Tag Archives: Electoral Votes

Most Recent Polling Shows Tight Race

Uncategorized, , , , ,

“Trump’s chance of victory have doubled over the last two weeks,” notes FiveThirtyEight, and this is in accord with what I’ve been saying.

I suggested a few days ago that while Clinton would probably win, there is a nowhere near zero chance that she won’t. FiveThirtyEight came out with an analysis today very similar to mine, suggesting that Trump has abut a 3 in 10 chance of winning. Historically, races tighten near the end, I think FOR THIS REASON mainly, and that has been happening. The actual national difference between Clinton and Trump by Tuesday will probably be about 2.5 percent or so.

Now, before you jump in to tell me that the national number isn’t what counts, yada yada yada, let me note right away that I do know about the Electoral College and stuff.

Anyway, see this for my most recent Electoral College analysis, and I’ll have a new one out in a day or so, though I expect it to be similar.

Meanwhile, here are some notes on some of the more interesting and important races.

No cherry picking here. All of the really recent, high ranked (by FiveThirtyEight) polls in states of interest. All these polls were released over the last few days, though they may cover earlier days. The data are all taken from FiveThirtyEight, but using the original poll numbers, not FiveThirtyEight’s adjustment.

Note: Polls that weight on the basis of motivation seem to favor Trump; his voters say they are more likely to vote.

Arizona, when it isn’t busy shooting something, generally votes for the Republican. There was hope this would not happen this year, but the latest polls suggest otherwise
Arizona CNN/Opinion Trump +5
Arizona Emerson Trump +4
Arizona Google CS Clinton +5

Florida is a very important states, and there are signs of Clinton weakening there, but most indicators suggest a Clinton win. Also, the TargetSmart study (not shown here) indicates that 28% of Republicans who voted early are voting for Clinton.
Florida CNN/Opinion Clinton +2
Florida Google CS Trump +3
Florida Quinnipiac Clinton +1
Florida TargetSmart (Not rated by 538) Clinton +8

People mention Georgia now and then. We’ll be watching Georgia, because if Clinton wins there, the world has changed. But she won’t.
Georgia Emerson Trump +9
Georgia Google CS Trump +9

I’ve been predicting a Clinton win in Iowa, many polls indicate otherwise, the latest Google Consumer Survey suggests a Clinton win.
Iowa Google CS CLinton +7

Nevada. I hear people saying that Clinton has Nevada in the bag. She doesn’t. My model currently has her winning there, but clearly there is ambiguity.
Nevada CNN/Opinion Trump +6
Nevada Google CS Clinton +7

New Hampshire has not been declared a solid sate for anyone, yet many seem to insist it is solid for Clinton. It isn’t, but also, there isn’t much good polling there, so really, we don’t know.
New Hampshire Google CS Trump +1

North Carolina is totally uncertain for many reasons, including polling all over the map, an active voter suppression campaign by the Republican party, and because it is, well, North Carolina.
North Carolina Elon Clinton +1
North Carolina Google CS Trump +6
North Carolina Quinnipiac Clinton +3
North Carolina SurveyUSA Trump +7

Everyone I know who is from Ohio or lives in Ohio loves Ohio and hardly ever shuts up about it. Time to shut up about it! You’all are about to go for Trump, so you suck.
Ohio Google CS Trump +2
Ohio Quinnipiac Trump +5

Pennsylvania seems solidly Clinton, though if I recall, Pennsylvania has sometimes thrown a surprise. But not likely this year.
Pennsylvania CNN/Opinion Clinton +4
Pennsylvania Franklin & Marshall Clinton +11
Pennsylvania Google CS Trump +2
Pennsylvania Monmouth Clinton +4
Pennsylvania Quinnipac Clinton +5
Pennsylvania Susquehanna Clinton +2

We fully expect Clinton to take Virginia.
Virginia Emerson Clinton +4
Virginia Google CS Clinton +5
Virginia Hampton Trump +3
Virginia WaPo Clinton +6

How Will The Swing States Swing?

Uncategorized, , , , , , ,

The election is one week off. I think I’ve convincingly demonstrated, here, that Clinton is likely but not certain to win, that Trump has something of a chance, but not a great one, and that the swing states, therefore, matter.

There are a lot of states that are called swing states but are not. There are non-swing states that are slowly becoming swing states. For example, Georgia and Texas may well be swing states for the next presidential election. Virginia has been considered a swing state for so long that this now reliably semi-progressive/centrist vote-for-the-Dems-for-POTUS state probably shouldn’t be considered a swing state any more. Of course, once a state is a swing state, it should probably not be trusted for several election cycles thereafter.

And, of course, there are swing states that are currently busy swinging back and forth and must be paid close attention to. Here are a few observations on this subset of swing states, based on this morning’s polling and my previous model. (A LOT, perhaps a record number, of polls came out over the last 36 hours, most of which are fairly low quality, and I’m mostly ignoring them.)

Right now, it looks like Trump will win Arizona. My model puts Arizona in Trump’s column. Before you object, FiveThirtyEight agrees with me.

My model puts Iowa in Clinton’s column, but polls disagree, and it looks like Iowa is going to be Trump. This may be where my model fails (likely, paying too much attention to Iowans of the past?) Or, this could be where I get to say, later, “I told you so.” This contrast has been developing for weeks, but there hasn’t been a lot of poling data.

Proposal: If Iowa votes for Trump, take Iowa out of the first slot for the next primary season. (Unless Trump wins the election, then, move to Iowa.)

Nevada really is very, very, close but all indicators suggest that Clinton will win Nevada. My model says Clinton will win Nevada.

New Hampshire probably is not on the table any more as an unreliable state, or a swing state. Does anyone know if this has anything to do with Massachusetts and New Hampshire cross border commuting and car insurance? Eric?

Even though my model is very iffy about North Carolina, it does give it to Clinton by a very small margin, and polls suggest that North Carolina is firmly Clinton.

My model currently puts Ohio barely in the Clinton column. Previous runs of this model put Ohio in Trump’s column. Polls suggest it is very iffy. FiveThirtyEight puts Trump one percent above Clinton, suggesting a fair sight better than 50-50 chance of Trump winning there.

Verily, Ohio is the swingiest of states.

I think everyone and every poll and every model is agreed: Pennsylvania is Clinton. But, Pennsylvania has pulled surprises in the past, so don’t turn your back on Pennsylvania. If you find yourself in the elevator with Pennsylvania, check your wallet.

People have been talking about Utah like it matters. It does not and never will. But it is interesting. Don’t confuse “interesting” with “matters.” Trump will win in Utah.

Are we done calling Virginia a “swing state” yet? Clinton.