Why the Clinton Strategy worked in Connecticut but not in the Electoral College.

I am not going to hide my shock at the result. Nor will I attempt to claim I saw it coming when I didn’t. However when it comes to Connecticut, given the national surprise, I feel I fared as well as could be expected. At a basic level, Clinton was in no danger in Connecticut. She won here by more than Kerry in 2004. Where the election day pattern held up better than I could have imagined was in tracking the primary map.

Clinton was able to win 47 out of the 53 towns she won in the primary, including Republican strongholds of Greenwich, New Canaan, Darien, Wilton, Easton, Ridgefield and Newtown. The Greenwich margin was 18%, and she ran stronger than Obama in almost all wealthy towns, but almost exclusively there. Some of the improvements were mind-boggling. Greenwich, Darien, New Canaan and Wilton averaged shifts of more than 30 points.

If this had been the end of the story a Clinton walk would have been in hand, but as big as some of her gains were, her losses were larger. She was pretty well slaughtered across all of rural Connecticut but also had big losses in inner-ring suburbs and small cities.

Romney got 55% or more in only 23 towns. Trump got 55% or more in 41 towns. Romney won about 60; Trump won 88. Trump won some very Democratic places headlined by Enfield, Bristol, Plainville, East Haven, Killingly. 

To make matters worse for Democrats, they also saw small erosions in their urban core, driven primarily by decline in voter turnout. Hartford was down 4%, New Haven was down 5%, Bridgeport was down 8%. Yet the cities still produced over 80% of the landslide here and a ton of votes.

The smaller cities and suburbs were not as forgiving. Margin for Clinton in New Britain was down 11%, Windham dropped 16% , Middletown 11%, Meriden 18%, West Haven 18%, Waterbury 28%, and Norwich was down 18%. These are all warning signs. A Democrat simply should not do better in Greenwich than Norwich.

While this math enabled Clinton to have a comfortable win across the state, this map when played out across the swing states spelled doom. Connecticut is simply more educated and more urban than the country, and that is why we voted how we voted. The Clinton strategy to pick off elite Republicans clearly worked here, but the losses with non-elites everywhere else was fatal.

When we look down ballot, we see improvements for Trump in areas with the targeted races. Democratic bleeding in Meriden and Ansonia cost them two Senate seats in districts where Obama previously hit 60%. Clinton performed nowhere near as strongly. Democrats did not gain any compensating seats in Fairfield County as Democratic challengers were somewhat weak or non-existent. On the House side, Republicans gained by the same regional math with a Trump factor or the national ticket mattering in almost every race, except for Pat Boyd in the 50^th District, who flipped a Republican seat blue despite the Trump wave. Democrats salvaged gains in three of the seats where the Republican wave was strong in 2010, and overall are far weaker.

The Connecticut fault lines matched the national ones in important ways. One interesting thing for Connecticut Democrats is whether Presidential strength will be mirrored down ballot in the gubernatorial election year, or whether shifting interests will bring other forces to the fore. At this point a Republican sweep in 2018 would seem to be the most likely outcome, though predicting has clearly proven harder than we imagined.