make the rational default the easy option. In retirement savings, that means auto-enrollment. In your own sunk cost management, it means building exit evaluation into your project governance, rather than requiring a special event to trigger it. Give every major project a scheduled quarterly review. Ask not just how things are going, but whether, given where you actually are, you should be doing this at all. That question, asked routinely, drops the activation energy for exit dramatically.

Organizations that build this kind of governance report consistently better outcomes on large projects. Partly that’s because they catch failures earlier. Partly it’s because the exit criterion changes the risk tolerance of the initial decision itself. When everyone knows a project gets evaluated against specific criteria at six months, the initial investment tends to get scoped more carefully, the criteria tend to be more realistic, and the team tends to be more honest along the way. The exit structure improves the entry decision. That’s the full value of designing the system, not just managing any single decision inside it.

James and the Flight From Chicago

Earlier, you heard about James and his nine years in a job that was wrong for him. His fuller story is worth telling, because it walks through nearly every dimension of the sunk cost problem at once, and because the way he eventually exited is worth learning from directly.

James joined the company at thirty-two, which, in retrospect, was his first error. A better opportunity had fallen through at the last moment, and he needed income, so he took the job as a fallback he told himself was temporary. By year three, it had stopped feeling temporary. Compensation had grown substantially. The team was good. Leaving would have meant constructing a whole new professional narrative from scratch. By year five, the equity vesting schedule had created a specific financial milestone he told himself would be the right exit point. By year seven, he’d vested most of the equity, but the remaining tranche was worth enough that leaving felt financially irrational. By year nine, he was forty-one, earning a salary that had become his family’s lifestyle anchor. His professional identity felt inseparable from his organization. He carried a persistent low-grade misery he’d been successfully numbing through compensation and routine.

The moment that broke his sunk cost trap wasn’t a particularly dramatic one. Picture him on a flight from Chicago to San Francisco, doing the math on how many working years he had left before a realistic retirement date. The answer was approximately twenty-five. He asked himself a simple question: would I choose to spend twenty-five more years doing this, if I had the option of doing something different? The answer took him about thirty seconds. He landed in San Francisco and called a recruiter he’d been avoiding for two years.

Notice what happened there. The zero-base question, framed as a calculation about future years rather than a retrospective about past years, broke the sunk cost frame for him. He wasn’t being asked to disavow the nine years. He was being asked what he wanted to do with the next twenty-five. Those are different questions, and the second one had a clear answer the first one had been obscuring the whole time.

Fourteen months into his new company, he described the environment as categorically different from what he’d spent nine years inside of. His own engagement felt like something he hadn’t experienced since the early years of his career. The nine years weren’t wasted, he’d say if you asked him now. He learned a lot. He built skills. He made relationships that still matter to him.

“But every year I stayed past year three was a year I was choosing inertia. The sunk cost was keeping me in place. It wasn’t keeping me safe. There’s a difference.”

There’s a difference, and it’s the difference this whole episode is about. The sunk cost is not a reason for you to stay. It is a weight that makes moving feel impossible, while the actual cost of staying compounds silently and invisibly, until something like a flight from Chicago makes it undeniable. Run the numbers. Ask the zero-base question. Set the quitting criteria. Move when the criteria are met. The past investment is not coming back to you. The future years are yours to allocate however you choose.

The Broader Cultural Problem: Why We Celebrate the Wrong People

The sunk cost fallacy doesn’t exist in a cultural vacuum for you. Your culture tells you specific stories about persistence, commitment, and what it means to be a person of integrity. Those stories are deeply embedded in how you talk about admirable people. And they systematically make rational exit harder for you.

The dominant narrative goes like this: people who stay with hard things are admirable. People who leave hard things are quitters, failures, or moral weaklings. This narrative has legitimate applications. There are real virtues in persistence, and real costs to a culture that treats all commitment as contingent. But it gets applied to you indiscriminately, with no mechanism for telling apart the persistence that builds something valuable from the persistence that just compounds a mistake.

Duke’s Quit makes this argument in detail. Survivorship bias is the mechanism that makes the narrative feel true to you. You see the people who persisted through hardship and succeeded. You hear their stories in books, films, and podcasts. You don’t see, or you quickly forget, the far larger population who persisted through hardship and failed even bigger, because they couldn’t exit when exit was warranted. Your sample is systematically biased toward success stories. That makes persistence look like the explanation for success, when it’s really a correlate of a subset of successes that was always going to work out anyway.

The cultural celebration of the entrepreneur who “never gave up” is the business version of this same distortion. For every founder who persisted through near-failure to eventual success, there are dozens of untold stories of founders who persisted through near-failure to actual failure, at far greater cost than an earlier exit would have produced. The selection effect is total. Only the success stories reach you at scale. The failure stories are the signal that never gets transmitted.

Here’s the practical implication for you personally. The cultural narrative that staying is virtuous and leaving is weak is a biased information source, built on a selected sample. Using it to calibrate your own exit decision is roughly like estimating your odds at Russian roulette by interviewing only the survivors. The relevant sample includes everyone who faced the decision. The cultural narrative includes only the people who made one particular choice and got a particular outcome.

This doesn’t mean quitting is always right, or that commitment isn’t a genuine virtue. It means your decision to continue or exit should rest on the actual prospective evidence in front of you, not on a narrative filtered through success. Use the protocol in place of the narrative, not alongside it.

Sunk Cost and the Identity Trap in Relationships

The relationship application deserves its own extended treatment. For you, this is where the stakes are often highest and the psychological forces most powerful. Kevin’s restaurant involved money. Relationship sunk cost involves your identity and your attachment. It often involves other people too — children, mutual friends, extended family — whose lives intersect with your decision in ways that make it more complex, but no less important.

The identity trap works like this. After enough sustained investment in a relationship, the relationship gets built into your identity. You’re not just in this relationship. You’re the kind of person who is in it. Leaving means changing not just your circumstances but who you understand yourself to be. That identity revision genuinely costs you — not in some performative “that’s hard” sense, but in real cognitive and emotional work that a lot of people find overwhelming. It’s also, when the relationship isn’t right, genuinely necessary.

The zero-base question for your relationships isn’t “would I stay if I revisited this right now.” That still activates the sunk cost frame through the word “stay.” Here’s a better version: if you met this person for the first time today, knowing everything you now know about them, would you choose to pursue a relationship with them? Your knowledge isn’t zero. It’s full and real, built over years. Framed this way, the choice isn’t contaminated by sunk cost, because it’s a fresh choice with full information, not a continuation of an existing state.

Picture a woman — call her Patricia — who spent eleven years in a marriage she’d describe as not actively terrible, but clearly wrong for both of them. She and her husband had grown into different people, with incompatible values and incompatible visions for their lives. Neither made the other miserable in dramatic ways. They made each other quietly smaller instead — not pursuing what they wanted to pursue, not becoming who they might become, because the relationship didn’t leave room for it anymore. She stayed because of what they’d built together: a house, a shared social world, the narrative of a long marriage both families had invested in emotionally. When she finally ran the zero-base question honestly on herself, the answer arrived immediate and clear. She’d describe it, later, as coming up for air after a very slow suffocation she hadn’t noticed until she stopped breathing it.

The divorce was hard. The social costs were real. The identity revision was demanding. None of that changed the accuracy of the evaluation she’d run. Two years after, she described her daily life as categorically better, in ways she couldn’t have predicted from inside the marriage, because her comparison set had been so limited back then. The sunk cost — the eleven years, the house, the social world — had been real. The prospective cost of continuing — more years of the same quiet diminishment — was also real. The calculation, done honestly, had a clear answer. She made it at forty-four. Earlier would have been better. But the years that followed were far better than the years that would have followed without it.

What Richard Thaler Would Say to You

Richard Thaler won the 2017 Nobel Prize in Economics for work on mental accounting and nudge theory, and he spent decades studying exactly the kind of decision Kevin faced. His framework of mental accounts, the psychological ledger-keeping that makes losses feel more costly to you than equivalent opportunity costs, is the cleanest explanation for why Kevin kept the restaurant open twenty-four months past the point his accountant told him to close it.

In Thaler’s framework, Kevin had a mental account for the restaurant. The account opened with his first dollar. Every subsequent dollar went into it. The balance was the total investment — not in a strictly financial sense, but as a psychological construct that needed to be settled before it could close. Closing it at a loss required a settlement that loss aversion made roughly twice as painful for him as an equivalent gain would have been pleasant. Kevin wasn’t being irrational in the sense of being random or illogical. He was being rational inside a psychological accounting system that weights losses roughly twice what classical economics predicts.

Here’s what Thaler’s research suggests for you about interventions. The most effective ones change the choice architecture, rather than trying to change how you feel about the choice itself. Make the rational exit the default, not something that needs a special action from you. Present the exit calculation in a format that makes future costs visible and concrete, not abstract and deferrable. Remove the social cost of exit by building it into the process as a normal outcome, not a failure. None of this requires you to overcome loss aversion through raw willpower. It changes your environment so the loss aversion matters less.

Here’s the practical translation for you. If you’re an entrepreneur, investor, or leader who wants to protect yourself against sunk cost traps, invest in the decision architecture, not just your own personal discipline. Build the quitting criteria into your governance documents. Build the cost analysis into your regular review cadence. Make the exit path as clear as the continuation path already is. That architecture protects you more durably than any amount of individual resolve, because resolve degrades under exactly the conditions where you need it most: high stakes, high emotional investment, high sunk cost.

The Compounding Cost: What Staying Too Long Actually Costs

One of the most insidious features of the sunk cost trap is that its real costs aren’t in the continuation period itself. They’re in what that period prevents you from doing. Kevin didn’t just lose an additional two hundred thousand dollars in his final twenty-four months. He lost two years in his mid-to-late forties when he could have been building something new. He lost the compounding of whatever he’d have learned, earned, and created elsewhere. He lost his health — real stress-related deterioration during those final two years that took another year to reverse. He lost relationships — two friendships strained past repair by the stress of a failing business. None of that shows up in any accounting of the restaurant’s losses. These are the real costs of the sunk cost fallacy, and they’re larger than the financial costs in almost every case, including yours.

Time is the resource you cannot renew. Money can be rebuilt, slowly and with difficulty. Time cannot. Every year you spend in a failing business, a wrong career, or a diminishing relationship is a year that’s gone for good. The sunk cost fallacy’s greatest damage isn’t that it makes you spend more money on bad investments. It’s that it makes you spend more of your finite time on situations that are costing you, while you could be doing something else entirely.

The opportunity cost calculation is the one that finally breaks the frame for a lot of people. Not “I’ve already lost this much, I can’t stop now,” but “how many more years am I willing to spend here, when those years could be doing something else?” That second question has a number attached to it. Most people, once they make that number explicit, find it’s much smaller than the default of indefinite continuation the sunk cost fallacy produces.

“If someone had told me in July 2018 that I could keep going for two more years and lose another two hundred thousand dollars, or stop now and start the next thing, I would have stopped. But nobody asked me that question. I was asking myself a different question — whether I could get the restaurant to work. Those are not the same question. One has a clear answer. The other never does.”

That’s Kevin, looking back, putting it in exactly those terms. Ask the clear question for yourself. The right question is always prospective. Given everything true right now, what’s the best use of the years ahead of you? Not “can I justify what I’ve already spent?” Not “does stopping mean I failed?” Those questions have sunk cost built into them already, and they produce the wrong answers every time. The right question is forward-looking, concrete, and honest. Everything else follows from getting that one question right.

Applying the Protocol: A Complete Worked Example

• You cannot afford to make less money while transitioning.
• Your twelve years of experience are not transferable to other fields.
• At your level, changing fields means starting over at entry level.
• The right thing for your family is to maintain current income, regardless of your own engagement.
• Your professional identity is permanently tied to this specific field.

Let’s make this concrete for you. Here’s a full walk-through on a realistic scenario — a composite of the most common sunk cost situations men bring into a conversation like this one.

Here’s the situation. You are a professional in your late thirties, on a specific career track for twelve years. You have a graduate degree in this field, real seniority, a title that took a decade to reach, and compensation that’s grown into your family’s financial anchor. You know, in honest moments, that the work is wrong for you. It doesn’t engage you. It doesn’t build toward anything you want. You’re increasingly certain the field itself is declining. But you have twelve years invested, a mortgage, two children in school, and a professional identity that feels inseparable from your role.

Step one: define the decision precisely. Not “should I think about maybe eventually doing something different.” The real decision is this: in the next twelve months, will you take specific, concrete action toward a career change, or not? Precision matters here, because vague decisions produce vague plans, and vague plans produce no action.

Step two: identify the assumptions embedded in staying. Here are five you’re probably carrying:

Step three: challenge each one directly. Can you really not afford a transition period? What’s the actual financial analysis? Is six months of reduced income survivable for you? Have you actually assessed which parts of your experience are transferable? Most senior professionals, doing this honestly, find far more of their skills transfer than they’d assumed. Are there real examples of people at your level who’ve changed fields without starting over? Almost certainly yes. A few conversations would tell you whether your assumption is accurate or just assumed. Is maintaining your income really right for your family, if the cost is your sustained unhappiness and years more in a declining field? These deserve your honest answers, not a rhetorical dismissal.

Step four: identify the bedrock truths. The prospective value of your current path is declining — that’s knowable from industry data. Your engagement affects your performance, your health, and your presence at home — that’s documented. The longer you stay, the more field-specific your skills become — that’s structural. Your financial constraint is real, but far more granular than “I can’t afford a change.” It has a specific number attached to it. That number is usually smaller than the barrier you’ve been assuming.

Step five: rebuild the plan. Not “quit my job and hope for the best,” but a structured, gated process:

1. Spend three months researching three specific alternative directions that use your most transferable skills.
2. Identify the credential, experience gap, or network development that makes you viable in each direction.
3. Build a twelve-month financial buffer while actively working toward the transition.
4. Set a specific transition date, not “someday,” and treat it as a hard commitment to yourself.

This is not a fantasy. It is a plan. The sunk cost fallacy was preventing you from making it, because a plan requires you to acknowledge that the sunk cost is sunk.

Step six: assign a probability. What’s the probability that staying five more years produces an outcome you’ll be satisfied with? What’s the probability that the transition plan, imperfectly executed, produces something better? Write both numbers down. In most versions of this scenario, honest probability assignment favors the transition strongly. The only remaining question is whether you’ll let that analysis drive your decision, or whether twelve years of sunk cost weight will drive it for you instead.

The protocol doesn’t make this decision easy for you. The transition is genuinely hard, genuinely risky, genuinely costly in the short term. What the protocol does is strip away the false costs — the psychological weight of a past investment that isn’t a real cost of your future — and leave you with the actual calculation. That calculation should drive your decision. Not the twelve years. Not the mortgage. Not the identity. The actual numbers, assessed honestly, compared honestly, decided on honestly. That’s the Exit Calculation Protocol. That’s how you stop staying too long in bad situations. Not because it’s easy, but because you now have the tools to see clearly what the situation actually is.

The Questions You’re Still Asking

You might be wondering how you apply this to your relationships without it feeling cold and transactional. Fair question. It isn’t about treating relationships as financial transactions. It’s about stopping your emotional investment in a relationship’s history from overriding your clear-eyed read on whether it’s currently good for both of you. The zero-base question isn’t “what’s the economic value of this person.” It’s “if I met this person today, knowing everything I now know, would I pursue this?” That question doesn’t strip the emotion out. It focuses the emotion on present and future reality, instead of on past investment. Genuine love and genuine commitment survive this question easily. Only the inertia of sunk cost fails it.

You might also be asking what the most reliable warning sign is that you’re in sunk cost thinking rather than rational persistence. Listen to your own language. If your justification is backward-looking — I’ve already invested so much, I can’t walk away, I’d be admitting it was wasted — you’re standing in sunk cost territory. If it’s forward-looking — I have specific evidence the inflection point is near, the next twelve months carry positive expected value — you’re standing in rational persistence instead. The sunk cost narrative always has a particular texture to it. It’s about what you’ve already done, never about what’s actually ahead.

Isn’t there something to be said for commitment, you might ask, for following through on your word even when it costs you? Yes, to other people. Commitments to other people carry real moral weight, and that weight belongs in your exit calculation as a genuine cost. The point isn’t that commitments don’t matter. It’s that a commitment to yourself, the internal story that you can’t quit because of what you’ve invested, is not the same as a commitment to others: agreements, obligations, relationships that touch someone else’s wellbeing. The first is sunk cost wearing a virtue label. The second is genuine moral accounting. Tell them apart, and weigh both honestly in your calculation.

Kevin’s story involves money specifically. You might wonder whether this works differently for time investments. The psychology is similar, but your sense of recoverability differs. Money feels countable — you can see exactly what you’ve lost. Time feels more ambiguous. It’s already gone regardless of what you do next, so the sense of throwing good after bad feels less visceral to you. That means your time-based sunk cost traps can be harder to catch, precisely because the loss accounting is less visible. The years on a career path, the months on a dead project, the days in a living situation you hate — these are real sunk costs. They just don’t show up as a number in your bank account. Arkes and Blumer found the same pattern for time as for money. The detection is just harder, because the meter isn’t as visible to you.

Last one, and it’s probably the one sitting closest to home for you right now. How do you handle the social stigma of being someone who “gave up”? Duke addresses this directly in Quit, and her answer is uncomfortable but clear. Other people will judge your exit decision mostly by whether it looks prescient or cowardly in hindsight, and you can’t control that. What you can control is whether your decision was analytically sound. Run the analysis. See whether the future expected value actually supports continuation. Decide on that basis, not on what other people would say about you. That’s the right process. Kevin’s accountant, his business partner, and three separate consultants all told him to close. Had he closed in July 2018, some people around him might have said he gave up too easily. He’d have kept two hundred thousand dollars and a lesson. Instead he has a much harder lesson. The social evaluation is a red herring. The analysis is the thing that actually matters, and it’s the thing you actually control.

One Number Worth Writing Down

Before you go, here’s one number worth writing down tonight, on paper, where you’ll actually see it again. Not a business number or a relationship number specifically — just the one that applies to whatever situation flashed into your mind earlier in this episode. How many more years, or months, or dollars are you currently willing to spend on it, on autopilot, without ever running the zero-base question on it honestly? Most men, when you push them to actually name that number, are shocked at how large it already is, and how quietly they’d agreed to it without ever deciding anything.

You didn’t choose the situation you’re in today by accident. You chose it, one small non-decision at a time, by letting the sunk cost carry you forward instead of asking the only question that ever actually mattered. That’s not a condemnation of you. It’s just what your brain does, what everyone’s brain does, under exactly this kind of psychological pressure. Kevin’s brain did it. James’s brain did it. Maria’s brain did it. Patricia’s brain did it. Yours is doing it right now, somewhere, on something you already know the answer to and haven’t let yourself say out loud.

You now have the protocol. You have the zero-base question, the recoverable-versus-sunk inventory, the future cost calculation, the quitting criteria, the outside perspective, and the discipline of separating the story from the numbers. None of it works if it stays theoretical for you. It only works if you actually run it, this week, on the one thing you already know needs it. Pick the situation. Write the zero-base question at the top of a page. Answer it honestly. Then do what the honest answer tells you to do, on whatever timeline you can manage, starting now instead of at some vague point later that never actually arrives.

The past investment is gone either way, for you and for everyone reading their own version of this same number. That part of your story is already decided, and no amount of staying will bring it back to you. What’s still open, entirely and only, is what you do with everything still ahead of you. Make that choice a real one, yours, made on purpose. Don’t let the weight of what you’ve already spent keep making it for you by default, the way it has been making it for you until now.

If you take nothing else from the last hour, take this. You are allowed to change your mind about something you were once completely certain of. Nobody who matters is keeping a ledger of your consistency, waiting to punish you for updating. The people worth having in your life will respect the honesty it takes to look at a bad number and say it out loud, far more than they’ll respect the stubbornness it takes to keep pretending the number is fine. You are not the sum of the decisions you’ve already made. You are the sum of the decisions still in front of you, starting with the very next one, and that one is entirely yours to get right.

Editorial StandardsCorrectionsMedical DisclaimerAbout Our ContentAffiliate DisclosureSite Map