: I think there's a whole swamp of cognitive & journalistic biases at play in why people thought Clinton was a sure thing...
-
-
Replying to @NateSilver538
: ...that people have trouble interpreting probabilistic odds is definitely somewhere high on the list, but it's a long list.
6 replies 9 retweets 36 likes -
Replying to @NateSilver538
It is no doubt a long list. People's priors caused them to resist. That said; this one is in your corner, where you can lead.
1 reply 0 retweets 5 likes -
Replying to @zeynep @NateSilver538
NYT had their "90% Clinton!" bar on every election story. Even if the reporting had been better... Holy priming batman.
2 replies 1 retweet 10 likes -
Replying to @zeynep @NateSilver538
On 538, topline dominated people's attention, I think, even with stories & fine-print that correctly conveyed uncertainty.
2 replies 1 retweet 11 likes -
Replying to @zeynep @NateSilver538
But the 538 top line did convey much of the needed uncertainty!
1 reply 0 retweets 0 likes -
Replying to @deaneckles @NateSilver538
What people don't understand is what an underlying *correlated* error of 1-2% means for topline probability.
2 replies 0 retweets 0 likes -
Replying to @zeynep @NateSilver538
If readers interpreted the top line probability correctly, that wouldn't be necessary.
1 reply 0 retweets 1 like -
I think those, like NYT, that had >90% are the ones needing more criticism & changes for future forecasts.
1 reply 0 retweets 1 like -
Replying to @deaneckles @NateSilver538
Yes, NYT was worse—especially if you consider it topped every story. But must try to make readers understand, no?
2 replies 0 retweets 0 likes
Odds or probabilities may be best indicators we have, but have real limitations in conveying volatility & error.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.