Why Good Leaders Make Poor Decisions Under Sustained Pressure

judg·ment | noun | the ability to make considered decisions or come to sensible conclusions.

Great definition. Also completely useless for explaining why a normally thoughtful executive can spend eighteen months building a reputation for careful, well-reasoned decisions, then during one particularly brutal quarter approve something so baffling that the board minutes just say “discussion followed.”

Here is the uncomfortable possibility. It might not have been him making that call. Not entirely. There is a decent chance someone else was driving.

Meet the Strategist (and His Understudy)

Under normal conditions, your decisions get made by what we will call the Strategist. He is patient. He can hold two conflicting ideas in his head at once without panicking. He notices nuance, remembers long-term consequences, and is generally the reason people trust you with hard calls. In brain terms, he lives mostly in the prefrontal cortex, the part responsible for working memory, impulse control, planning and emotional regulation.

The trouble is, the Strategist has an understudy. Let’s call him the Autopilot. He is fast, decisive, allergic to nuance, and has exactly one setting: handle it now. Under normal circumstances he stays backstage. Under sustained pressure, he quietly steps in, and here is the unsettling part. Nobody actually approves this handover. It just happens.

It does not happen all at once, either. Think of it less like a light switch and more like a dimmer. The room does not go dark. It just gets steadily less bright, slowly enough that nobody in the meeting notices, including the leader running it. The slides are still visible. The conversation still sounds normal. But perspective has narrowed, patience has thinned, and ambiguity now reads as a personal threat rather than an interesting problem. The Strategist is technically still in the building. He is just not the one answering the phone anymore.

Why the Swap Happens

This is not a character flaw. It is closer to a staffing decision made by your own nervous system. Research on stress and the prefrontal cortex has found that stress signalling can rapidly impair exactly the functions the Strategist relies on, including working memory, flexible thinking and impulse control, while simultaneously strengthening more reflexive, faster-acting circuits (Arnsten, 2009). In plain terms, stress does not just distract the Strategist. It actively hands the controls to someone else.

This is not necessarily a design flaw either. In an actual emergency, you want the Autopilot. He is quick, decisive, and does not waste time weighing six options while a bus is bearing down on you. The problem is that he has been given a permanent parking spot in the boardroom, where the threats are rarely buses and mostly things like ambiguous market data and a colleague who disagreed with you in the all-hands.

The Autopilot’s Operating Manual

Once he is in charge, the Autopilot does not run your normal decision-making process, just faster. He runs a completely different one. A meta-analysis of stress and decision-making under uncertainty found that stress tends to push people toward more reward-seeking, more risk-taking and less reliance on careful executive control (Starcke & Brand, 2016). Related research found that stress can shift behaviour away from deliberate, goal-directed choices and toward old habits instead, meaning people fall back on whatever they have done before rather than actually evaluating what is in front of them right now (Schwabe, Höffken, Tegenthoff, & Wolf, 2011).

Which is a fairly precise description of why a leader who normally listens starts interrupting, why someone who values empowerment suddenly wants to approve every decision personally, and why a person who usually thinks in years starts obsessing over this week’s numbers. It is not that their values evaporated. It is that the Autopilot does not have access to them. He is running an older, blunter playbook, and it is the one that gets grabbed first whenever the Strategist clocks out early.

Why Your Best People Are the Most at Risk

Here is the part that should worry you more, not less. It is usually your strongest people who hide this the longest. Good leaders absorb tension, protect their teams from unnecessary noise, and keep functioning long after a less capable person would have visibly cracked. Which sounds like a strength, and mostly is, right up until it becomes the exact reason nobody notices the handover has already happened.

They do not collapse. They compensate. The Autopilot, run by someone genuinely skilled, can look a lot like confident leadership for quite a while. It is just leadership running on a much shorter, blunter script than the one that got them the job in the first place.

The Blind Spot at the Top

Most companies review bad decisions the same way. What did the data say, who signed off, where did the process break down. Nobody on that review ever asks the one question that might actually explain it. Who was actually driving when this decision got made?

That is not a soft question. It is closer to a supply chain question, the same way asking whether a machine was properly maintained is a supply chain question. A leader can complete every leadership programme you throw at them, learn every framework, nod along in every workshop about psychological safety, and still hand the wheel to the Autopilot the moment three bad weeks stack up in a row. That is not the training failing. That is the training being aimed at the Strategist while the actual decision got made by someone who never attended a single session.

What Leaders Might Reconsider

Worth asking, honestly, next time a decision goes sideways and nobody can quite explain why. Were the people making this call actually running on empty? Do your meeting rhythms leave any room for reflection, or is it wall-to-wall reaction? Is there a real difference in your culture between useful urgency and someone quietly panicking with better posture? Are your best people ever rewarded for pausing, or only for being available at all hours?

None of these are HR questions dressed up to sound important. They are closer to a maintenance schedule for the machine that actually makes your decisions.

The Bottom Line

Good leaders do not usually make bad decisions because they have suddenly become bad leaders. They make bad decisions when the Autopilot has quietly taken the wheel and nobody, including them, noticed the handover.

Which changes the actual question worth asking. Not “do we have good leaders,” because you probably do. The better one is: are we creating the conditions where the Strategist actually gets to stay in the room, or are we just handing the Autopilot the keys every single quarter and calling it resilience?

This article is part of our ongoing exploration of Sustainable Performance through the lens of human physiology.

Author: Giorgio Bruins

Date: September 2026

Sources

Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience, 10(6), 410–422.

Starcke, K., & Brand, M. (2016). Effects of stress on decisions under uncertainty: A meta-analysis. Psychological Bulletin, 142(9), 909–933.

Schwabe, L., Höffken, O., Tegenthoff, M., & Wolf, O. T. (2011). Preventing the stress-induced shift from goal-directed to habit action with a β-adrenergic antagonist. Journal of Neuroscience, 31(47), 17317–17325.