Human Performance & Leadership
The Pocketknife Decision
An outfielder who cannot compute a parabola still catches the ball, by running to hold one angle steady.
2026-09-151 min read
Leaders treat decision quality as a function of how much they consider, so the instinct under pressure is to gather one more dataset. Gerd Gigerenzer's work on fast-and-frugal heuristics points the other way: in environments with the right structure, a rule that ignores most of the available information can match or beat a model that weighs all of it. The reason is not that less thinking is magically better. It is that complex models fit noise as readily as signal, and a volatile environment supplies noise in bulk.
The cleanest illustration is the gaze heuristic. An outfielder chasing a fly ball does not solve for the trajectory, a problem that would demand the ball's velocity, its spin, and the wind at three different altitudes. He fixes his gaze on the ball and adjusts his running speed so that the angle of his gaze stays constant. Follow that one rule and you arrive where the ball lands, without ever having calculated where that is. The heuristic works because it exploits a regularity of the environment rather than modeling it, and the ignoring is the point rather than a compromise.
The discipline this demands of a leader is not choosing a simple rule but naming the cue the rule rides on, and the condition under which that cue stops predicting. Take a rule like escalate any customer issue that a second team touches. It works because handoffs correlate with unresolved ownership. Reorganize into a structure where handoffs are routine and the cue goes dead, while the rule keeps firing exactly as confidently as before. Most decision rules in an organization die this way, not overruled but quietly outliving the regularity that made them work. Write the cue down next to the rule, and you have something you can audit.
Key insights
A frugal rule beats a weighted model when the environment is noisy, because the model fits the noise too.
Every rule of thumb rides on one environmental regularity; when that regularity shifts, the rule fails silently.
The gaze heuristic solves an intercept problem without ever computing the trajectory.
Why it matters
A leader who cannot name the cue behind a rule cannot tell whether the rule still works or is merely still running.
Adding information to a decision in a noisy environment can lower accuracy, which makes gathering more data an expensive default.
Use this tomorrow
1Write out the single cue that triggers your most-used escalation rule, then count how many of your last ten escalations actually contained that cue.
2Pick one recurring decision, cut the inputs you consult down to three, and log for two weeks whether any outcome came out differently.
Go deeper
Gigerenzer frames this as ecological rationality: a heuristic is not smart or dumb on its own, only well or badly matched to the structure of the environment it runs in. The same rule that is excellent for catching a fly ball is useless for pricing a bond, because the underlying regularities are nothing alike. That reframes the question a leader should ask about any rule of thumb, from is this rigorous enough to what does this rule assume about my world and is that still true. It also explains why imported best practices so often disappoint, since the rule arrives without the environment that made it work.
The opposite failure is real and worth naming. In stable environments with abundant, clean data, a model that weighs many cues can extract signal that a one-cue rule leaves on the table, and insisting on frugality there is its own kind of laziness. The practical split is where your uncertainty comes from: a world that keeps moving rewards the frugal rule, because yesterday's weights are already stale, while a world that holds still rewards the model that has had time to learn it. Most leaders misclassify their own environment in whichever direction flatters the tool they already own.
