Stop Overthinking

Overview

Thinking is a tool for reducing uncertainty — and overthinking is what happens when the mind keeps applying that tool to problems the tool cannot solve. You cannot think your way to certainty about a future that does not exist yet, but the loop keeps spinning anyway, because the alternative — committing to an answer — feels premature and exposed. This chapter is not an argument for thinking less. It is an argument for deciding at the right resolution: big enough to act on, small enough to revise. The fix for a mind that cannot stop is not silence; it is a smaller decision, made sooner.

01 — Why the Loop Runs

Unfinished goals intrude on attention until they are discharged or planned. That single mechanism explains the entire experience: an undecided question is a permanently open loop, and your attention is drafted into it every time it surfaces — in the shower, at the wheel, at three in the morning. The loop is not a failure of character; it is attention doing its job with no exit sign. The escape is not more analysis. It is completion: deciding, and then writing the next action down. A plan is the loop's discharge valve — the goal stops intruding because the mind knows where it stands. You are not required to solve the universe; you are required to close the question you actually hold.

02 — The 70% Solution

The Marine Corps teaches officers to make decisions under uncertainty at around seventy percent of the information they might want — doctrine taught at The Basic School and popularized in David Freedman's Corps Business, and the same logic Jeff Bezos describes using. The doctrine exists because the cost of waiting is real: by the time you reach ninety-five percent, the decision has usually been made for you — by events, by other people, by the clock. The honest version of the rule is a question about reversibility. If the decision is cheaply reversible, decide at seventy percent today. If it is expensive to reverse, do not wait for certainty — shrink the stakes instead, and test the decision at a size you can afford.

03 — Decision Drift and the Long Session

There is a well-known study of Israeli parole judges in which favorable rulings declined across long sessions and rebounded after breaks. Read it honestly: the case order was not random, the effect size is contested, and a later registered study of physicians found nothing. What survives a close look is the direction: decision quality degrades over long sessions. The pattern shows up across domains — not because the brain is a fuel tank with a limited supply, but because sustained mental work degrades judgment. The practical translation is unglamorous. Do your big decisions early in the day. Take breaks. Remove trivial choice loads from decision-heavy days. Never quote the headline percentages; they are the part of the story that did not replicate.

04 — The Decision Protocols

Run every parked decision through the resolution ladder. First: is this reversible? If yes, decide now. Second: if it is expensive to reverse, can you test it at small stakes — a trial month, a pilot, a smaller version? Third: decide at seventy percent and write the next action. Fourth: set a tollgate — a scheduled review date when the decision gets revisited with new information. "Sleep on it" is reserved for the genuinely irreversible, high-cost cases; everywhere else it is procrastination wearing a wisdom costume. Then run the audit: keep a list of the decisions the loop actually improved. You will notice the count stays near zero. The loop's claim to fame is that it prevents mistakes — and the audit shows what it mostly prevents is motion.

Overthinking is a refusal to let the answer be small.

05 — Attention as the Antidote

Overthinking is not a thinking problem; it is attention parked on an open loop. The One-Task Rule and the Winner's Loop give that attention somewhere else to be: one object at a time, closed loops, and a record of what got done. When the question reopens mid-day — and it will — the move is not to argue with it but to write the next action and let it go. Paper is cheaper than rumination, and the written action is the loop's discharge valve in miniature. You cannot order yourself to stop thinking. You can give the thought a plan, and the plan releases the attention.

Key Ideas

• Overthinking is a tool mismatch: analysis cannot remove uncertainty.
• Open loops occupy attention until they are discharged or planned.
• Decide at seventy percent; waiting for ninety-five is waiting for someone else's decision.
• Decision quality degrades over long sessions; schedule big decisions early.
• A written next action closes a loop more cheaply than another hour of thought.

Apply Today

1. Pick one decision you have been circling for more than a week.
2. Rate its reversibility: can it be undone, and at what cost?
3. Take the seventy-percent stance: decide with the information you have, today.
4. Write the single next action on paper, in one sentence.
5. Move your standing decision reviews to the morning, before the day fills.

The Science

Danziger and colleagues (2011) examined 1,112 parole decisions made by eight Israeli judges and found that favorable rulings declined across a decision session and rebounded after breaks. The finding is graded B/C because the case order was not random, later analyses contested the magnitude, and a 2025 registered report found no similar effect in physicians. What it licenses is the modest, well-replicated claim that decision quality degrades over long sessions — which is why the protocol here is about scheduling, not about willpower reserves.

The Marine Corps 70% rule is doctrine rather than study: taught at The Basic School, popularized in David Freedman's Corps Business (2000), and echoed by Jeff Bezos. It is graded A as doctrine — a decision-making standard under uncertainty — and it licenses acting at seventy percent rather than waiting for certainty, with reversibility as the qualifier.

Masicampo and Baumeister (2011) showed that unfulfilled goals intrude on attention until they are discharged or planned — the empirical basis for the plan-making advice in this chapter. Note what this study does and does not license: the older claim that interrupted tasks are remembered ninety percent better has not replicated, but the pull to resume unfinished work is real. Planning closes the loop; the memory lore does not survive scrutiny.