November 2024
- Nov 30, 2024ConceptsWhat Is the Pareto Principle: The 80/20 Rule ExplainedPareto observed in 1896 that roughly 80% of Italian land was held by 20% of the population; Joseph Juran generalized it into the vital few and the trivial many for manufacturing quality in the 1940s. Fenton and Ohlsson, in IEEE Transactions on Software Engineering, found about 20% of code modules held roughly 80% of defects, and that those modules could be flagged in advance by complexity metrics. The article is equally clear on why the low-value 80% is rarely separable.
- Nov 30, 2024ConceptsIntermittent Reinforcement: Unpredictable Rewards ExplainedWhy unpredictable reward produces the most persistent behavior. Running low on food pellets in 1943, Skinner began delivering them intermittently rather than after every response, expecting behavior to weaken. The pigeons pecked faster and kept pecking long after pellets stopped entirely. The variable ratio schedule he stumbled into resists extinction better than any other arrangement, which is the architecture of slot machines and social feeds.
- Nov 29, 2024ConceptsChoice Overload: When More Options Produce Fewer DecisionsAt Draeger's Market in 1995, Iyengar and Lepper's 24-jam tasting table drew 60 percent of passing shoppers against 40 percent for the 6-jam table, yet 30 percent of those who stopped at six jams bought a jar versus 3 percent at twenty-four. The effect is real but conditional: Scheibehenne's 2010 meta-analysis of 50 studies found a mean effect size near zero, and Chernev's 2015 review of 99 studies pinned it to four moderators.
- Nov 28, 2024ConceptsThe Endowment Effect: Why We Overvalue What We OwnIn the autumn of 1990, Kahneman, Knetsch and Thaler handed Cornell students a bookstore coffee mug at random. Sellers would not part with theirs below a median of $7.12, while buyers offered a median of $2.87 for the identical object, and a third group choosing between mug and cash valued it at $3.12. Twenty minutes of possession moved the price by a factor of about 2.5 and falsified the Coase theorem's prediction that trade would clear.
- Nov 27, 2024ConceptsThe Overconfidence Effect: The Mistakes of the SmartAlpert and Raiffa asked people for ranges they were 90 percent certain contained the right answer. The truth fell outside those ranges roughly 40 to 50 percent of the time. The error is not boastfulness, it is compressed uncertainty, and Moore and Healy showed in 2008 that lumping it in with the better-than-average effect produced contradictions, since overplacement reverses on hard tasks while overprecision does not. Gigerenzer argues frequency formats shrink the gap.
- Nov 26, 2024ConceptsIn-Group Bias: The Psychology of Group FavoritismHenri Tajfel told Bristol schoolboys in 1971 they were Klee types or Kandinsky types, assigned effectively at random. They had never met their group, could not communicate, and stood to gain nothing personally. They still favored their own, and chose the maximum difference strategy: 7 points for us and 1 for them over 19 and 25, forgoing 12 points purely to widen the gap. Categorization alone, with no history or shared interest, produces favoritism.
- Nov 25, 2024ConceptsFraming Effect: How Presentation Influences DecisionsMcNeil, Pauker, Sox and Tversky gave 238 patients, 491 students and 424 physicians a choice between surgery and radiation for lung cancer, with identical data shown as survival or mortality figures. Surgical preference fell from 82 percent in the survival frame to 53 percent in the mortality frame, and the shift was statistically indistinguishable across groups. Trained physicians were as susceptible as medically naive patients. The words, not the numbers, did the work.
- Nov 23, 2024ConceptsWhat Is the Recency Bias: How Recent Events Distort JudgmentGoyal and Wahal examined over 3,400 pension funds across a decade and found they consistently hired investment managers after periods of strong recent performance and fired them after weak ones, with no improvement in subsequent returns from the hiring decisions. These are fiduciaries with professional teams and research budgets, making the same error as retail investors with far more money. Also covers the serial position effect and what actually corrects it.
- Nov 22, 2024ConceptsThe Optimism Bias: Why We Expect the Best OutcomesNeil Weinstein's 1980 study asked 258 undergraduates to compare their own odds of major life events against an average peer. They rated themselves below average on every negative event and above average on every positive one, knowing full well that not everyone can be below average. Tali Sharot's later work found the mechanism that keeps it going: beliefs update strongly toward good news about real risk and much more weakly toward bad news.
- Nov 21, 2024ConceptsGambler's Fallacy: Misconceptions About RandomnessTheir 2016 study found the same corrective pattern in loan officers approving applications and umpires calling pitches, controlling for the merits of each case. A roulette wheel has no memory, but the people watching it keep a ledger, and so do professionals trained for case-by-case objectivity. The page is careful on the hot hand too, noting Miller and Sanjurjo's 2018 correction and that the debate continues rather than treating it as settled illusion.
- Nov 20, 2024ConceptsThe Halo EffectNisbett and Wilson filmed a lecturer behaving warmly for one group and coldly for another, with identical lecture content. The warm-condition students rated his appearance, his accent and his mannerisms more favorably, then denied his warmth had affected those judgments. Their 1977 paper is why awareness is a weak defense: people invent independent reasons for a judgment a single impression produced. Thorndike saw the same pattern in 1920.
- Nov 19, 2024ConceptsAnchoring Bias: The First Number Shapes Your MindHow an arbitrary number bends judgment even when everyone knows it is meaningless. Kahneman and Tversky spun a wheel of fortune rigged to land on 10 or 65, then asked what percentage of African countries belong to the UN. Those who saw 10 averaged 25 percent; those who saw 65 averaged 45. Participants watched the wheel being rigged and still could not escape it, because adjustment from an anchor is typically insufficient.
- Nov 18, 2024ConceptsWhat Is a Framework: Principles for Structured ThinkingDemographic segmentation and asking customers what they wanted produced nothing useful. Reframing the question to what job the customer hires a milkshake to do revealed morning commuters buying something to last a long drive, competing against bananas and bagels rather than other milkshakes. The page separates frameworks from models, methodologies and processes, and is blunt that domain mismatch, like applying Five Forces to a platform business, is the common failure.
- Nov 18, 2024TechnologyUnderstanding Automation and Job Risk InsightsWhat the headline automation numbers actually measure. Frey and Osborne rated 702 occupations in 2013 and classified 47 percent as high risk, but the study explicitly described potential susceptibility over an unspecified number of years rather than forecasting unemployment. The OECD applied a stricter task-level definition and reached 9 percent. Autor's work finds polarization rather than mass joblessness.
- Nov 16, 2024ConceptsThe Fundamental Attribution ErrorIn 1967 Jones and Harris told participants that essay writers had been assigned their pro-Castro or anti-Castro position, which licenses no inference about the writer's actual views. Participants drew it anyway, and decades of replication failed to remove it. Gilbert, Pelham and Krull later showed why: the dispositional read is automatic and the situational correction is effortful, so participants running a memory task never corrected at all.
- Nov 14, 2024ConceptsConfirmation Bias: Why We Find What We're Looking ForIn 1979, Lord, Ross and Lepper gave supporters and opponents of capital punishment the identical pair of studies, one favoring deterrence and one undercutting it. Both groups rated the study that agreed with them as better designed, and both left holding their original view more strongly than before. Balanced evidence deepened the disagreement. The article traces the pattern from Wason's 2-4-6 task to Iraq, N-rays and Theranos.
- Nov 13, 2024ConceptsLoss Aversion: Why Losing $100 Hurts More Than Winning $150Kahneman and Tversky's 1979 prospect theory paper found that most people needed stakes of roughly $200 to $250 before accepting a coin flip risking $100. De Martino, Camerer and Adolphs, in PNAS in 2010, then studied two patients with focal bilateral amygdala damage: both tracked expected value and probability normally, yet both showed dramatically reduced loss aversion. Also covers Boston housing sellers, NFL fourth downs and the cultural variation in the coefficient.
- Nov 12, 2024ConceptsUnderstanding Parkinson's Law: Time Management InsightsParkinson published that arithmetic anonymously in The Economist in 1955, and the satire stuck because the data held. The useful modern test is Ariely and Wertenbroch's deadline experiment, where students who set their own staggered due dates did better than those with none, but students given externally imposed evenly spaced deadlines did best of all. People know they need constraint and still grant themselves too much room.
- Nov 11, 2024ConceptsRegression to the Mean: Statistical Insights ExplainedA senior flight instructor told Kahneman that praising a clean maneuver was followed by worse flying and criticizing a sloppy one by improvement, so reward clearly did not work. He was right about what he saw and wrong about the cause. Extreme runs contain luck that does not repeat, so the next run drifts toward the pilot's true average whatever anyone says. An entire training culture had been built on evidence the mathematics produced by itself.
- Nov 10, 2024ConceptsThe Availability Heuristic and Media: Why What Comes to Mind Feels Like What Is TrueSchwarz and colleagues showed in 1991 that the feeling of retrieval matters more than what gets retrieved: people asked to recall twelve examples of their own assertive behavior rated themselves less assertive than those asked for six, because straining to reach twelve reads as evidence that assertiveness is rare. The cost can be counted: Gigerenzer estimated roughly 350 extra American road deaths in the three months after September 11, as travelers moved from planes to cars.
- Nov 9, 2024ConceptsDunning-Kruger Effect: Misjudging CompetenceMcArthur Wheeler robbed two Pittsburgh banks in 1995 with lemon juice on his face, convinced it made him invisible to cameras. Dunning was struck by the sincerity, not the brazenness. In the 1999 paper, bottom-quartile performers estimated the 62nd percentile and averaged the 12th; top performers guessed 68th and scored 86th. The popular stupid-people-think-they-are-smart curve distorts this: it is an asymmetric effect at the tails, about competence structure rather than ego.
- Nov 8, 2024ConceptsFeedback Loops: How Systems Talk to ThemselvesJames Watt's 1788 centrifugal governor made steam engines stable rather than faster, the first widely deployed engineered feedback device. The same structure ran in reverse on October 19, 1987, when the Dow fell 508 points, or 22.6%: the Brady Commission concluded portfolio insurance had created a reinforcing loop, selling futures as prices dropped and driving them lower. Sterman's 1989 beer-game analysis of 48 trials found every single one produced the boom-bust pattern.
- Nov 7, 2024ConceptsThe Map Is Not the Territory: Why Models Always LieLong-Term Capital Management managed 126 billion dollars of assets on 4.8 billion of equity, with two Nobel laureates among its partners, because it trusted models built on stable correlations. When Russia defaulted in August 1998 those correlations collapsed toward one, LTCM lost 4.6 billion, and the New York Fed brokered a 3.6 billion private bailout. The partners had not miscalculated; they had stopped seeing the model as a map.
- Nov 6, 2024ConceptsEmotion Regulation: The Science of Managing What We FeelGross had participants watch a graphic surgical film while either reappraising its meaning or hiding their expressions. Both groups looked composed. Suppressors showed elevated cardiovascular activation throughout, while reappraisers showed none of that cost: changing what the film meant changed what it did to them. Troy later found the picture is conditional, since reappraisal helped for uncontrollable stressors but gave no advantage for controllable ones.
- Nov 5, 2024ConceptsInversion: Thinking Backward to Solve ForwardAt Scutari in 1854, Florence Nightingale stopped asking how to save more soldiers and asked what was killing them: ventilation, sewage in the water, overcrowding, unwashed linen. Mortality fell from roughly 42 percent to 2 percent in six months. The same reversal drives Snow's Broad Street map and Amazon's working backward memos. Mitchell, Russo and Pennington found in 1989 that treating an outcome as already past surfaced about 30 percent more reasons for it.
- Nov 4, 2024ConceptsChesterton's Fence: Importance of Understanding ChangeThe 1929 passage is explicit that a reformer who grasps the fence's purpose and clears it anyway is acting within the principle, and only the one who removes what he cannot explain is indicted. Mao's Four Pests Campaign is the hard case: sparrows were destroyed as grain eaters, but they also ate locusts, and the locusts arrived without a predator. The page covers action bias, the illusion of explanatory depth, and when removing a fence anyway is right.
- Nov 3, 2024TechnologyWhat Is Machine Learning: How It Actually WorksA model is a function with adjustable parameters: the loss function scores how wrong it is, and gradient descent nudges each parameter downhill, millions of times. The article covers supervised, unsupervised, and reinforcement learning, and does not flinch at the limits. Northcutt, Athalye, and Mueller found popular benchmark test sets carry label error rates averaging at least 3.3 percent, so some state-of-the-art results may reflect adaptation to noisy labels.
- Nov 3, 2024Work SkillsProven Salary Negotiation Strategies for ConfidenceA precise figure signals you have actually researched the market rather than picked a number, and counterparts treat it as better informed. The page pairs this with Bidwell's finding that external hires are paid more than people promoted internally into comparable jobs, which is the structural reason an outside offer carries leverage. It also scripts the counter itself and advises taking a day rather than accepting on the spot.
- Nov 2, 2024Work SkillsGuide to Effective Job Interview PreparationTheir 2013 paper calls the belief in such interviews an illusion, because interviewers form a fast impression and then weight later evidence to confirm it, adding noise rather than signal. That is why the opening minutes carry disproportionate weight and why structure helps. The page covers layered company research, STAR answers with quantified results, and practicing aloud rather than mentally, drawing on Roediger and Karpicke on retrieval.
- Nov 1, 2024TechnologyImproving Website SEO: Effective Strategies for 2026The page is strongest on why ranking works the way it does, tracing the path from TF-IDF, which could be gamed by simply repeating a keyword, to PageRank, which moved the signal to what other publishers do, to BERT in 2019, which reads words in relation to the whole sentence. That history explains why matching search intent now beats matching vocabulary, and the page is careful to label one widely shared ranking experiment as anecdotal.
No guides match that search. Try a shorter phrase, or clear the search.