in 
, via 
| she/her
The primary model is back on after Iowa. It's not good news for Biden. https://projects.fivethirtyeight.com/2020-primary-forecast/ …pic.twitter.com/8oWspWH2zj
Show this thread
And @NateSilver538 with more detail:
"The model shows former VP Joe Biden’s chances of winning a majority of pledged delegates being halved — from 43 % before Iowa to 21 % now.
Who gains from Biden’s decline? Well, a little bit of everyone."https://fivethirtyeight.com/features/our-post-iowa-primary-forecast-is-up-and-bidens-chances-are-down/ …
-
-
2 things: 1. The model has since updated again: https://projects.fivethirtyeight.com/2020-primary-forecast/ … 2. Iowa was weird! The assumptions about the post-Iowa bounce are particularly uncertain. But once new polls come out, the model will replace those assumptions with actual polling data to measure that bounce
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Shows the same as 2016.. WE DONT KNOW ANYTHING
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
rt others: the model is either too sensitive to bounces or not reflecting how bounces affect different populations differently— It's showing huge Buttigieg boost and Biden crash in SC, but there's little reason to think Buttigieg's Iowa bounce will inform SC forecast (unlike NH)
-
The new polls seem to confirm bounce hypothesis for NH. But SC voters, especially nonwhite, are not likely to switch from Biden to Buttigieg after IA/NH bounce. Adjust bounce effects to be tied more closely to fundamentals, then rerelease before this misinforms media narratives
End of conversation
New conversation -
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.