Judging likelihood or feeling regret based on how easily an alternative outcome can be mentally imagined.
Imagine you're at a carnival and you throw a ball at a stack of bottles. If you miss by a mile, you shrug it off. But if the ball just barely grazes the top bottle and it wobbles but doesn't fall, you feel way worse — because your brain instantly plays a little movie of the bottle falling over with just a tiny bit more force. Your brain judges how bad or good something is by how easily it can imagine a different ending.
The simulation heuristic describes the mental process by which people construct imaginary scenarios—alternative versions of events that did or could happen—and use the ease of that construction to judge probability, causality, and emotional impact. When an outcome feels easily 'undone' through small mental tweaks (such as leaving five minutes earlier to catch a flight), the person experiences stronger emotional reactions like regret or relief, regardless of whether those tweaks would have actually changed the objective probability. This heuristic is especially powerful in near-miss situations, where the psychological distance between what happened and what could have happened is small and vividly imaginable. It also drives anticipatory regret, causing people to avoid decisions where they can easily picture a negative outcome unfolding.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors experience outsized regret over stocks they nearly bought that subsequently soared, leading to impulsive 'revenge' purchases of similar assets. The ease of mentally simulating 'I almost clicked buy' distorts future risk assessment more than actual portfolio losses from positions they held.
Patients who narrowly miss qualifying for a clinical trial or treatment threshold experience greater distress than those who were far from eligibility, even when the medical prognosis is identical. Clinicians may also feel disproportionate guilt over adverse outcomes that followed small, easily reversible decisions (choosing drug A over drug B) compared to outcomes from standard protocols.
Students who fail an exam by a single point experience dramatically more regret and self-blame than students who fail by twenty points, despite the identical consequence. Teachers may also dwell more on students who 'almost' passed, allocating disproportionate mental energy to those near-miss cases.
People tend to ruminate far more intensely about relationships that 'almost worked' — the partner who was nearly perfect but for one issue — than about clearly incompatible matches. This mental replaying of small mutable details ('If only I hadn't said that one thing') can delay emotional closure.
Users who experience a system crash right before saving their work report far greater frustration than users who lose work early in a session, because the near-save is trivially easy to mentally undo. Product teams can exploit this by implementing auto-save to eliminate the high-regret 'near miss' of lost work.
Hiring managers who nearly chose a candidate that went on to succeed spectacularly elsewhere experience outsized regret that biases their future hiring toward similar profiles, regardless of whether the pattern is statistically meaningful. Teams also dwell more on project failures caused by last-minute reversible decisions than on those caused by deep structural problems.
Election near-misses (razor-thin margins) generate vastly more counterfactual analysis and emotional intensity than landslides, with media and voters endlessly debating the small, mutable factors ('If only turnout had been slightly higher in one county') while ignoring broader structural causes.
Daniel Kahneman and Amos Tversky, first presented in a 1979 lecture and formally published as a chapter in 'Judgment Under Uncertainty: Heuristics and Biases' (1982).
The ability to mentally simulate alternative outcomes likely evolved as a planning and learning mechanism. Ancestors who could vividly imagine 'what would have happened if I had gone left instead of right' could extract causal lessons from near-misses with predators, failed hunts, or environmental hazards — refining future behavior without paying the full cost of repeated trial and error.
Generative language models trained on human text absorb the simulation heuristic's patterns by overrepresenting vivid, easily-narrated counterfactual scenarios in their outputs. When asked to assess risk or generate explanations, LLMs may emphasize 'near miss' narratives and mutable factors over statistically more relevant but harder-to-imagine base rates, mirroring human tendencies to equate imaginability with probability.
50 cards are free to preview. Buyers unlock the rest of the deck plus the interactive training — Spot-the-Bias Quiz unlimited, Swipe Deck with spaced repetition, My Blindspots, Decision Pre-Flight, the Printable Deck + Cheat Sheets, and the Field Guide e-book. $29.50$59.
