Philip E. Tetlock

@PTetlock

Annenberg University Professor, Wharton & School of Arts and Sciences (Psychology & Political Science)

Philadelphia, PA
amzn.to/1FWLGAD
Se unió en octubre de 2011
280 fotos y videos Fotos y vídeos

Tweets

Bloqueaste a @PTetlock

¿Estás seguro de que quieres ver estos Tweets? Ver los Tweets no desbloqueará a @PTetlock

  1. 19 mar.

    I love the company. And “dark arts” is ironic. All three books promote transparency in the standards of evidence & proof we use in judging claims. Transparency is best check on the duplicity, rigidity & demagoguery that permeate a depressingly large fraction of the twitterverse

    Delphic oracle, crystal ball, or predictive analytics? Futurologists use data (Bayesian statistics) and psychology to perform the dark arts. I've been re-reading ; and poker player . All are recommended. anyone?
    1 respuesta 4 retweets 31 Me gusta
    Deshacer
  2. 16 mar.

    so many ways to be superb

    0:59
    ! low-skilled work?
    11 respuestas 137 retweets 437 Me gusta
    Deshacer
  3. 15 mar.

    A rare sighting in academia: a self-critical article. The great Keith Stanovich reflects on his blind spots. It is hard to be open-minded, even if open-mindedness happens to be your research specialty.

    Reading researchers as Superforcasters?- ‘The need for intellectual diversity in psychological science: Our own studies of actively open-minded thinking as a case study’ - ScienceDirect. ⁦⁩
    2 respuestas 9 retweets 33 Me gusta
    Deshacer
  4. 15 mar.

    Hardball politics imputes worst semi-plausible motives to opponents (covert/unconscious racism, sexism, anti-Semitism,…)—& dumbs down discourse. But sometimes it is so tempting…

    3 respuestas 2 retweets 26 Me gusta
    Deshacer
  5. 14 mar.

    Bryan saw the deeper truth before scandal hit the headlines: higher education is a well-oiled form of organized hypocrisy. And of course, mea culpa, the research professoriate is a big beneficiary. Much for the left & right to complain about but the center is sitting pretty happy

    4 respuestas 27 retweets 95 Me gusta
    Deshacer
  6. 13 mar.

    Below are some reasons to be wary of conventional wisdom of elites. And, of course, mocking elites is lots of fun. But cherry-picking is cherry-picking, whether elites do it to self-promote or skeptics do it to debunk elites. We need fair score-keeping across the board

    4 respuestas 14 retweets 67 Me gusta
    Deshacer
  7. 13 mar.

    Undergraduates’ non-negotiable demands seemed less transparently absurd 45 years ago—I vaguely recall

    If you want to dispute the narrative that students are overprotected at elite liberal arts colleges such as Sarah Lawrence, you probably should not put a demand for unlimited free fabric softener in your 3rd demand (out of more than 90) to the president
    Mostrar este hilo
    2 respuestas 1 retweet 21 Me gusta
    Deshacer
  8. 12 mar.

    Spyros is right: these long-range forecasts are pretty awful—& the forecasters, chronically over-confident. But why not try crawling before running—& learn how well we can do in shorter-range exercises and discover where we are going wrong faster

    2 respuestas 11 retweets 45 Me gusta
    Deshacer
  9. 11 mar.

    Editor’s conduct would have been par for the course in academia of 1919. By 1979, when I got my phd; it would’ve been 1 standard deviation out on rudeness/sexism scales. In 2019, 2 standard deviations out. Or am I failing again as intuitive statistician?

    When the editor wrote back, he removed all the women from the email and addressed only Matt & Paul. For me, that was the last straw. There is no justifiable reason to reply only to the men, especially when the 4 of us had signed the email. 9/
    Mostrar este hilo
    1 retweet 5 Me gusta
    Deshacer
  10. 8 mar.

    Best forecasters have fine-grained maybe zones. They more reliably distinguish 60/40 from 40/60 bets, 45/55 from 55/45. They're neither algorithms (probability as infinitely divisible) nor typical humans (yes maybe no) The secret: don’t hide your mistakes behind vague verbiage

    1 respuesta 22 retweets 64 Me gusta
    Deshacer
  11. 7 mar.

    Money confers so much status that tweet below will strike some as sacrilegious But how much epistemic deference do billionaires deserve? Not zero but not nearly as high as many assume. Umpteen standard deviations in wealth doesn't translate into umpteen sds of wisdom

    10 respuestas 14 retweets 99 Me gusta
    Deshacer
  12. 7 mar.

    Uri is incisiveness incarnate. Here he does to hard-nosed publishers of scientific journals what he did to the soft side of psychology:

    The problem with Elsevier 1. Journals' value = reputation 2. Publishers didn't create it 3. But do own it 4. Sell it too expensively 5. Trans.+coordination costs prevent true value creators reclaiming it Seems Rationally Speaking missed it. cc:
    3 retweets 11 Me gusta
    Deshacer
  13. 7 mar.

    Jay nails it. Whether it’s the success of marriages or mergers or the outcomes of civil wars, top forecasters start with base rates/outside view & then adjust in response to the news. Anchor on something sensible. Ignoring this advice guarantees losing to the algorithms

    So, in 2013, a simple base-rate forecast would have predicted that the war would end around now with a government military victory. Even in all that horror and complexity, some statistical regularity. /end
    Mostrar este hilo
    3 respuestas 30 retweets 130 Me gusta
    Deshacer
  14. 6 mar.

    Beneath a cute joke lurks a serious question that will pop up more as we see more algorithm-human-hybrid forecasting. How many degrees of uncertainty/probability can Nate-plus-algos distinguish across domains of questions? Superforecasters could do 10-to-15 in world politics

    Or maybe he's overrated. Only I, , properly rate Beto, whose chances of winning the nomination are exactly 13.81345143713%.
    Mostrar este hilo
    3 respuestas 2 retweets 19 Me gusta
    Deshacer
  15. 5 mar.

    I very much admired Lucian—& am happy to report that EVERY finding from our forecasting tournaments/prediction markets meets his challenge. It’s ALL a matter of degree: how much better can people get at distinguishing 70/30 bets from 30/70; 60/40 from 40/60; 55/45 from 45/55…

    When I was in grad school at MIT, Lucian Pye used to challenge the quants in the Poly Sci Dept to name a quantitative social science finding that depended on more than the sign of the relationship. That was early 1970s. It's easier to respond now, but not a lot.
    5 retweets 23 Me gusta
    Deshacer
  16. 5 mar.

    Wolfgang Pauli’s devastating “it’s not even wrong” comes to mind. But politics isn’t physics. We need a space for broad-brush, big-picture arguments. And we need more accountability for accuracy. The world gives us yet another dissonance-tolerance test

    7 retweets 27 Me gusta
    Deshacer
  17. 4 mar.

    There is a psychometric caveat: it is tricky distinguishing “open-mindedness” from “empty-headedness” (not impossible but tricky)

    2 respuestas 7 retweets 35 Me gusta
    Deshacer
  18. 2 mar.

    Taken as friendly amendment, Bob Are we playing a: 1.Pure-power game (activists trying to change world)—emotion essential 2.Pure-accuracy game (superforecasters trying to understand world)—emotion as biasing factor 3.Murky mix of the two (90% of us 90% of time)?

    2 respuestas 4 retweets 20 Me gusta
    Deshacer
  19. 1 mar.

    Bryan asks great questions (why I love talking with him). Anger: 1Encourages ideologically incestuous social networks 2Fuels over-justification (tamp down probability of anything good linked to policies you hate) 3Rigidifies thought (harder to change your mind) 4All of above

    En respuesta a @PTetlock
    What *does* anger do to forecasting accuracy?
    6 respuestas 26 retweets 86 Me gusta
    Deshacer
  20. 1 mar.

    Hardly news that angry people don’t like Nussbaum's argument that anger is impairing their judgment. Angry people are even less eager to discover what anger does to their forecasting accuracy. Righteous wrath ain't cheap.

    9 respuestas 48 retweets 164 Me gusta
    Deshacer

@PTetlock no ha twitteado todavía.

Parece que el contenido está tardando un poco en cargarse.

Puede que Twitter esté saturado o experimentando un problema momentáneo. Inténtalo de nuevo o visita el Estado de Twitter para más información.

    También te puede gustar

    ·