Voter registration data post November has not been kind to Democrats. Forget the four special election losses. The longer term trend—despite the futility and weakness of the Republican Congress—has the Democrats trending toward obsolescence in the battleground states.
Since November, many states have been cleaning up voter rolls (a common occurrence) so totals for both parties have fallen in several states. This is nothing unusual. But the direction of the change and the size of it are instructive. Oh, and it’s worth noting that a recent PRRI survey of white voters found that “party affiliation” was the single best predictor of a Trump vote. In other words, these aren’t just “Republican” voters, but rather statistics tell us that more than 93% of them would be Trump voters.
I’ve been providing running updates and this week we have new numbers for Florida, New Hampshire, and New Mexico. Trump lost the latter two while winning 306 votes in the electoral college. Clinton carried New Hampshire by 2,736 votes (with Gary Johnson siphoning off 30,694 additional votes). Since November, however, both parties lost registrations—but Democrats lost 7,000 more in this Republican state. Still think a Democrat can win New Hampshire? Especially with no Johnson on the ticket? Think again.
In New Mexico, the margin was greater (it’s a Democrat state): 65,000 votes, with, again, Johnson taking 74,500, or more than enough to win if they had voted Republican. Since November? Republicans have seen a net gain of 10,000. Not enough yet to win, but, absent a strong third party candidate, very close.
Florida, the only one of the three states Trump won in November, Trump won fairly decisively—by 112,911 votes. There was no Gary Johnson on the ballot in Florida to save Hillary. But today? Republicans have seen a net gain of 49,000 since November against Democrats. That means that as of July 2017, Trump would win the Sunshine State by over 160,000 votes. Moreover, the margin has grown every month, indicating a trend in Florida toward Republicans of about 8,000 net gain per month. That means that by 2020, Republicans would be an outright majority in Florida.
Previously I have posted voter registration statistics on AZ, CO, DE, IA, ME, NC, NH, NJ, OR, and CA. Only in Colorado, Oregon, and California have Democrats seen net gains since November. But Trump lost all three (only Colorado seen as a real “battleground”) and California is so “out of play” that it doesn’t factor into any real strategic analysis. Pennsylvania, on the other hand, has seen a steady shift over 104,000 to the GOP; IA some 2,000: and Arizona (where both parties are gaining voters) nearly 10,000 more than the Democrats. In other words, Republicans have seen net gains in all but two of the battleground states for which we have voter registration data since the election. In no state where Trump won has there been any GOP slippage at all. Quite the contrary, as of today he’d win 306 electoral votes again, as well as New Hampshire and most likely Minnesota. (That state does not register by party, but the narrow victory by Clinton was won despite Trump carrying all non-Minny/St.Paul precincts by 2:1 on election day. And, again, Johnson played a role. While Democrats can take solace that they are “holding” blue Oregon and recently-blue Colorado, and that California is as locked in as a state can possibly get (with a net gain for Democrats since November of one million!), they are losing the rest of the country. Even blue states such as Maryland, New Jersey, and Delaware are seeing the Democrats’ edge shaved. And in places such as Rhode Island, Hillary saw a phenomenal drop-off from Obama’s 2012 totals.
As I’ve repeatedly said, based on the performance of Congress, there is only one interpretation to the meaning of these registrations: these are Trump voters, pure and simple. The only chance Democrats will have in 2020 in some of these states is to find a compelling third party candidate who can erode that “party identification” factor. Otherwise, in large swaths of the country the Democrats are trending to extinction.