@NateSilver538 ತಡೆಹಿಡಿಯಲಾಗಿದೆ

ನೀವು ಖಚಿತವಾಗಿಯೂ ಈ ಟ್ವೀಟ್‌ಗಳನ್ನು ನೋಡಲು ಬಯಸುವಿರಾ? ಟ್ವೀಟ್‌ಗಳನ್ನು ನೋಡುವುದು @NateSilver538 ಅವರನ್ನು ತಡೆತೆರವುಗೊಳಿಸುವುದಿಲ್ಲ.

  1. Reminder: Cubs will win the World Series and, in exchange, President Trump will be elected 8 days later.

  2. 8. It's usually not worth it to diagnose why an individual poll deviates from the consensus. Think 'macro' not micro—look for robust trends.

  3. What if last week was just a trick so he could finally shake Kasich loose?

  4. 7. Looking at Electoral College is great once you have rich data — multiple recent polls of each state. We won't have that for a few months.

  5. 6. The election will go through a lot of twists and turns, and polls are noisy. Don't sweat individual polls or short-term fluctuations.

  6. 5. Watch whether polls are likely or registered voters. Usually GOP gains a point or two with likelies. Possible Trump will be an exception.

  7. 4. Possible there are effects from Trump wrapping up his nomination while Clinton still competes against Sanders. We'll know more in June.

  8. 3. State polls are broadly consistent with that ~6% Clinton lead + noise + house effects. Not nearly enough data to say more than that.

  9. 2. The data is consistent with Clinton having a ~6% nat'l lead over Trump. It's early. Trump could win. Also, he could lose in a landslide.

  10. 1. For fuck's sake, America. You're going to make go on a rant about general election polls -- in May?

  11. He broke the record, guys. 17.

  12. Steph Curry has 15 points in OT. All time NBA record is 16, apparently.

  13. How anecdote journalism blew the Trump story 🔥🔥🔥 (27:00)

  14. If I were WaPo or Politico, I'd seriously consider hiring a reporter to ride the Acela back and forth all day.

  15. Hypothetically, who from the world of politics would you want to see interview?

  16. As an aside, I've soured on the term "data journalism". A better one for what 538, Upshot et. al. do is probably "empirical journalism".

  17. Data journalists -- who often disagree with one another, BTW -- replicated some of these errors but avoided others, especially re: Clinton.

  18. Apart from Trump, there were many other failures for traditional reporting this cycle.

  19. NYT's reporters literally could not think of a way that Trump could win the Republican race.

  20. Shorter : Polls underrating Sanders indicate fraud. Also, polls *overrating* Sanders indicate fraud.

  21. A lot of the animus the NYT politics desk had toward 538 is because our forecasts pissed off their sources, especially in Romneyworld.

  22. No, I didn't predict that the Republican Party would lose its fucking mind.

  23. If you premise a "take" on polls having been wrong in Indiana, without mentioning how well they did in other states, you're being dishonest.

  24. Overall, the FiveThirtyEight "polls-only" model has called 51 of 56 (91%) races correctly this year. "Polls-plus" has called 50 of 56 (89%).

  25. Both tend to treat lessons drawn from a small number of recent elections as immutable laws of politics. Polling is more robust to change.

  26. In some important ways, the "fundamentals" from political science models resemble "conventional wisdom". Maybe a reason to be wary of them.

  27. There are lessons from that, but they're almost totally orthogonal to the "data journalism" vs. "traditional journalism" beef.

  28. Roughly speaking (I'm generalizing a LOT) polls got Trump *right* while other types of empirical evidence (e.g. endorsements) got him wrong.

  29. So let's look at polling *and* let's get out into the field, but less time laundering opinions from the consultant class into a "narrative".

  30. BTW, you know what polling is? It's a method of **talking to voters** in a structured way that reduces bias.

  31. ಗೆ ಪ್ರತ್ಯುತ್ತರವಾಗಿ

    .: The NYT politics desk was especially hostile to 538 when it tried to do stories that blended data with traditional reporting.

  32. ...but in avoiding the previous mistake, you'll often make an equal and opposite mistake in the other direction, or a whole new mistake.

  33. Relatedly: It's easy to overcorrect after an "unexpected" event occurs. It's easy to say "we'll never make THAT mistake again..." (1/)

  34. Prior to Trump, most surprising nominees include Carter, McGovern, Obama, Dukakis, in roughly that order. Pretty meh in general election.

  35. Candidates who beat pundits' expectations in the primaries haven't done particularly well in the general election.

  36. Big spike in Google searches for Gary Johnson.

  37. This chart is interesting although what happened recently is less interesting than the early-mid part of the race.

ಲೋಡಿಂಗ್ ಸಮಯ ಸ್ವಲ್ಪ ತೆಗೆದುಕೊಳ್ಳುತ್ತಿರುವಂತೆನಿಸುತ್ತದೆ.

Twitter ಸಾಮರ್ಥ್ಯ ಅಧಿಕವಾಗಿರಬಹುದು ಅಥವಾ ಈ ಕ್ಷಣದ ತೊಂದರೆಯನ್ನು ಅನುಭವಿಸುತ್ತಿರಬಹುದು. ಮತ್ತೆ ಪ್ರಯತ್ನಿಸಿ ಅಥವಾ ಇನ್ನಷ್ಟು ಮಾಹಿತಿಗೆ Twitter ಸ್ಥಿತಿ ಗೆ ಭೇಟಿ ನೀಡಿ.

    ಇದನ್ನೂ ಸಹ ನೀವು ಇಷ್ಟಪಡಬಹುದು

    ·