This is the website of Abulsme Noibatno Itramne (also known as Sam Minter). Comments here or emails to me at abulsme@abulsme.com are encouraged... or follow me on Twitter as @abulsme.

Get Posts by Email

Categories

AbulTags

Calendar

July 2017
S M T W T F S
« Jun    
 1
2345678
9101112131415
16171819202122
23242526272829
3031  

Electoral College: Trump Surge Sputtering Out?

States with new poll data added since the last update: All 50 states and DC.

Notable changes in: Florida (29), Georgia (16), North Carolina (15), Virginia (13), Wisconsin (10), Colorado (9), Iowa (6), and the tipping point.

National Summary

It has been about two days since the last update. Since then the changes have been mixed, indicating that perhaps the recent movement toward Trump has topped out.

Good for Trump

  • The tipping point moves from Clinton by 2.2% in NH to Clinton by 1.8% in NC
  • Trump’s best case has improved from a 100 EV win to a 112 EV win

Good for Clinton

  • The expected case moves from Clinton winning by 32 EV to Clinton winning by 108 EV
  • Clinton’s best case moves from a 178 EV win to a 210 EV win

Charts

chart-366

chart-367

chart-368

screen-shot-2016-11-06-at-12-04-49647

screen-shot-2016-11-06-at-12-19-58777

Discussion

While we have some indicators moving in each direction, the two to pay attention to are the expected electoral college result and the tipping point. The two “best case” scenarios give all states with a margin under 5% to one candidate or another. Those are both highly unlikely scenarios less than three days out from polls closing.

Even those two indicators are mixed though. If you had to decide who is favored in this update, you would still pick Trump though. Even though his expected electoral total has gone down, the tipping point moves a bit in his favor, meaning that the amount he has to move polls in order to win has lessened.

So while he is losing by more, the amount of effort to change that has still decreased.

The one caveat there is that while the tipping point has improved since my last update, looking at the actual trend line it seems to be down from a recent peak.

If you put all these mixed messages together, it looks like the recent “Trump Surge” may have reached a high water mark… or at least a pause.

At these levels Clinton is of course still winning. The electoral result according to the current averages would be Clinton 323 to Trump 215, a 108 electoral vote win for Clinton. Trump does better than Romney did here, but still loses.

Trump still has ways to win though. The easiest way right now would be for him to pull in Nevada (0.1% Clinton lead), New Hampshire (0.7% Clinton lead), Florida (1.8% Clinton lead), and North Carolina (1.8% Clinton lead). That would get him to a 269-269 tie and throw the election into the House of Representatives, where he would probably win. To win outright, he would need to also pull in Maine’s 2nd congressional district, where Clinton is currently leading by 2.0% in the Election Graphs average.

Moving things by 2% doesn’t seem that big a move in the grand scheme of things, but given the history of the race in terms of what seem to be the candidate’s floors and ceilings, as well as reports of early voting trends in some of those states (specifically Nevada and Florida), moving those states that last little bit is harder than it may seem just looking at the percentages. If anything, the preliminary early voting data seems to indicate Clinton may end up over-performing the polling, not the other way around.

Just to check in with the sites that give odds (which we do not), the chances of a Trump win given all the current information:

The median of these 8 estimates is 12%. That seems reasonable and in line with what I would expect from the Election Graphs numbers and the small amount of time left before the election is over.

2.5 days until polls start to close. We are almost there.

State Details

The following are the detailed charts for all the states that influenced the national changes outlined above.

Moving from Trump to Clinton

chart-369

chart-370

Moving from Clinton to Trump

chart-371

Now possible Trump pickups

chart-372

chart-373

Now possible Clinton pickups

chart-374

No longer possible Trump pickups

chart-375

Other states that helped move the tipping point

chart-376

A note about faithless electors

In the past few days there have been reports of first one, then two, expected electors from Washington state saying that that have decided to or are considering being “faithless electors” by voting for someone other than who they are “supposed to” when the Electoral College votes in December.

Yes, they can do that. In Washington state there would be a $1000 fine if they did, but they can do it, and it would count.

There was also one other potential elector from Georgia who suggested they would be faithless back in August. That potential elector came under extreme pressure and resigned, to be replaced by someone who would presumably vote the “right” way.

That might well happen to these two as well. But faithless electors can and do happen every few election cycles. In the last 50 years there were faithless electors in 2004, 2000, 1988, 1976, 1972, and 1968. That is 6 out of 12 elections during that period. Half. It would not be surprising if there were one or more faithless electors in 2016, especially given the contentious nature of the race. They have never made a difference in the final outcome.

The Election Graphs estimates show what the results “should be” given who wins various states. They do not take into account statements by individual electors on their intentions. If a scenario arises where it is close enough for electors who have made statements like this to potentially make a difference in the outcome or throw the race to the House, we will of course make note of that. And once electors vote in December, we will show the final tallies, including any votes cast by faithless electors.

A note about McMullin

Right now the Real Clear Politics average in Utah has McMullin 12.4% behind the lead. The 538 average in Utah has him 10.3% behind the lead. Although McMullin was a lot closer than anyone expected, he never got closer than 5% to the lead in Utah in these averages and now appears to be fading. 538 still has him in second place, but it doesn’t seem likely at this point that he will win electoral votes.

If that changes, I will of course note it in these updates and on the Election Graphs site.

For more information…

This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks a state poll based estimate of the Electoral College. 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.

Edit 13:56 UTC to add an additional paragraph break.

Electoral College: Tighter and Wider at the Same Time

States with new poll data added since the last update: All 50 states, DC, and the Maine congressional districts.

Notable changes in: Florida (29), Pennsylvania (20), Arizona (11), Iowa (6), Kansas (6), Utah (6), Alaska (3), and the tipping point.

National Summary

It has been about five days since the last update. Since then the movement has been decidedly mixed.

Note that as of this update, no polls have been included that contain interviews done after the release of the “Comey Letter”.

Good for Trump:

  • The expected result has moved from a 130 EV Clinton win to only a 62 EV Clinton win
  • Clinton’s best case has declined from a 234 EV win to only a 216 EV win

Good for Clinton:

  • The tipping point moves from Clinton by 4.1% in PA to Clinton by 5.3% in PA
  • Trump’s best case declines from a 34 EV win to a 6 EV loss

Charts

chart-31

chart-32

chart-33

screen-shot-2016-10-31-at-21-57-53

Discussion

While some metrics have moved in Trump’s direction and others in Clinton’s, in the end this picture is bad news for Trump. Why is that? Fundamentally, doesn’t the fact that the “expected case” moved toward Trump mean that he is doing better?

Well, it means the margin he will lose by is getting smaller, yes. However, if you are trying to look for scenarios where Trump would actually catch up and win, the tipping point is what you want to be looking at. To win he has to move the tipping point state (or states he is even further behind in) over to his side of the line. And the tipping point is moving further away at the moment.

If you look at the tipping point chart you can see that Trump had indeed been making progress on the tipping point starting about when the Access Hollywood tape dropped. But that started to reverse around the 23rd. At the same time that his expected electoral college result was still improving.

This is still confusing. What is happening? Well, all the states aren’t moving in lock step in reaction to events. A number of close states have moved toward Trump, like Florida for instance. But key states like Pennsylvania have been moving away from him. Because of this divergence, we can have Trump improving his electoral college “score” while simultaneously the path to 270 electoral votes gets harder.

And in fact, at the moment, 270 looks out of reach for Trump. Look at the spectrum of the close states above. Right now Trump could win every state he is ahead in, then also pull over the states where he is close…  Nevada (0.4% Clinton lead), Iowa (1.5% Clinton lead), Maine CD2 (2.5% Clinton lead), and North Carolina (3.8% Clinton lead)…  and Trump would still lose by 6 electoral votes.

To win, Trump has to win all of the close states, plus Pennsylvania, where Clinton has a 5.3% lead at the moment… or one of the other larger “Strong Clinton” states… Wisconsin, Michigan, Colorado, Minnesota, or Virginia… all of which Clinton currently leads by even bigger margins. (New Hampshire isn’t enough to push Trump over the edge with only 4 EV.)

This is a really big pull with so little time left. The Trump side seems to be hanging their hopes on the fallout from the Comey Letter. This may cause some movement, but early indications are that almost all voters made up their minds on the Clinton email issue and if it mattered to them long long ago, and this is unlikely to change many minds.

We’ll start to see if there is any movement due to the letter within a few days. But there just are not that many days left. Sites that give odds for Trump winning range from less than 0.01% on the low side to 24.4% on the high side. The median odds seem to be around 5%. That isn’t impossible. But those odds are low enough that a Trump win would be a very big surprise.

There is another goal that Trump has a much better shot at though. If the election ends up exactly how things look today, Trump will lose by only 62 electoral votes. That would mean that Trump would do better than both McCain (lost by 192 EV) and Romney (lost by 126 EV). It may not be an actual win, but Trump improving on the performance of both previous Republican candidates may give the message to Republicans that a Trump like candidate in 2020 might be the way to go rather than returning to Romney or McCain style Republicanism.

We shall see.

8.1 days left until we start getting actual election results.

We are almost there.

State Details

The following are the detailed charts for all the states that influenced the national changes outlined above.

Moving from Clinton to Trump

chart-34

chart-35

Moving from Trump to Clinton

chart-36

Now a possible Clinton pickup

chart-37

Now out of Trump’s reach

chart-38

Now out of Clinton’s reach

chart-39

chart-40

Also influencing the tipping point

chart-42

chart-41

A note about McMullin

Election Graphs is designed around showing the Republican – Democrat margin, and there is not enough time before the election to revamp the site to properly account for 3rd parties that are strong enough to have a shot at winning electoral votes.

Right now the Real Clear Politics average in Utah has McMullin 6.0% behind the lead. The 538 average in Utah has him 7.3% behind the lead. If/when McMullin is within 5% according to either of these averages, I will add notes to both the Utah and National pages about the possibility of McMullin winning Utah’s 6 electoral votes.

For more information…

This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks a state poll based estimate of the Electoral College. 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.

Electoral College: Trump improving, but not enough

States with new poll data added since the last update: All 50 states, DC, and the Maine congressional districts

Notable changes in: Iowa, Georgia, Utah, Alaska, Indiana, Missouri, Maine (CD2), and the tipping point

National Summary

It has been about six days since my last update and the trend since then has been toward Trump. Here are the stats:

Good for Trump:

  • The expected result changed from a 140 EV Clinton win to a 130 EV Clinton win
  • The tipping point moved from Clinton by 4.9% in PA to Clinton by 4.1% in PA
  • Clinton’s best case declined from a 238 EV win to only a 234 EV win

No change

  • Trump’s best case remained a 34 EV win

Charts

chart-20

chart-21

Discussion

With all the data we have now, looking back to the release of “the tape” on October 7th the overall movement has been mixed. Both candidates have improved their “best cases” as states that were not close have gotten closer… on both sides. So this doesn’t show a move toward one candidate or another, but rather some red states getting less red while simultaneously some blue states got less blue.

Some of the really close states moved across the center line… in both directions… resulting in a net movement of the expected electoral college margin toward Trump by 24 electoral votes… but the most indicative sign of movement toward Trump is actually the tipping point.

Since the release of the tape the tipping point has moved from Clinton leading by 6.0% (in Virginia) to Clinton leading by only 4.1% (in Pennsylvania). So… with some of the worst news cycles for a politician in decades, with day after day after day of negative revelations…  Trump improved his position by nearly 2%.

How to interpret this? My best guess… Most people were pretty well set on their Trump vs Clinton choice prior to the reveal of the tapes and the subsequent sexual assault allegations. In the wake of the first debate the part of the public that was persuadable had already swung us from near Trump’s ceiling to near Trump’s floor. With the string of additional bad news, very few die hard Trump supporters moved away from him, while some number of those persuadables viewed this as the press piling on Trump and went back to him thinking this was unfair.

But to be honest, the above is trying to make up a story to fit the numbers. Is it right? I don’t know.

In the end though, the reality is that in the face of a hurricane of bad coverage, Trump has actually improved his position.

Despite the improvement, Trump is still losing by a significant margin, but Trump’s best case does still includes a win.

If he keeps all the states he leads, then flips Arizona, North Carolina, Florida and Pennsylvania, he could still pull out a narrow win. Arizona is close enough it could flip back easily enough. North Carolina and Florida have both been back and forth over the past couple months, so seeing them go red again would not be shocking.

Pennsylvania is the hard pull. While he made it close in September, Trump has never led in Pennsylvania. And Trump needs Pennsylvania to win. Right now, this is Clinton’s firewall.

Clinton would clearly like to be leading in Pennsylvania by more than 4.1%… but even at that margin, it is a tall wall for Trump to climb.

There are 13.7 days left until the first polls close. With less than two weeks left to go now, it is almost certainly too late for Trump to flip all the states he would need to flip in order to win. But if he manages to continue the trend of the last couple of weeks, it is enough time for him to make it close enough that Democrats will start to panic.

State Details

The following are the detailed state charts for states that influenced the status changes since the last update.

Weak Clinton to Weak Trump

chart-22

Weak Trump to Weak Clinton

chart-23

Influencing the tipping point

chart-24

chart-25

Now possible Clinton pickups

chart-26

chart-27

No longer possible pickups for Clinton

chart-29

chart-30

chart-28

A note about McMullin

Election Graphs is designed around showing the Republican – Democrat margin, and there is not enough time before the election to revamp the site to properly account for 3rd parties that are strong enough to have a shot at winning electoral votes.

Right now the Real Clear Politics average in Utah has McMullin 5.5% behind the lead. The 538 average in Utah has him 8.5% behind the lead. If/when McMullin is within 5% according to either of these averages, I will add notes to both the Utah and National pages about the possibility of McMullin winning Utah’s 6 electoral votes.

For more information…

This post is an update based on the data on ElectionGraphs.com. Election Graphs tracks a state poll based estimate of the Electoral College. 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.

Electoral College: Trump’s Polling Meltdown

States with new poll data added since the last update: All of them. Multiple times. There are additional new polls since I had to cut things off for this update as well. Those will be in the next update.

National Summary

The last week of polls has been devastating to Trump’s position. In the last update we were just starting to see the reaction from the first debate. With this batch the post-debate polling continued to roll in, and while there were a couple of minor moves toward Trump, the net change in this update was a dramatic move toward Clinton:

  • The expected result moved from Clinton by 6 electoral votes to Clinton by 154 electoral votes
  • The tipping point moved from Clinton by 2.1% in CO to Clinton by 4.5% in MI
  • Trump’s best case moved from a 60 EV win to a 26 EV win
  • Clinton’s best case moved from a 208 EV win to a 196 EV win

Looking at the charts:

chart-321

The previous update mentioned some Trump losses starting a week or two before the debate. With a lot more polls filling in the time period right before the debate as well as right after, it now looks like those losses were reversed before the debate.

In the time after the debate though, all three indicators on the chart have moved in Clinton’s direction. The race is now essentially back to where it was at the beginning of September. We’re not yet back at Clinton’s high point from August, but it is getting close.

chart-322

The tipping point chart is even more stark. As with the electoral college view, it looks like Trump’s mid-September declines were reversed by debate time, but after the debate everything falls off a cliff. Most of Trump’s gains from August and September are gone. Right before the debate he topped out with Clinton only ahead in the tipping point metric by 0.4%.

Clinton now leads by 4.5%. The tipping point state is Michigan. Clinton only has to increase her lead slightly there to not just take Michigan out of reach for Trump, but the Presidency.

What we are seeing in the polling results right now at this point probably fully factors in the debate itself. But we are just starting to see the impact from Trump’s “worst week in presidential campaign history ” that started with the debate, continued with his attacks on Alicia Machado, and ended with the leak of some of his tax information. The events of that week are not yet fully baked in.

And of course, the polls do not yet reflect the VP debate, or more importantly, the 2005 Trump video that was leaked on Friday. It is hard to imagine the events of the last few days NOT having an impact on the polls.

And the next debate is coming up. It should be interesting. If you were not planning on tuning in before, change your plans.

31.7 days left until the first actual election results start coming in.

State Details

A lot of states moved in order to produce the summary results above. For those interested in looking at the state by state details, they are below:

Weak Trump to Weak Clinton

chart-324

chart-326

chart-327

chart-325

chart-323

Moving out of reach for Trump

chart-330

chart-328

chart-329

Moving out of reach for Clinton

chart-331

Moving into reach for Trump

chart-332

States that changed categories but then moved right back again

chart-335

chart-333

chart-334

For more information…

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.

Electoral College: Trump Close Enough to Smell Victory

States with new poll data added since the last update: All 50 states, DC, and the Maine congressional districts. Notable changes in Colorado, New Mexico, Iowa, Nevada, Maine (CD2), Kansas, and the tipping point.

National Picture

Trump continues on the upswing, and at this point Clinton only leads by the very narrowest of margins. One more round of polls in his favor could easily put Trump in the lead for the first time this cycle.

With this update, we have the following changes:

  • The expected case moves from Clinton by 32 EV to Clinton by only 6 EV
  • The tipping point moves from Clinton by 3.4% in MI to Clinton by 0.8% in NM
  • Trump’s best case improves from winning by 66 EV to winning by 94 EV
  • Clinton’s best case drops from a 190 EV win to a 178 EV win

The full current summary:

screen-shot-2016-09-19-at-00-36-56579

chart-287

There is no denying this is now a razor thin race. This is close. If the election was held today, it could go either way.

Looking at the trends:

chart-285

chart-286

In both measures, Trump bottomed in mid-August and has been on a strong upswing ever since. Or is it that Clinton has been collapsing? In the end, the result is the same.

Trump is now far outside the “McCain to Romney” zone where he had been stuck. The question is if he is establishing a new normal, with many people permanently converted to his cause, or if this is an unsustainable high that will soon undergo reversion to the mean.

The race is not today. There are 51.0 days until we start getting the first real election results. A lot will happen between now and then. The first debate is now less than eight days away. Will it change anything? We shall see…

Individual States

I won’t provide individual commentary on all of these, but for those wanting to see the details on the states that changed the national summary this time around, here they are. Click through on any chart for the full detail page with individual polls listed.

Moving from Clinton to Trump:

chart-288

chart-289

chart-290

Now close enough to be in play for Trump:

chart-291

chart-292

No longer close enough to be in play for Clinton

chart-293

Another state whose movement influenced the tipping point

chart-294

Note on “50 State” polls

Since Election Graphs tries to be as inclusive as possible with polls, we are including some “50 state” polls that some other folks have decided to exclude because of various methodological concerns. Others, like 538, include these polls, but give them very low weights in their models. Because Election Graphs tries to “keep it simple” we don’t do that kind of weighting. Everything just counts and we do a straight average.

Two of these “50 state” polls are now producing weekly results, Google Consumer Surveys and Reuters/Ipsos.

Honestly there are some crazy things in there. For instance:

  • Google thinks Clinton is leading in Kansas, some weeks by a very large margin. No other pollster has seen this. Everybody else shows a significant Trump lead.
  • Google thinks Trump is only 7% behind in DC. In DC? Really? In DC Obama beat Romney by 83.6% and McCain by 85.9%. The idea that Clinton is only 7% ahead of Trump in DC is pretty much ludicrous.
  • The last few Ipsos polls have shown a Trump lead between 13% and 23% in Texas. The most favorable poll to Trump other than Ipsos has a 11% Trump lead, and most show it significantly narrower than that.
  • In general it seems to be very common for these two polls to produce results which are outliers compared to other pollsters… Sometimes they show a result far more favorable to Clinton than anyone else, sometimes far more favorable to Trump than anyone else. The direction isn’t uniform, they just sometimes seem way off the mark.
  • In many states the sample sizes are very low, and you see wide swings. For instance in West Virginia the Google poll moved from Trump +29% to Clinton +1% in one week! Yes, when you average them you get a 14% Trump lead, which is probably reasonable for West Virginia… but still!
  • There are more examples, but you get the idea. In most states, these polls give roughly the results you would expect, in line with other recent polling, but occasionally, especially in smaller states, they seem to be clear outliers.

The craziness of some of these polls leads me to sometimes question my tenets of including everything and keeping it as close to a straight average as possible. But we’re sticking with that and we will see how it does.

And even with this the general notion of this site, that even with very simple assumptions you end up with similar results to what folks with complex models get seems to be holding up. Although the exact state by state results may differ slightly, every analysis I have read has the same general conclusion right now. The race is now much much closer than it was, although Clinton retains a narrow lead.

While these polls are a bit crazy, in the competitive states there are enough other polls so the average seems to be at least somewhat resilient when outliers like these pop up. And in the non-competitive states, well, these 50 state polls may be the only data in the average, so they may screw up the margins a bit, but in the end it doesn’t change the status of the state.

For more information…

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.

Electoral College: Trump breaks the Romney barrier

States with new poll data added since the last update: All 50 states and DC. Multiple times each. Notable changes in Florida, North Carolina, Ohio, Iowa, Pennsylvania, Kansas, Colorado, Wisconsin, Maine (All), and Georgia.

There have been tons of polls, including “50 state” polls from three different sources, one of which had weekly data going back four weeks that needed to be added. There were many shifts caused by this batch of polls. We’ll look at all the charts, but lets start with the headline:

The Romney Barrier

chart-273

The averages in Florida, Ohio, and North Carolina all moved from Clinton to Trump. Iowa did move in the opposite direction, but it is much smaller, so the net change was dramatically toward Trump.

Before this batch of changes, the “expected” result if everybody won the states they lead was Clinton 341 to Trump 197… a 144 electoral vote win for Clinton.

With this batch of polls, the expected result is now Clinton 285 to Trump 253. This is still a Clinton win, but now by only 32 electoral votes.

From February onward this race had been stuck in the zone with Trump losing by margins between McCain’s loss (192 electoral votes) and Romney’s loss (126 electoral votes).

For the first time since February, Trump is now looking like he would do better than Romney did. Trump now matches Romney’s map, but adds Florida and Ohio. It isn’t quite enough to win, but Trump is now closer than he has ever been… and doing better than the last two Republican candidates.

Some folks were looking at Clinton’s lead in mid-August and were trumpeting the possibility of a Clinton landslide in early August. It was just looking like a solid win for Clinton back then, not a landslide, but things have actually moved in the opposite direction. A 32 electoral college win is more like a squeaker. It would only take Michigan flipping to get us to a 269 to 269 electoral college tie.

(There were also changes in the “best cases” from a lot of states moving in both directions. Trump’s best case went from winning by 68 electoral votes to winning by 66 electoral votes. Clinton’s best case went from winning by 210 electoral votes to winning by 190. So the overall range of possible outcomes reduced.)

Tipping Point

chart-274

Now, despite the expected case moving in Trump’s direction, changes in Florida, Michigan, Wisconsin and Pennsylvania had the net effect of… zero. Well, to be precise the tipping point changed from a 3.4% Clinton lead in Florida to a 3.4% Clinton lead in Michigan. So the state that holds the balance shifted, but the actual percentage did not.

Looking at the trends though, rather than just the change in this update, while the expected case has clearly been moving in Trumps direction the last few weeks, the tipping point has been trending toward Clinton.

What does this mean?

At a basic level it means that while Trump has made the race closer, pulling over the next few states he needs to win is getting harder.

3.4% is still close, but if Trump is at or near his ceiling, it may be a tough last 3.4% to move.

To be absolutely clear, with all of the above Clinton is still a strong favorite. Trump has made it a lot closer than it was, but he still has a lot of work to do to pull the remaining states he would need past the line to secure an electoral college victory. It isn’t impossible. It is just very hard.

Individual States

OK, with the two national summaries out of the way, time for a run down of all the states that shifted in ways that caused changes to the expected or best cases. I’ll just do a run down of all the graphs without commentary on each one. For the most part they speak for themselves. Keep an eye out for the polls that look like outliers. In some cases if new polls don’t back those up, the changes described here may reverse themselves quickly as new polls come in.

States moving from Clinton to Trump:

chart-275

chart-276

chart-277

States moving from Trump to Clinton:

chart-278

States now in play for Trump:

chart-279

States now in play for Clinton:

chart-280

States no longer in play for Trump:

chart-281

chart-282

chart-283

States no longer in play for Clinton:

chart-284

What’s Next

This batch of polls was huge, so in the time it took me to complete it, there have already been a bunch of new polls which I will be adding as soon as I can. Even if those polls don’t change anything, there will be polls soon enough that do. This is a dynamic race, and as I post this we have 54.7 days left until we start getting the actual results. Hang on tight, we’re in for a nice ride before we are done.

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.

[Update 07:16 UTC to add comment about outliers.]

[Updated 07:27 UTC to fix some awkward wording.]

Electoral College: Iowa Back on the Red Side

States with new poll data added since the last update: Virginia, Iowa, New Hampshire, Ohio, California, Texas, Pennsylvania, Florida, New York, Illinois, North Carolina, Indiana, Georgia, Missouri, Arizona, Massachusetts, Wisconsin, Tennessee, New Jersey, Minnesota, Washington, Michigan, South Carolina, Maine (All), Louisiana, Utah, Oklahoma, Alabama, Nebraska (All), Oregon, Kansas, Maryland, Kentucky, Mississippi, Nevada, Connecticut, Arkansas, West Virginia, Colorado, Idaho, New Hampshire, Alaska

We have the weekly batch of polls from Ipsos, plus a handful of other scattered polls. One state changes category:

chart-262

After having been “Weak Clinton” through July, for the last month the average in Iowa has been hovering right around the center line. It had become “Weak Trump” for a short time in the second week of August, but then quickly moved back to Clinton. With the latest couple of polls showing Trump leading, the average moves once again to the Trump side of the line.

This is a narrow lead… 0.5%… it could move back to the other side of the line with the next poll. But for now, Trump leads in Iowa, and so this changes the expected results line:

chart-263

This brings the expected case to a 144 electoral vote win for Clinton. This matches the best Trump has done since April. In terms of the expected case, the metric that actually matters the most, this has seemed like Trump’s ceiling ever since he locked up the nomination. He has been able to move his best case, but in terms of moving blue states over to his side of the line, he just hasn’t been able to push past this level.

With average changes in Ohio and Florida, the tipping point also moved:

chart-264

Prior to the current batch of polls, the tipping point was Clinton +3.8% in Ohio. The new polls exposed a brief spike to Clinton only leading by 2.4% in Michigan, but then falling back to Clinton +3.0% in Florida. The net change in this update is still a 0.8% movement toward Trump though.

As with the expected electoral college result, Trump is near the top of his recent range. A 3.0% Clinton tipping point means that only a small uniform shift across all states would push Trump into the lead. The race is actually relatively close here. So why is most of the talk still about how Clinton is clearly winning?

Well, first of all, because she is. Trump has never led in either the electoral college expected case or the tipping point. She is just objectively in the lead and always has been.

Second, we have seen the whole race wiggle around in a zone. Things move up and down within that zone, but seem to have trouble escaping it. On the one side Clinton can’t seem to break out to move things from just a regular comfortable win to doing better than Obama did against McCain, and on the other side, Trump can make the race tighter, but hasn’t been able to break the barrier of matching or improving on Romney’s performance.

Since April, it has looked like Trump would do better than McCain (192 EV loss), but not as well as Romney (126 EV loss). It just seems like that is where this race wants to be, at least in terms of the electoral college. Of course, both of those men lost. This range does not include winning.

So for Trump to win, things need to change and move into a zone they have not been in before. And he is running out of time. The next opportunity will likely be the debates. The first one is in 22 days. Then we have 65.3 days until results start rolling in.

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.

Electoral College: Trump continues to tighten the race

States with new poll data since the last update: Iowa, Colorado, Michigan, Ohio, Pennsylvania, Virginia, Wisconsin, Florida

In advance of the conventions there has been a large volume of recent polling. For the first time with this update the polling average in some states is based fully on polls with their middates within the last seven days. Right now there is no shortage of state level polling and we are seeing lots of movement. Some of this may be random movement depending on which polls are most recent at any given time, but it is likely we are also detecting actual changes on the ground as well.

With the current batch of polls there are notable changes in four states. Pennsylvania, Ohio, and Iowa move toward Trump, while Colorado moves toward Clinton. Lets look at all four of these changes individually, then we’ll review the national picture.

Pennsylvania [20 EV]

chart-201

Polls for both Quinnipiac and Marist were added with this update. These were released and added on the same day. You can see above though that the Quinnipiac poll (which covers a slightly earlier date range) actually briefly spiked Pennsylvania into Weak Trump before the Marist poll pushed things back to the Clinton side.

Net result together though, Pennsylvania tightens, but Clinton is still ahead… by a narrow 2.0% margin.

With Pennsylvania in play again, Trump’s best case improves. In addition, this change contributed to moving the tipping point toward Trump.

Ohio [18 EV]

chart-202

With the newest polls Trump sees a bump as some polls very favorable to Clinton from mid-June roll off the average. Clinton now leads Ohio by only 0.8%. There is no change in categories here. Weak Clinton before, Weak Clinton after. But the movement in Ohio combines with the movement in Pennsylvania above to impact the tipping point.

Colorado [9 EV]

chart-203

After spending what seemed like forever as the least polled “close state”, we all of a sudden have lots and lots of polls in Colorado. The end result? Colorado had looked like a very close state based on the average results of the last few elections. But now that we finally have a decent volume of Clinton vs Trump polls, Clinton has a clear lead. She is now ahead by a healthy 7.0%.

So in the only one of today’s moves that is in Clinton’s direction, Colorado moves out of reach for Trump. At least for now. This reduces Trump’s best case.

Iowa [6 EV]

chart-204

New Gravis and Marist polls push Loras polls from the end of June off the average… polls which now look like clear outliers… and so the average moves in Trump’s direction. Iowa now looks very close, sitting at a 1.0% Clinton lead.

With Iowa close again, it is once again included as a possible Trump pick up, improving his best case.

National View

chart-205

Once you factor in the two states moving toward Trump and the one state moving forward Clinton, Trump’s best case moves from winning by 6 electoral votes to winning by 40 electoral votes. No net change for the expected case despite the temporary spike.

The more impressive change though is actually with the tipping point:

chart-206

With two states moving in ways that impact the tipping point, it moves from Clinton by 4.3% in Ohio to Clinton by 2.0% in Pennsylvania. This is the best position Trump has had in this metric since last August.

Trump is of course still behind. But a 2.0% tipping point means you only would need 1% of people flipping from Clinton to Trump to push Trump into the lead, or undecideds breaking in his direction. For the first time in a long time, this is looking like a close race again.

Screen Shot 2016-07-14 at 18.47.29595

Looking at the center of the spectrum, Trump’s shortest path to being in the lead goes through Ohio, Nevada, Iowa, and Pennsylvania. All of those states have Clinton in the lead, but by less than 2.0%, a margin that could easily evaporate overnight.

With the Republican convention starting in less than 4 days, if we see Trump get the traditional “bump” we may well see him actually in the lead for the first time this cycle sometime in the next couple weeks. This assumes of course the convention ends up coming off smoothly. If the convention becomes chaotic, the bump may evaporate.

Also acting against the bump, the Democratic convention starts only a few days after the Republican convention ends. There is no significant gap between the two conventions. Which may make any bump too transitory to measure in an electoral college based view.

If we see even a minor bump though, at this point it would be enough to put Trump in the lead.

To close up today, a quick look at the current map:

chart-207

117.2 days until polls start closing on Election night 2016. The conventions are upon us. Things will be nuts from here until November. Hold on tight! :-)

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.

Electoral College: Clinton now strong in Iowa

New polls since last update: Iowa, Florida, New York, North Carolina, Nevada, Arizona, Ohio, New Hampshire, Pennsylvania, Wisconsin, Michigan.

This is now the sixth day in a row that there has been movement in the Election Graphs models. I didn’t expect this kind of pace until September or October. Hopefully it won’t be this way the entire time until November, or your faithful commentator will be a burned out cinder by then.

In any case, in today’s update two changes contribute to the changes:

Iowa [6 EV]

chart-192

The new Iowa polls by Loras (one with Johnson and Stein, one without) initially look like outliers. But there has been a wave of these polls showing bigger Clinton leads than we had seen previously this week. Are they all outliers? There will be no way to tell for sure until there is more polling. But do not be surprised if this “movement” reverses once these particular polls roll off of the average.

In the mean time though, we include all the polls in the average, and Iowa moves from Weak Clinton to Strong Clinton, which means it is no longer included in Trump’s best case.

chart-193

With Iowa not in play, Trump’s best case moves from losing by 40 electoral votes to losing by 52 electoral votes. This is still better than the picture was at the beginning of May, which Trump’s best case was to lose by 78 electoral votes, but even so, this isn’t a pretty picture for Trump.

After weeks of saying here that a Trump collapse was not yet visible in the state polls, you can now clearly see a Trump bump, followed by a collapse. Notable though is the fact that this is only visible in Trump’s “best case”, not the “expected case”. Trump made a number of states look close, but aside from some transitory short term changes that were erased as more polls were added, Trump has not moved any states over to his side of the line.

The expected electoral college result has been static at Clinton 338, Trump 200 since March.

Florida [29 EV]

chart-194

Florida does not change categories this time, but along with Iowa moves the tipping point.

Iowa moved toward Clinton. The Iowa poll covered a slightly earlier timeframe than the Florida poll. Iowa moved the tipping point from Clinton by 5.5% in Minnesota to Clinton by 6.5% in New Mexico.

But then Florida moved a bit toward Trump and almost immediately moved the tipping point back to Clinton by 5.7% in Florida.

chart-195

You can see the spike down to 6.5% then immediate movement back to 5.7% on the chart. Net change is still a 0.2% movement toward Clinton.

The Trump bump followed by a collapse is now clearly visible on the tipping point chart as well. Note that the “bump” only moved Trump to the zone he was in back in March. It never completely reversed the negative trend of the primary months. And this entire graph is on the blue side of the center line. Trump has never had a lead.

On this metric, with the recent collapse just hit his worst level ever (then bounced back a bit).

The way to read this trend along with the flatness in the expected electoral college result is that while Trump hasn’t fallen further behind in the electoral college, the states he would need to flip in order to win are moving further away from him, making the task of catching up with Clinton look harder and harder.

The picture has changed dramatically in just the last two to three weeks though, so it is important to keep in mind how volatile these numbers can be. Especially since a number of the polls that caused these changes initially look like they might be outliers.

It would not be surprising at all to see some movement back in the trump direction in the two weeks prior to the Republican convention. And of course the “usual” pattern is a bump after the Convention.

So keep watching. 130.3 days until polls start to close on election day.

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.

Republicans: Iowa results officially certified, Huckabee gets a delegate too!

The Iowa results on the Republican side were officially certified on Wednesday, earlier than anybody expected, and with a surprise. Almost every source believed 27 delegates would be decided by the caucus results, but the remaining 3 “automatic delegates” who are party leaders, would attend the convention unbound and able to support whoever they wished. But it looks like those three delegates will be bound by the caucus results as well.

When you do the required math on 30 delegates instead of 27, everything remains the same, except Huckabee gets a delegate too.

Updating the chart showing the “% of remaining needed to win” graph:

chart-33

And the raw numbers:

Screen Shot 2016-02-04 at 13.17.12929

Huckabee and Paul have suspended their campaigns, but since these are bound delegates, they keep them unless the candidates officially release them. I haven’t seen a report of that yet, so won’t remove them from the list at this point. Not that it matters much for single delegates. This is a much more important factor when candidates drop out later in the race once they have already accumulated a significantly sized chunk of delegates.

Huckabee’s one delegate makes no difference to the analysis I posted Tuesday or the follow up from Wednesday.

We still have three candidates showing strength going into New Hampshire It is within the realm of possibility that New Hampshire will add a fourth to that list, but just as likely it will solidify the “three way race” picture. And as I outlined in the Wednesday post the specific New Hampshire delegate rules matter a lot. Specifically the rules essentially give extra bonus delegates to the winner, and don’t allow delegates at all for anyone coming in under 10%. This is a recipe to transform a plurality win to a healthy delegate majority.

So if you are watching the polls as they start to include post-Iowa data, watch carefully not only who is in the lead, but exactly who is above or below the 10% threshold, because that will make a huge difference in the delegate counts.

And an 8-7-7 split with only 1.21% of delegates determined is essentially a tie. And a tie that will soon be overwhelmed numerically by the results from New Hampshire and beyond. So don’t get sucked too deeply into the spin from all quarters taking about who is up and who is doomed. It is still too early for that. (Well, at least if you don’t drop out.)

Five days until New Hampshire.

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. All charts above are clickable to go to the current version of the detail page the chart is from, which may contain more up to date information than the snapshots on this page, which were current as of the time of this post. 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.