What the data actually shows
Decision researcher Annie Duke, in her 2022 book Quit, argues that people systematically quit too late rather than too early. Her account draws on well-established behavioural findings: the sunk-cost fallacy, where we keep investing in something because of what we have already put in rather than what it will return; loss aversion, which makes giving up a position feel like an unbearable loss; and the identity trap of seeing yourself as 'not a quitter,' which turns leaving into a threat to your self-image.
A core part of the problem is that the costs of persisting are largely invisible. Economists call this opportunity cost — the value of the best alternative you forgo by staying. When you persist at a poor-fit job, project or relationship, the bill does not arrive as an obvious loss; it arrives as the better thing you never got to do, which is hard to see and easy to discount. The sunk-cost fallacy compounds this by making the visible past investment feel more decisive than the invisible future return.
At the same time, persistence genuinely matters for the right goals. Angela Duckworth's research on grit — passion and perseverance toward long-term goals — finds it predicts success in some demanding settings. The two views are not actually in conflict. Grit is valuable when applied to a well-chosen, good-fit goal, and costly when applied to a poor-fit one. The unanswered question grit alone does not settle is whether this particular goal is the one worth your perseverance.
Why this feels different from how it actually is
Quitting feels worse than the evidence says it should because the language around it is almost entirely negative. 'Quitter,' 'giving up,' 'throwing in the towel' — the vocabulary frames leaving as a character flaw, while persistence gets the heroic words. We absorb a one-sided story in which the only admirable choice is to keep going, no matter the cost.
Loss aversion makes it sharper still. Abandoning something you have built means accepting a definite, immediate loss now, while continuing keeps alive the hope, however faint, that it might still work out. Our minds weight the certain loss heavily and the dim future possibility generously, which biases us toward staying past the point of good sense.
And much of our identity gets quietly fused to our commitments. Quitting a career, a course of study or a long relationship is not just exiting an activity — it can feel like contradicting who we have told ourselves and others we are. That identity cost is real, but it is separate from whether continuing is actually the better decision, and it tends to keep people in place far longer than the underlying facts warrant.
What the research says to do about it
The most practical idea from this literature is setting 'kill criteria' in advance — deciding, before you are emotionally entangled, what signals would tell you to stop. Duke recommends defining specific conditions ('if by this date I have not reached X, I walk away') while you can still think clearly, because in the moment, sunk costs and identity make honest reassessment very hard. Pre-commitment moves the decision to a calmer vantage point.
Weighing the decision in terms of opportunity cost rather than what you have already invested is the central correction. The right question is not 'how much have I put into this?' but 'knowing what I now know, would I start this today?' If the answer is no, the past investment is sunk and irrelevant; the only thing that matters is the future return compared to the best alternative.
Reversibility is worth factoring in too. Decisions that can be undone cheaply deserve faster, lower-stakes calls, while genuinely irreversible ones merit more caution before quitting. The aim throughout is not to quit more or less as a default, but to make the persist-or-quit decision deliberately, on the merits, rather than by reflexively defaulting to 'never give up.'
What the research says does not help
Treating 'never give up' as a universal rule does not help, and the research suggests it actively contributes to staying too long. Persistence is a tool, not a virtue to be maximised; applied to a poor-fit goal it simply spends finite time and energy on something the evidence already says will not pay off.
Justifying continuation by pointing to what you have already invested is the sunk-cost fallacy by definition, and it is one of the most reliable ways to make a bad situation worse. Money, years and effort already spent cannot be recovered by spending more; they should not be in the decision at all.
Equally, quitting impulsively at the first sign of difficulty is not the lesson here. The point is deliberate assessment, not a bias toward leaving. Bailing out the moment something gets hard can mean abandoning a good-fit goal right before the part where perseverance would have mattered — which is exactly where grit earns its keep.
Real numbers in context
This is a topic where the honest material is conceptual rather than a tidy statistic, and it is worth being upfront about that. The core claims — that people quit too late more often than too early, that sunk costs and loss aversion bias us toward staying, and that opportunity cost is the hidden bill — come from well-replicated behavioural-economics findings and from Duke's synthesis of them, not from a single headline number you should over-trust.
The two reference points worth holding are the directions, not figures. Persistence (grit) has measurable value for well-chosen, long-term goals, per Duckworth's research. And the cost of staying too long is real but invisible, paid in forgone alternatives rather than obvious losses — which is precisely why it is so easy to ignore and why deciding on the merits, rather than by slogan, is the skill that matters.