The survey that you extensively quote @dhume, is at odds with your own piece. You say in the piece that the major states with IA's (NY, NJ, Texas, CA) have already been decided (which i agree with) and the real battle is for IA votes in some swing states. (1/n)
https://twitter.com/dhume/status/1316891237195091968
The survey that you quoted @dhume, still talks to> 400 people in those states. The total sample size is only 936. Given that in those 4 states, majority of IA's vote for democrats, isn't this heavily tilted towards dems and not focusing as much on IA's in the swing states? (2/n)
Establishment surveys forget one big thing, surveying IA's in CA, NY, NJ, TX doesn't give an accurate picture of what is happening in swing states. 60% of the IA's might still vote for Dems, but if 70% of them come from those 4 states how is it representative? @dhume (3/n)
Lets look at the ratings that the survey quotes, i have my doubts about the numbers. But lets assume this data to be accurate. The largest number of respondents in the survey @dhume quotes seem to in the 'dont know' category about how Trump is handling relations with India. (4/n) 
The problem with this survey, is the overwhelming assertions it makes about the data based on flawed assumptions. Although finally the numbers may be 65-35 towards biden. If IA's swing the votes in MI, FL - we might see some interesting results given the margins last time (5/n)
FWIW, i still think Biden has a lead over Trump. But my issue with this piece by @dhume is it basis its analysis primarily on this survey which seems to be not representative. I would expect Sadanand to raise those questions about it as well, and not take it at face value (6/6)
• • •
Missing some Tweet in this thread? You can try to
force a refresh
