Lots of good stuff in this article, but this quote in particular is worth reading:
Polls of congressional districts don’t draw the usual attention because so few are made public. But Dave Wasserman of the Cook Political Report used that data in 2016 to say the election was very close and Trump could win. This time he notes:
the scores of district and state-level polls conducted by the parties to make spending decisions in down-ballot races generally align with national polls showing Trump running behind his 2016 pace, including in key states. In 2016, these same polls had shown flashing red warning signs for Hillary Clinton, particularly in districts with lots of white working-class voters.
A few districts are instructive. In 2018, Sharice Davids knocked off four-term Congressman Kevin Yoder in a district heavily based in Suburban Kansas City. She won by nearly 10 points in 2018, portending the GOP’s collapse in Suburban America (she also won formerly reliably Republican Johnson County by 15,000 votes against a strong candidate in Yoder). Usually a party that lost a seat they held for eight years would ostensibly claim they could win that seat back. But in 2020, every major ratings agency has the race rated as safe/solid Democrat. An internal poll had Davids up 20 points.
The district polls were a phenomenal sign Trump had real strengths in 2016. That’s not happening in 2020. The GOP would release internal polling if they showed Trump doing as well as he did in 2016. Instead, Biden is running well ahead of Trump across the board, and that’s why Wasserman has been so bullish on Biden.