Biases and Blind Spots

“Just the facts. Ma’am,” (1) may be the perfect catchphrase in a police drama, but facts are not enough to solve problems. Certainly not the complex problems we find in our world today. Everyone has their own worldview, a bias that blinds one person from seeing not only what another person sees, but the way he or she sees it. We each are packed with a set of assumptions. We have likes, dislikes, and prejudices. Different things trigger us and cause us to be afraid. An obsession for one individual means nothing to another. One situation causes a conflict of interest for one, where the other has none. These limitations keep us from seeing what we could and would see if we did not have them.

We are not inflicting these intrusions on truth on one another maliciously. They are usually unconscious. Unaware of these internal roadblocks, it is hard to address them.

Unless you had an economics professor who punished you with impossible models of behavioral bias!

While I won’t take you through the math, I can reassure you there are rational explanations for the seemingly random choices different individuals make in the face of the same set of “facts.” The theories call these “mistakes”, but they are really predictable behaviors in decision-making that depart from rational choice.

Let’s use a scenario called “disaster response” to examine a common bias called “overconfidence.” Individuals who lean towards overconfidence see the world as more manageable than it is. They trust their judgment, minimize uncertainty, and assume the situation will work out favorably because they are the ones handling it. Their Pollyanna-like approach leads them to underestimate obstacles, make impulsive decisions and double down after mistakes. They need “just one more try,” then feel blindsided when reality contradicts their beliefs.

It is one thing if an individual is overconfident about whipping up the perfect Bechamel. It is light years different if the individual is a civic leader addressing an approaching hurricane. Overconfidence can translate into an overestimation of preparedness. Leaders may rely too heavily on forecast models tracking the storm and put off evacuation orders. Meteorological models are becoming more accurate over time, but are still imperfect. Assuming too much precision can cause under-preparation or misallocation of resources.

Anticipating local shelter capacity and medical surges is a key part of disaster response management. Failure to plan for these with clear sight has life and death consequences. With any hurricane comes transportation logistics. The coverage area is spread out; downed wires and trees block roads. And debris management is monumental.

Residents too, display overconfidence bias. Living in a hurricane-prone area, I have heard these irrational (my definition) statements: “It won’t be that bad,” “My house survived Hurricane (fill in the blank)’” and my favorite “I’m going to ride this one out.” Residents delay evacuation or stay in place because leaving feels like losing their home, possessions, or control. They stay until the risk is undeniable. Sometimes it is too late. Now we are into loss aversion, a different kind of blind spot.

If we telescope out from this one disaster and one set of responses based on distorted information (that is what bias really is), we can see overconfidence bias on display everywhere in government at all levels, no matter who is in power. Forecasts, by their nature, are wrong (2). Few politicians will get in front of a microphone touting the huge economic losses that are coming, when it is easy to be overconfident about growth and vague about the timing (in economics, we always establish the timing with our short-run and long-run curves). Likewise, with the unrolling of new programs. Budget savings are sure to compensate for, even over-compensate for, up-front investments. There is a compounding of unrealistic timelines, missed targets, and cost overruns that conspires to derail what we think are solid plans.

The best-laid plans can fall off the cliff if there is overconfidence in implementation. Policymakers who assume communities can instantly absorb mandates and execute them are fooling themselves. Capacity constraints, staffing needs or training, lack of data, and unclear communication are just a few of the bottlenecks between planning and achievement.

Even if the forecasts are on track and comprehensive implementation plans have been developed, there remains the hurdle of compliance. We have come full circle to the behavioral quirks of individuals who do not behave predictably. Opening a shelter doesn’t mean people will go there. These are unintended consequences.

Getting out in front of all these problems means we can’t use reason alone to get to the best solution. What we need here is a reality check. Some feedback to check the inclination towards bias.

The best way to do a reality check is to create friction by seeking disconfirming evidence. Ask, “What would prove me wrong?” In the disaster case, we can challenge overconfidence about how many shelters to open by asking many questions like: “What if the storm affects a larger population area?” and “What if certain shelters flood and can’t be used?” and “How can we make the most effective use of the spaces we have allocated?” Questions like these whittle away at the certainty of the original forecast, allowing for adjustments, not in facts, because there are no solid facts, but in judgment. There is a shift from looking for supporting evidence to gathering contradictory evidence, the fastest way to puncture false certainty.

Similarly, before committing to a plan of action, conducting a pre-mortem bypasses the Pollyanna-ism and highlights weak spots. By imagining the plan has failed, the idea is to work backwards to identify how and why the plan failed. This approach will drain the overconfidence out of most.

Expressing outcomes as a specific number or point conveys a confidence that few situations outside of strict engineering or computer design warrant. Even medical lab results are reported in ranges of normality. Communicating outcomes as intervals of uncertainty rather than exact numbers right-sizes the level of confidence in those outcomes.

A common practice seen in business and in academic writing is the independent review. By bringing in someone with no stake in the outcome, there is less danger of that person being caught up in the enthusiasm or assumptions of the originating group. This review can include stress tests that ask, “What if this takes twice as long?” or “What if that costs four times the forecast?” Such pessimistic scenarios create a more accurate view of risk by offsetting the overconfidence bias.

A humbling exercise is to track predictions. When people record their forecasts against reality, they can learn how often they miss (ask any accountant). A forecast versus actual record addresses the blind confidence that biases towards unfavorable outcomes. You can also compare the current situation to historical data or base rates. Asking “What normally happens in cases like this?” is more reliable than asking, “What do I think will happen?”

Overconfidence thrives on speed. The adrenaline surge that happens when we are in the driver’s seat wielding decisions out of sheer gut instinct is addictive. Requiring structured checklists, decision gates, and deliberate pauses forces people to examine facts and assumptions, and slow down decision-making instead of speeding towards potential disaster.

Group dynamics can do much to temper overconfidence bias. Groups become overconfident when no one pushes back. By encouraging dissent, like assigning a rotating “challenger role” or explicitly requesting objections, the likelihood of making decisions skewed toward bias can be reduced. By rewarding people for well-founded forecasts instead of boldness or ego, the role of overconfidence bias is diminished. When self-assurance is equated with competence, overconfidence flourishes.

It is true: people can’t see what they can’t see. Biases fall into this category. But the simple awareness that everyone comes to every interaction with some set of biases should be motivation for us to look for our own. How are they disabling us? When do they show up?

Where are our blind spots?

Love,

Alice

(1) “Just the facts, Ma’am,” is the misremembered phrase used by Sgt. Joe Friday in the TV series Dragnet. The actual phrase was “All we want are the facts, Ma’am.”

(2) There is a running joke attributed to political activist Lyndon LaRouche (1922-2019) that meteorologists and economists are paid to lie.