ClaimsPeople

Nassim Taleb

Nassim Nicholas Taleb is an author and risk researcher, known for the five-volume Incerto series, which includes Fooled by Randomness.

783 claims collected, 182 researched, 94 resolved. Report of 5 November 2026, method 0.1.0+28da144b.

Scores by kind of claim

Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.

  • Factual claims

    Statements about the present or the past, checked against the record: did they hold up?

    73 67 79

    High confidence, 94 factual claims resolved. From 18 documents, 2008 to 2025; 19 documents sampled from 33 candidates found.

    Not scored: 65 unresolvable, 9 checkable only against their own records.

  • Predictions

    Statements about events outside the person's control, checked once they fell due: did they come true?

    Insufficient record: 0 predictions resolved, 5 needed for a score.

    Not scored: 11 unresolvable. 2 more fall due later.

    Does not yet allow for how likely each prediction was when made.

  • Promises

    Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?

    Insufficient record: 0 promises resolved, 5 needed for a score.

    Not scored: 1 checkable only against their own records.

All types combined: 73 (likely 67 to 79, high confidence), from 94 claims: 94 factual claims. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Nassim Taleb's claims

577 claims · page 4 of 6

577 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Not checkable Not checkable Jul 2020 Moreover, it is not correct to assume, more or less implicitly, that a disease brings no or little costs, while mitigation is burdensome. Not checkable · Jul 2020
Not checkable Not checkable Jul 2020 And lockdowns are the costs of ignoring arguments such as increased connectivity in our environment Not checkable · Jul 2020
Not checkable Not checkable 30 Dec 2019 This regression's sole aim is to make IQ Eugenists and Quilettists look good. Not checkable · 30 Dec 2019
Not checkable Not checkable 30 Dec 2019 In addition to SPURIOUS correlation from data insufficiency under fat tails, correlation requires strick linear relationship Y=a+b X. Not checkable · 30 Dec 2019
Not checkable Not checkable 30 Dec 2019 you don't feel dishonorable promoting a plagiarized book, after being made aware of it? Not checkable · 30 Dec 2019
Not checkable Not checkable 30 Dec 2019 Otherwise uninformative. Not checkable · 30 Dec 2019
Not checkable Not checkable 29 Dec 2019 These are largely newspaper articles under bindings. Not checkable · 29 Dec 2019
Not checkable Not checkable 28 Dec 2019 Likewise, in any field, the higher the number of academic publications the higher the ratio of garbage. Not checkable · 28 Dec 2019
Not checkable Not checkable 28 Dec 2019 Recall that the more information you get, the more disproportionately higher the ratio of noise to signal. Not checkable · 28 Dec 2019
Not checkable Not checkable 28 Dec 2019 The optimal strategy is the #barbell: Not checkable · 28 Dec 2019
Not checkable Not checkable 28 Dec 2019 Bravo w mamnunin! Not checkable · 28 Dec 2019
Not checkable Not checkable 28 Dec 2019 I hated the compromise. Not checkable · 28 Dec 2019
Not checkable Not checkable 27 Dec 2019 I was told it was uncurable. Not checkable · 27 Dec 2019
Not checkable Not checkable 27 Dec 2019 Among the best skills for real life: the ability to be bored early. Not checkable · 27 Dec 2019
Not checkable Not checkable 27 Dec 2019 I have severe Bullshitphobia. Not checkable · 27 Dec 2019
Not checkable Not checkable 26 Dec 2019 In other words SKIN IN THE GAME. Not checkable · 26 Dec 2019
Not checkable Not checkable 26 Dec 2019 If anyone tries to bully you, there is one and one way to react: ignore or overreact. Not checkable · 26 Dec 2019
Not checkable Not checkable 26 Dec 2019 The right balance betw noise & signal REQUIRES producing forecasts as if they were prices, buying/selling at each. Not checkable · 26 Dec 2019
Not checkable Not checkable 24 Dec 2019 The same with infinite expectations under Fat Tails! Not checkable · 24 Dec 2019
Not checkable Not checkable 23 Dec 2019 I wonder why these IQ idiots didn't realize it before. Not checkable · 23 Dec 2019
Not checkable Not checkable 23 Dec 2019 Psychology =astrology Not checkable · 23 Dec 2019
Not checkable Not checkable 23 Dec 2019 Dead people are very, very good at keeping secrets. Not checkable · 23 Dec 2019
Not checkable Not checkable 23 Dec 2019 Which shows why the overly used "heritability index" is nonsense. Not checkable · 23 Dec 2019
Not checkable Not checkable 2019 Way too often, the decision science and economics literature uses one as a proxy for another. Not checkable · 2019
Not checkable Not checkable 2019 The misinterpretation is still made four decades after Kahneman-Tversky (1979). Not checkable · 2019
Not checkable Not checkable 2019 This is typically the case with venture capital: the riskier the project, the less likely it is to succeed but the more rewarding in case… Not checkable · 2019
Not checkable Not checkable 2019 likewise I2, because its construction is similar to VaR, violates the inequality. Not checkable · 2019
Not checkable Not checkable 2019 The well known Kahneman-Tversky result proved robust, but interpretations make erroneous claims from it. Not checkable · 2019
Not checkable Not checkable 2019 A binary forecasting record is likely to be a reverse indicator under some classes of distributions or deeper uncertainty. Not checkable · 2019
Not checkable Not checkable 2019 More significantly, the tracking error for probability guesses will not map to that of the performance. Not checkable · 2019
Not checkable Not checkable 2019 Furthermore, one can overestimate the probability yet underestimate the expected payoff. Not checkable · 2019
Not checkable Not checkable 2019 The methods, for the reasons we have shown here, are highly flawed except in narrow circumstances of purely binary payoffs Not checkable · 2019
Not checkable Not checkable 2019 if x moves by up to 1 unit (say, standard deviation), there is a profit, and losses beyond that. Not checkable · 2019
Not checkable Not checkable 2019 So in both cases a forecast of < 1 is correct, but the lumping of the volatility –the fatness of tails– made a huge difference. Not checkable · 2019
Not checkable Not checkable 2019 The Conditional Value at Risk, CVar, measures the expected total losses conditional on default, which is altogether another entity… Not checkable · 2019
Not checkable Not checkable 2019 which can allow in absurd cases p∗ to exceed 1 when the distribution is grossly misspecified. Not checkable · 2019
Not checkable Not checkable 2019 The difference is exaggerated when the variable under consideration lies in the power law classes of probability distributions. Not checkable · 2019
Not checkable Not checkable 2019 An economic "collapse" or a climate "disaster" must not be expressed as an event in {0, 1} when in the real world it can take many values. Not checkable · 2019
Not checkable Not checkable 2019 as games are necessarily truncating a dimension from reality Not checkable · 2019
Not checkable Not checkable 2019 Probability is just a kernel inside an integral or a summation, not a real thing on its own. Not checkable · 2019
Not checkable Not checkable 2019 In fact under fat tails, there is no "typical" collapse or disaster, owing to the absence of characteristic scale, hence verbal binary… Not checkable · 2019
Not checkable Not checkable 2019 Succession 1 (thin tails): {1, 1, 1, 1, 1, 0, 0}. Mean variation= 0.71. P/L: g(x) = 2. Prediction is correct. Not checkable · 2019
Not checkable Not checkable 2019 We can finally assess the speed of convergence, the rate at which higher moments map to those of a Gaussian distribution Not checkable · 2019
Not checkable Not checkable 2019 Decision making is sequential. Not checkable · 2019
Not checkable Not checkable 2019 Note that we are operating under the mild case of known probability distributions, as it gets worse under parametric uncertainty. Not checkable · 2019
Not checkable Not checkable 2019 almost always have their valuations react in opposite ways when one increases the kurtosis of Not checkable · 2019
Not checkable Not checkable 2019 arguments for the consolidation of beliefs via prices does not lead to prediction markets as discussed in such pieces as [25] Not checkable · 2019
Not checkable Not checkable 2019 One suggestion by the author is for the M5 competition to correct for that by making "predictors" predict the minimum (or maximum) in a… Not checkable · 2019
Not checkable Not checkable 2019 it is a central property of fat tails that higher moments are explosive (even "infinite") and count more and more. Not checkable · 2019
Not checkable Not checkable 2019 We hold that calibration in frequency (probability) space is an academic exercise (in the bad sense of the word) that mistracks real life… Not checkable · 2019
Not checkable Not checkable 2019 This is rather a problem with the experiment: people do not necessarily separate probabilities from payoffs. Not checkable · 2019
Not checkable Not checkable 2019 meaning the fatter the tails, the worse the uncertainty about the expected payoff and the larger the difference between I1 and I2. Not checkable · 2019
Not checkable Not checkable 2019 This example is particularly potent as it has no verbalistic correspondence but can be understood by option traders and machine learning. Not checkable · 2019
Not checkable Not checkable 2019 unwarranted conclusions that people overestimate tail risk have been chronically made since Not checkable · 2019
Not checkable Not checkable 2019 Succession 2 (fat tails): {0, 0, 0, 0, 0, 0, 5}. Mean variation= 0.71 (same). P/L: g(x) = −18 (bust, really bust, but prediction is… Not checkable · 2019
Not checkable Not checkable 2019 Simply, their g(.) function was nonlinear not properly adapted to naive forecasting of the first moment of the random variable. Not checkable · 2019
Not checkable Not checkable 2019 the probability distribution of probability (sic) is robust –in other words the distribution of the probability of X doesn’t depend on the… Not checkable · 2019
Not checkable Not checkable 2019 making "calibration" mistakes where it doesn’t matter or can be helpful should be encouraged, not penalized. Not checkable · 2019
Not checkable Not checkable 2019 for under Gaussian distributions the effect can be marginal, which may have lulled the psychology literature into the conflation. Not checkable · 2019
Not checkable Not checkable 2019 It is practically impossible to hedge one with another Not checkable · 2019
Not checkable Not checkable 2019 Some nonlinear payoff functions, while not lending themselves to verbalistic expressions and "forecasts", can be well captured by ML or… Not checkable · 2019
Not checkable Not checkable 2019 We start with yj = (f − 1Aj), the difference between a continuous Beta distributed random variable and a discrete Bernoulli one, both… Not checkable · 2019
Not checkable Not checkable 2019 reducing these events to verbalistic "collapse", "disaster" is equivalent to a health insurance payout of a lump sum if one is "very ill" Not checkable · 2019
Not checkable Not checkable 2019 If a mistake doesn’t cost you anything –or helps you survive or improve your outcomes– it is clearly not a mistake. Not checkable · 2019
Not checkable Not checkable 2019 Owing to Jensen’s inequality, the discrepancy (I1 − I2) increases under parameter uncertainty, expressed in higher kurtosis Not checkable · 2019
Not checkable Not checkable 2019 Many perceived "biases" are shown to be just mischaracterizations by psychologists. Not checkable · 2019
Not checkable Not checkable 2019 prices are not binary bets. Not checkable · 2019
Not checkable Not checkable 2019 Binary forecasts are not particularly prone to model error; real world payoffs are. Not checkable · 2019
Not checkable Not checkable 2019 which works well in situations where one has to forecast the first moment of the distribution and the dispersion around it. Not checkable · 2019
Not checkable Not checkable 2019 There can be considerable mathematical and statistical differences between the following two items: Not checkable · 2019
Not checkable Not checkable 2019 P/L tautologically corresponds to the real world distribution, with an absorbing barrier at the survival condition. Not checkable · 2019
Not checkable Not checkable 2019 This insures the predictor meets the constraint of staying above water. Not checkable · 2019
Not checkable Not checkable 2019 We have shown that, in risk management (and while forecasting socio-economic and financial variables), one should never operate in… Not checkable · 2019
Not checkable Not checkable 2019 most businesses have severely skewed payoffs, so being calibrated in probability is meaningless. Not checkable · 2019
Not checkable Not checkable 2019 given that our exposures in the real world rarely map to I1. Not checkable · 2019
Not checkable Not checkable 2019 Accordingly, Fig. I-5 makes little sense empirically. Not checkable · 2019
Not checkable Not checkable 2019 These appear to be at best one order of magnitude fewer in decisions entailing formal predictions. Not checkable · 2019
Not checkable Not checkable 2019 At the core, calibration metrics such as the Brier score are always thin-tailed, when the variable under measurement is fat-tailed, which… Not checkable · 2019
Not checkable Not checkable 2019 As we can show here, prediction markets may be useful for gamblers, but they cannot hedge economic exposures. Not checkable · 2019
Not checkable Not checkable 2019 Risk management is about changing the payoff function g(.) rather than making "good forecasts". Not checkable · 2019
Not checkable Not checkable 2019 Further, we can extract its probability density function of the Brier score for N = 1 Not checkable · 2019
Not checkable Not checkable 2019 Binary forecasts and calibration metrics via the Brier score belong to the thin-tailed class. Not checkable · 2019
Not checkable Not checkable 2019 Clearly we are dealing with 1) f beta distributed Not checkable · 2019
Not checkable Not checkable 2019 probabilistic calibration requires estimations of the zeroth moment while the real world requires all moments Not checkable · 2019
Not checkable Not checkable 2019 We can proceed to prove directly from there the convergence in distribution for the average Not checkable · 2019
Not checkable Not checkable 2019 Inconsistent opinions therefore would lead to a violation of arbitrage rules Not checkable · 2019
Not checkable Not checkable 2019 can be close to I2 only under thin-tailed distributions, not under the fat tailed ones. Not checkable · 2019
Not checkable Not checkable 2019 This in a nutshell shows how, in the world of finance, "bad" forecasters can make great traders and decision makers, and vice versa. Not checkable · 2019
Not checkable Not checkable 2019 The substitution comes from conflating I1 and I2, which becomes an identity if and only if g(.) is constant above Not checkable · 2019
Not checkable Not checkable 2019 A series of binary forecasts will converge very rapidly to a thin-tailed Gaussian even if the underlying distribution is fat-tailed Not checkable · 2019
Not checkable Not checkable 2019 Clearly one may be mismeasuring performance if the random variable is in the wrong probability class. Not checkable · 2019
Not checkable Not checkable 2019 Given that company sales are typically fat tailed, a very low probability of success might still be satisfactory to make a decision. Not checkable · 2019
Not checkable Not checkable 2019 While the trader story is mathematically trivial (though the mistake is committed a bit too often) Not checkable · 2019
Not checkable Not checkable 2019 The following simplified example can show us their error and the mismatch between correct forecasting and performance as well as what… Not checkable · 2019
Not checkable Not checkable 2019 Accordingly, miscalibration may be a good idea if it reduces the odds of being absorbed. Not checkable · 2019
Not checkable Not checkable 2019 This quantity can be quite different from the actual loss expectation in case of default. Not checkable · 2019
Not checkable Not checkable 11 Apr 2018 I favor dogs over cockroaches. Not checkable · 11 Apr 2018
Not checkable Not checkable 11 Apr 2018 the notion of ruin is lack of extinction. It’s very different. Not checkable · 11 Apr 2018
Not checkable Not checkable 11 Apr 2018 If I can explain my book, then I shouldn’t be writing books. Not checkable · 11 Apr 2018
Not checkable Not checkable 11 Apr 2018 I realized there was a class of people who have the upside and transfer the downside to others. Not checkable · 11 Apr 2018
Back to the list

Nassim Taleb

Open its page

Not checkable

“At the core, calibration metrics such as the Brier score are always thin-tailed, when the variable under measurement is fat-tailed, which worsens the tractability.”

Nassim Taleb 2019 On the Statistical Differences between Binary Forecasts and Real World Payoffs (arXiv)

At the core, calibration metrics such as the Brier score are always thin-tailed, when the variable under measurement is fat-tailed, which…

Not checkable An opinion, value judgement or statement with nothing to check.

Cite this claim

Nassim Taleb, “At the core, calibration metrics such as the Brier score are always thin-tailed, when the variable under measurement is fat-tailed, which…” (said 2019). Verdict: not checkable. Reliability API, claim c_b8e2b342b16f, method 0.1.0+28da144b. https://epistacklabs.com/claims/c_b8e2b342b16f

Claims without a verdict

Is a verdict here wrong, or do you speak for Nassim Taleb? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.