This is the website of Abulsme Noibatno Itramne (also known as Sam Minter). Posts here are rare these days. For current stuff, follow me on Mastodon

Categories

Calendar

Electoral College: Trump close in Virginia again too

Since the last notable change, Jill Stein became the presumptive nominee of the Green party. So I folded in 4-way poll results that included her in AZ/NC/OH/NJ/GA/PA/FL/CT. I also added new polls in VA/WI/CA. Of all of these updates, changes only resulted from the additions in Virginia and Pennsylvania.

Interestingly, although Trump has been plummeting in the national polls the last few weeks (see Pollster and RCP), 4 out of the 5 state level category changes and 4 out of 4 tipping point changes here at Election Graphs since Trump became the presumptive Republican nominee have been in Trump’s favor.

I have been attributing this to the natural lag in the kind of state poll averages tracked here… but the trend has gone on long enough that there may be something else happening. It is actually starting to look like Clinton’s lead has been narrowing in a number of states, even while Trump has been falling further behind at a national level. This is an odd pattern. It will be interesting to see if it continues.

In the mean time, lets look at the latest example:

Virginia

chart-172

In Virginia on the eve of Trump becoming the clear winner on the Republican side he was behind by 12% in the Election Graphs average. With every poll since then, his position has improved. He’s now behind by only 3.2%. The most recent results were from June 13-15… well after the narrative of Trump’s collapse in the national polls was taking hold, and after weeks of Trump doing and saying things that seemed to be damaging his prospects nationally.

But yet the newest results have Trump down only 3% in a state where he had a poll showing him down 17% back in January. This trend does not look like a collapse. Quite the contrary. It does look like Trump is in a downward spiral in the national polls. But in Virginia specifically, he is doing better than he has since last August.

It seems strange that Virginia would be rapidly moving in the opposite direction than the national polls. Perhaps if we had weekly polls in Virginia we might have seen a different picture. Perhaps there have been large swings up and down that are just invisible because there have not been enough polls to show it. Maybe. But with the data we have, it really looks like Virginia has been on a Trumpward swing, despite the national trends.

With Virginia now close, we now include Trump winning it in “Trump’s best case”. So a quick look at the national trend:

chart-173

Aside from the one bump showing Clinton making Utah close, the trend here at the state level looks like Trump making a number of states closer. First Florida, then North Carolina, then Pennsylvania, now Virginia.

Trump isn’t flipping states, but he is making more states close. The national picture may look different, but at the state level, which is what actually determines the winner in November, there isn’t a Trump collapse visible yet. Trump is actually looking stronger.

Pennsylvania

chart-174

With Jill Stein now the presumptive nominee of the Green party, I added in a number of polls I had been tracking but not yet including in the analysis that included Jill Stein along with the major party candidates and Gary Johnson for the Libertarians. Among those was a poll in Pennsylvania showing a 1% Clinton lead over Trump. This did not change the status of Pennsylvania, which had already moved to Weak Clinton in the last update, but since Pennsylvania was the tipping point state, it moved that metric:

chart-175

The tipping point state remains Pennsylvania, but the margin moves from a 3.8% Clinton lead to a 3.4% Clinton lead.

So with today’s updates, both Trump’s “best case” and his tipping point improve.

The downward trend in the national numbers may eventually show itself at the state level, but it hasn’t yet.

144.9 days until the polls start closing on Election Day 2016.

Note: This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks both a poll based estimate of the Electoral College and a numbers based look at the Delegate Races. All of the charts and graphs seen in this post are from that site. Additional graphs, charts and raw data can be found there. Follow @ElectionGraphs on Twitter or like Election Graphs on Facebook to see announcements of updates or to join the conversation. For those interested in individual general election poll updates, follow @ElecCollPolls on Twitter for all the polls as they are added. If you find the information in these posts interesting or useful, please consider visiting the tip jar.

8 comments to Electoral College: Trump close in Virginia again too

  • Alex Coolchien

    Could it be that the “noise” is mainly pollster driven. I checked the older polls, and it seems that the large Clinton lead was mainly due to Roanoke College (which has a C rating according to fivethirtyeight). Incidentally, RC also had a poll that showed a tie back in mid-May. So I suspect that Virgina was never as large as 12%. We’ll probably see a clearer picture of statepolls once the pace picks up.

  • Alex, you are absolutely right that the Roanoke polls show bigger Clinton leads than any of the others (aside from the one showing a tie, which is at the other extreme). Roanoke is responsible for pushing Virginia into the “Solid Clinton” category representing a lead of more than 10%. If you remove all of the Roanoke polls, instead of the average peaking at a 12.2% Clinton lead, it would have only peaked at a 6.0% Clinton lead, which would only be in my “Strong Clinton” category. With or without Roanoke, there has been a Clinton weakening though. Including Roanoke, Clinton’s lead has dropped from 12.2% to 3.2%. Excluding Roanoke, Clinton’s lead would have dropped from 6.0% to 4.2%. Not as dramatic a change, but still a movement from a category that I consider out of reach (for the moment) for the other side, to a close enough margin it is prudent to allow for the possibility of the state going either way in the models.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.