š Mind the Failure Gap
New research reveals that we systematically underestimate how often people fail, and this shapes our beliefs, politics, and relationships
Failure is more common than we realize. Each day, we might pick up someone elseās order at the coffee shop, say the wrong thing at a dinner party, or leave embarrassing typos in our newsletters. It was Benjamin Franklin who claimed that ānothing can be said to be certain, except death and taxesā. Well, Iād like to add a third depressing guarantee: failure is certain too.
Bizarrely, despite the ubiquity of failure around us, we canāt help but judge both ourselves and others a little too severely when failures happen. When we burn the meal weāre cooking, our stress levels hit the ceiling. When someone mistakenly cuts us off in traffic, we hurl expletives at our windows. When our partner forgets to pick up the one thing we requested from the grocery store, we momentarily consider whether to search online for ābest divorce lawyers in my areaā.
A new set of studies suggests much of this harshness and misunderstanding may be driven by a blind spot in our failure detection systems. People are systematically unaware of the true prevalence of failures around them, and this misperception can impact how we treat others at every level of society.
š§Ŗ What did the researchers do?
In a 2026 study published in Journal of Personality and Social Psychology, researchers tested peopleās intuitions about the prevalence of failure.
First, they recruited hundreds of online participants and asked them to estimate peopleās general rate of failure across more than 30 domains of everyday life. This included the percentage of crimes for which police fail to identify a suspect, the percentage of new businesses that fail, the percentage of new drugs that never receive FDA approval, and the percentage of romantic relationships that end in breakups.
The researchers then compared participantsā estimates to verified official statistics and national records. They also analyzed news articles, social media posts, and online consumer reviews, searching for evidence of how often people actually discuss failure in the real world outside the lab.
Finally, they also tested whether simply telling people the true rate of various failures in the world would meaningfully change their attitudes and behaviors. For this, participants were tested within their real-life roles: educators at a professional conference, hiring managers, and citizens weighing in on policy questions such as mass incarceration.
š What did the research find?
People consistently underestimated the rate at which people fail. On average, failures occurred around 61% of the time across the life domains studied, yet people estimated the failure rate to be only around 41%.
In many cases, this was not a subtle gap. For example, people thought TSA screening missed about 14% of weapons at airport security when the true rate is closer to 70%. They also estimated that roughly 35% of US college students fail to graduate on time when the true figure is closer to 64%. They even believed that only 25% of wildlife on Earth had gone extinct since 1970 when the actual figure is around 70% (I guess you could think of this as either a failure of wildlife survival or a failure of human conservation).
The researchers refer to this underestimation effect as āthe failure gapā.
They attribute this gap at least partly to under-reporting of failure in the information we share with each other, including in the news and social media. People are resistant to publicly discussing failureāperhaps because it threatens their self-esteemāso their perceptions of failure across society as a whole become misaligned with reality. Even online product reviews showed this kind of pattern: common medications like Tylenol fail to provide meaningful relief for around half of users, yet fewer than 4% of reviews on major retail platforms rate the product below four stars. People even hesitate to admit that a drug might not work for them.
So what happens when people are more willing to discuss failures? The #MeToo movement normalized the sharing of experiences of sexual misconduct. Sexual harassment and abuse went from failures that were swept under the rug to failures that were all over mainstream news and household conversations, so the researchers tested whether peopleās estimates of sexual misconduct would be more accurate in this specific case. Consistent with their predictions, the failure gap not only disappeared from peopleās estimates on this topic, it reversed entirely: people overestimated the likelihood of sexual misconduct in the general population.
The same participants simultaneously underestimated womenās health failures like heart disease and cancer diagnoses, showing that two very negative topics can produce opposing perceptions based on how widely they are discussed.
The researchers finally tested whether closing the failure gap on certain issues would change peopleās real beliefs and behavior across a range of policy domains. When people learned the true higher-than-expected rate at which people miss court summons, they were less likely to support imprisonment for failing to appear in court. Similarly, one in five educators retracted their support for school suspensions after learning how many students commit infractions annually.
In other words, knowing the true scale of a problem deeply impacts how people react to it. Far from being an abstract academic exercise, the failure gap affects peopleās decision-making about the most important issues on the planet. This is a serious challenge when you consider how widespread the failure gap really is. As the researchers put it in their paper:
āPeople underestimated tens of thousands, and in some cases, millions, of failures⦠they were unaware of millions of adults with poor educations, poor relationships, and declining mental healthā
āļø Takeaway tips
#1. Assume that everybody is failing
The failure gap is large, consistent, and everywhere. Whatever domain youāre thinking aboutāhealth, business, education, relationshipsāyour intuitive sense of how often things fail is probably an underestimate. And importantly, rare-seeming failures attract harsher judgment. Before you judge yourself or someone else for making a misstep, ask yourself whether you actually know the true prevalence of the failure in question. Knowing how many people make that same misstep every day may help to put things in perspective and make you feel a little more compassionate toward yourself and others.
#2. Be skeptical of what gets shared
Online reviews, news articles, and social media posts systematically under-represent failure. People share their successes more than their struggles, companies curate positive feedback, and the news tends to cover only the most dramatic failures rather than the most common ones. Rather than proving that failures are rare, this may simply mean that people find it uncomfortable to discuss failure. Treat the absence of bad news as a potential data gap, not evidence that everyone around you is flawless.
#3. Communicating failures is convincing
Correcting the failure gap for people has consequential effects. It not only aligns them more closely with reality but actively influences how they feel about important social issues ranging from employment to mass incarceration. Whether youāre trying to make up your own mind or trying to convince someone else about the best course of action on a consequential issue, make sure youāre aware of any possible failure gaps and double-check the stats.
āEvery failure is a step to success. Every detection of what is false directs us towards what is true: every trial exhausts some tempting form of error. Not only so; but scarcely any attempt is entirely a failure; scarcely any theory, the result of steady thought, is altogether false; no tempting form of Error is without some latent charm derived from Truth.ā
~ William Whewell



I didn't appreciate the failure rate size at all. I'm currently writing about failing to adhere to a medical regimen even with financial incentives and thought that a 25% adherence rate looked way too low.
But this context really shows that it is far more common.