You already know that sometimes the best move is no move. You still won't make it, because sitting still feels like giving up.
That framing is close but useless. Action Bias is not a general argument for doing less. Action Bias is the tendency to choose action over waiting when the evidence does not favour acting. The meme treats it as a personality setting. It is a situational trap and being good at your job does not exempt you from it.
How it works
The leading explanation is anticipated regret. When acting is the norm for your role, a bad outcome after doing nothing feels worse than the same outcome after doing something so waiting carries a penalty that action does not. The result: you act to discharge the discomfort, not to improve the outcome. The expected value of waiting never gets properly priced because waiting does not feel like a decision at all. Patt and Zeckhauser (2000) describe Action Bias as a penchant for action carried into settings where the reasons for acting no longer apply and found it in surveys of hypothetical environmental decisions: people preferred producing a gain over preventing an equivalent loss - a preference for being the one who makes something happen.
The canonical evidence comes from Bar-Eli and colleagues (2007), who analysed 286 penalty kicks in top leagues. Goalkeepers dived left or right roughly 93% of the time yet given where the kicks went, the authors concluded that staying central was the better strategy. The small number of central stays makes that advantage imprecise; it is not proof that standing still is always optimal. The dive rate is observed; the explanation is the authors' proposal. Drawing on norm theory, they suggested that the norm of diving may make conceding after staying central feel worse than conceding after a wrong dive. Their survey of 32 professional goalkeepers supported the claim that diving is the norm. It did not measure regret directly.
Also shows up in
- Look for it wherever acting is what the role is expected to do. A communications team in a crisis, an on-call engineer with executives in the incident channel, a manager with a board asking "what are you doing about this" - each sits in a role where doing something is the norm. The bias may intensify when someone is watching, because visible inaction creates social pressure but the sources cited here do not establish that effect.
- Knowing about it does not fix it because the discomfort is real-time and the correction is abstract. In the working two-system model CogniqOS uses (the cited studies did not test it), deliberate evaluation is slower and effortful, while urgency can arrive before that evaluation is complete. You can fully understand the goalkeeper statistics and still dive. Awareness does not automatically lower the anticipated cost of inaction. It only tells you the cost is miscalibrated.
Countermeasure
First ask whether delay can cause irreversible harm. If it can, see 'When this isn't a bias' below. If not, do not suppress urgency; interrogate what it is pointing to. Sometimes it is noise. Sometimes, especially in novel or high-uncertainty situations, it is picking up a real threshold crossing that deliberate calculation underweights.
Run a named hold whenever the action is reversible but feels urgent or new information arrives on a known timeline and waiting for it is cheap. Convert waiting into an explicit decision: "I am choosing to wait for the noon data." The point is to put waiting through the same evaluation that acting gets. Whether naming it also changes how an audience judges you is untested. Write the hold in five lines:
Decision: what action am I considering?
Waiting window: how long can I wait before waiting itself costs something?
New information: what exactly am I waiting for?
Trigger: what evidence would make me act sooner?
Review time: when will I reassess?
Stop rule: if the review time arrives and the evidence is still ambiguous, either make the smallest reversible move or extend the hold explicitly, with a new reason and a new deadline. If no reversible move preserves meaningful options, keep waiting or make the irreversible call explicitly. Ambiguity alone is not a trigger for action and it is not a reason to wait forever.
Match the response to the action:
Reversible, low-cost: tolerate a short delay and gather the information.
Irreversible or complex: write the expected-value comparison for acting and for waiting, plus a reversal or maintenance plan, before you move.
Time-critical: act but make the smallest reversible move that buys information or preserves options.
Where it hurts most
- Evidence status: the investing and crisis cases below are illustrations, not documented findings. The clinical case is survey data.
- Investing: a manager trades out of a sound position because holding feels passive. Whatever the replacement does, the trade itself creates costs and taxes, plus a new position you now have to watch and justify.
- Clinical care: Lyu and colleagues (2017) surveyed 2,106 US physicians; the interpolated median estimate was that 20.6% of medical care is unnecessary - a physician-reported figure, not a measured rate. The reasons they cited most were fear of malpractice (85%), patient pressure (59%) and difficulty accessing prior records (38%). Those pressures push toward intervening but they are not evidence that Action Bias causes overtreatment.
- Crisis communications: a team ships a fast response that reads as decisive internally, then has to publish a correction once fuller information arrives. The correction erases the credibility the speed was meant to buy.
The uncomfortable version
The uncomfortable part is that competence does not protect you. The goalkeepers in Bar-Eli's data were elite professionals and they still dived. What exposes you is less your competence than the role it earns you, one where you are visible and expected to respond. And if you pride yourself on being decisive and responsive, waiting cuts against how you see yourself.
The second-order consequence most people miss: premature action creates a new state that now has to be managed. If you add an approval step to calm one incident, you now own that step and every complaint about it. Every move made to relieve discomfort leaves behind something you now have to maintain or defend. You do not just waste the move. You buy its consequences on installment. That compounding tail is why the protocol above asks for a reversal plan before any irreversible move.
When it doesn't apply
Action Bias is not the problem when the cost of delay clearly exceeds the expected cost of acting too early. In crisis response, market reaction or live negotiation, delay can cede ground faster than error costs you and moving now is correct. Action Bias is most suspect in stable environments where reflection is nearly free and the choice is reversible. Waiting in a time-critical setting is its own failure mode.
Danger zone
Affect Heuristic + Overconfidence Effect + Action Bias - a vivid emotional trigger makes acting feel urgent, Action Bias makes acting feel necessary and overconfidence makes the specific action feel obviously correct. You then move fast, hard and without a reversal plan precisely when you should be waiting. Treat this as a warning about a dangerous combination, not a settled claim that the three interact. (The Affect Heuristic - the tendency to let emotional response substitute for probability judgment - is covered in a separate CogniqOS entry.)
Also a component of Convergence Trap 5: The Panic Decision →
Stress test
These questions surface Action Bias where it is operating. The recurring test: what does waiting cost and what evidence makes acting worth that cost? They are not a case against acting: if the evidence clearly favours acting, act.
1. You are a founder and a key hire just quit. Within an hour you have started drafting a reorg of the whole team. What evidence says a reorg is warranted now, rather than just making the situation feel addressed?
2. A dashboard metric dropped yesterday. You scheduled three meetings to address it before anyone confirmed whether the drop is real or a tracking artifact. What evidence would have justified the meetings if no one could see your calendar?
3. You are negotiating and the other side has gone quiet for two days. You feel compelled to send a new offer. Does the silence actually cost you something or does only your waiting feel costly?
4. You held an investment flat for six months and now feel the urge to rotate into something new. What evidence says the switch beats holding, separate from the fact that holding feels passive?
Testable prediction
If decision-makers write a named hold (decision, waiting window, evidence trigger, review time and stop rule) before acting, then premature actions in reversible, non-time-critical decisions decrease compared with decision-makers who are simply told to wait. Premature means reversed or abandoned before the stated review time or judged by an independent reviewer to have lacked a decision-relevant basis when taken. Research hypothesis: if people privately record an initial act-or-wait call before announcing it, then unnecessary actions decrease relative to the same decision made only under observation. Unnecessary means later reversed, abandoned or judged by an independent reviewer to have had no decision-relevant basis at the time.
Run this audit with your AI
Paste your decision in, and let your assistant audit it through this bias.
Audit the following decision for Action Bias - the tendency to choose action over waiting when the evidence does not favour acting. MY DECISION: [paste here] Audit: 1. Am I in a time-critical regime where delay cedes irreversible ground, a stable reversible one where reflection is nearly free or an irreversible and complex one? If it is reversible, why does it feel urgent anyway? 2. What assumptions drive the expected value of acting now and what assumptions drive the expected value of waiting? Which of them are observed evidence, which are forecasts and which are discomfort being treated as evidence? 3. Is new information arriving on a known timeline and what does waiting for it actually cost? 4. Would I make this same move if no one could see whether I acted? 5. What future workload does this action create that I will have to maintain, defend or correct? 6. What is the smallest action that preserves options without pretending the uncertainty is resolved? Keep this in your stack. Next trap in the chain: Plan Continuation Bias.
Related concepts
- Plan Continuation Bias - both substitute momentum for reassessment, except here the momentum is toward new action rather than toward finishing a committed course. Covered in an upcoming CogniqOS post.
- Decision Fatigue - depleted deliberation makes the fast somatic urge to act easier to follow and the slow expected-value check harder to run. Covered in a separate CogniqOS post.
- Omission Bias - a distinct tendency, where people prefer inaction because harm from acting feels worse than equivalent harm from doing nothing; it fires under different framings of the same choice. Action Bias and Omission Bias are competing preferences that arise from different perceived costs, not sequential steps in a causal chain - which one dominates depends on how a decision is framed. Covered in the prior CogniqOS post.
Also connected in the mapInformation Bias
Origin
Sources
- Bar-Eli, Azar, Ritov, Keidar-Levin & Schein 2007 - Action bias among elite soccer goalkeepers: The case of penalty kicks. Journal of Economic Psychology 28(5).
- Patt & Zeckhauser 2000 - Action Bias and Environmental Decisions. Journal of Risk and Uncertainty 21.
- Lyu, Xu, Brotman et al. 2017 - Overtreatment in the United States. PLOS ONE 12(9).
Read next
Entry #53 of 631 in The Lexicon · see the full Lexicon