Kalm is Calling the Election

[kalm]

With 0% of precincts reporting!

Trump is within striking distance in most polls (RCP average is Clinton + 5.7 and falling). Clinton fatigue will continue to deepen. Seriously, how many Clinton fanatics are out there? I'm not talking about the ones standing behind her, paid to hold up pre-printed signs at less than well attended rallies. How many Hillary supporters do you know who REALLY like her and all that she's about?

Two YUGE caveats...unless Clinton gets indicted or significantly damaged by the investigation, or an indie candidate throws their hat in the ring...The Donald will be our next president.

You heard it here first!

KalmNet - The Most Trusted Name in Fair and Balanced Leaning Forward News.

[kalm]

Speaking of independents or third party candidates, here's a great read taking us back to the election of 2012.

Conservatives cannot let this pass without a response, which is why an independent candidacy is essential. It isn't just a way to win the White House or hold the line in Congress, it is a statement of purpose and a warning to the Republican party that commitment to conservative ideas comes before party loyalty. If the party abandons the principles that have guided it for decades, it may no longer count on conservative support.

In this way, a conservative third-party candidate may serve the same purpose that Teddy Roosevelt's Bull Moosers did in 1912. Conservatives are inclined to castigate this party for its progressivism, and rightly so, but the Bull Moosers were reacting to a sense that the Republican party put special interests ahead of the general welfare. The first plank of the platform read:

Political parties exist to secure responsible government and to execute the will of the people. From these great tasks both of the old parties have turned aside. Instead of instruments to promote the general welfare, they have become the tools of corrupt interests which use them impartially to serve their selfish purposes. Behind the ostensible government sits enthroned an invisible government owing no allegiance and acknowledging no responsibility to the people. To destroy this invisible government, to dissolve the unholy alliance between corrupt business and corrupt politics is the first task of the statesmanship of the day.

This is a fitting description of the Republican party in 2016: the unprincipled Trump standing at the head of a corrupt political force, flanked on one side by K Street lobbyists who use government to line their pockets and on the other by professional politicians regurgitating Reaganesque talking points but lacking true beliefs. Having betrayed the principles that have animated it for so long, this party cannot secure responsible government anymore.

Here's how the Bull Moosers reacted to Republican corruption a century ago:

Unhampered by tradition, uncorrupted by power, undismayed by the magnitude of the task, the new party offers itself as the instrument of the people to sweep away old abuses, to build a new and nobler commonwealth.

Whatever one may say about the progressives, they weren't wrong about the GOP, which had been overrun by corruption. Instead of sitting idly by, the Bull Moosers rebuked their crooked old party and formed a new coalition.

http://www.weeklystandard.com/the-lesso ... le/2002374

[Ibanez]

With 0% of precincts reporting!

Trump is within striking distance in most polls (RCP average is Clinton + 5.7 and falling). Clinton fatigue will continue to deepen. Seriously, how many Clinton fanatics are out there? I'm not talking about the ones standing behind her, paid to hold up pre-printed signs at less than well attended rallies. How many Hillary supporters do you know who REALLY like her and all that she's about?

Two YUGE caveats...unless Clinton gets indicted or significantly damaged by the investigation, or an indie candidate throws their hat in the ring...The Donald will be our next president.

You heard it here first!

KalmNet - The Most Trusted Name in Fair and Balanced Leaning Forward News.

Did I miss a news story?

[Pwns]

Nope. Trump needs to flip 62 electoral votes from 2012 and win every state Romney did to win. Do you realize even if Trump somehow manages to flip Ohio, Pennsylvania, and Florida he will still need Arizona (where Hillary is winning in current polls) to win? And Arizona could very well go blue as Arizona has quite a large number of Hispanic voters. Not to mention (at least for now) that Georgia is in reach.

[OL FU]

With 0% of precincts reporting!

Trump is within striking distance in most polls (RCP average is Clinton + 5.7 and falling). Clinton fatigue will continue to deepen. Seriously, how many Clinton fanatics are out there? I'm not talking about the ones standing behind her, paid to hold up pre-printed signs at less than well attended rallies. How many Hillary supporters do you know who REALLY like her and all that she's about?

Two YUGE caveats...unless Clinton gets indicted or significantly damaged by the investigation, or an indie candidate throws their hat in the ring...The Donald will be our next president.

You heard it here first!

KalmNet - The Most Trusted Name in Fair and Balanced Leaning Forward News.

Win lose or draw, I am just happy to hear that is over

Wake me up in 4 years either way

[JohnStOnge]

Well, I hope you're wrong Kalm. Hopefully the Democrats will be successful in exploiting the target rich environment that is Trump's background of behavior and statements.

Otherwise: I don't think there is a "significant" falling trend for Clinton in polling at this point. You've had stuff like Trump up by 2 then Clinton up by 13 in two consecutive polls in the RCP average. There's a lot of variation. I did a trend thing for the poll results up right now and there's no significant trend. But you can see that by just noting that Clinton was +3 in the earliest poll up right now....conducted 4/17 through 4/20...and is +3 in the latest one...conducted entirely on 5/10. In between there were a couple of Clinton spikes to 11.. in a poll conducted 4/20 through 4/24...and to 13... in a poll conducted 4/28 through 5/1.

To list it out, it's like this for Clinton in the set of polls listed right now for Clinton:

+3, +7, +11, -2, +13, +6, +2.

Things could change but right now there's not a lot of evidence of any trend one way or another. Just an overall impression that Clinton is ahead by a modest margin and a lot of variability in the results of different polls.

One thing that gives me hope, though:

The polls that gave Clinton +11 and +13 (USA Today and CNN) are both rated A- by the Poll ratings at http://fivethirtyeight.com/interactives ... r-ratings/. And those ratings were constructed well before the current situation. The other four polls are rated C+, B, C, B-, and C.

In other words, the two most highly rated polls according to that site are the two that showed Clinton up by double digits. At least right now:

[Ivytalk]

Klam has a 50-50 shot, so what the hell.

[kalm]

Well, I hope you're wrong Kalm. Hopefully the Democrats will be successful in exploiting the target rich environment that is Trump's background of behavior and statements.

Otherwise: I don't think there is a "significant" falling trend for Clinton in polling at this point. You've had stuff like Trump up by 2 then Clinton up by 13 in two consecutive polls in the RCP average. There's a lot of variation. I did a trend thing for the poll results up right now and there's no significant trend. But you can see that by just noting that Clinton was +3 in the earliest poll up right now....conducted 4/17 through 4/20...and is +3 in the latest one...conducted entirely on 5/10. In between there were a couple of Clinton spikes to 11.. in a poll conducted 4/20 through 4/24...and to 13... in a poll conducted 4/28 through 5/1.

To list it out, it's like this for Clinton in the set of polls listed right now for Clinton:

+3, +7, +11, -2, +13, +6, +2.

Things could change but right now there's not a lot of evidence of any trend one way or another. Just an overall impression that Clinton is ahead by a modest margin and a lot of variability in the results of different polls.

One thing that gives me hope, though:

The polls that gave Clinton +11 and +13 (USA Today and CNN) are both rated A- by the Poll ratings at http://fivethirtyeight.com/interactives ... r-ratings/. And those ratings were constructed well before the current situation. The other four polls are rated C+, B, C, B-, and C.

In other words, the two most highly rated polls according to that site are the two that showed Clinton up by double digits. At least right now:

How accurate have the polls been during the present campaign? I seem to remember Bernie exceeding them a few times.

[TheDancinMonarch]

With 0% of precincts reporting!

Trump is within striking distance in most polls (RCP average is Clinton + 5.7 and falling). Clinton fatigue will continue to deepen. Seriously, how many Clinton fanatics are out there? I'm not talking about the ones standing behind her, paid to hold up pre-printed signs at less than well attended rallies. How many Hillary supporters do you know who REALLY like her and all that she's about?

Two YUGE caveats...unless Clinton gets indicted or significantly damaged by the investigation, or an indie candidate throws their hat in the ring...The Donald will be our next president.

You heard it here first!

KalmNet - The Most Trusted Name in Fair and Balanced Leaning Forward News.

Oddly enough I agree with you. It's still a long shot but not outside the realm of the possible. And "Clinton Fatigue" may well be the most important factor. That woman may be the most unlikeable candidate put up by either party since I first voted in 1968. Either way Costa Rica will be looking pretty good in 2017. Yo hablo español.

[houndawg]

With 0% of precincts reporting!

Trump is within striking distance in most polls (RCP average is Clinton + 5.7 and falling). Clinton fatigue will continue to deepen. Seriously, how many Clinton fanatics are out there? I'm not talking about the ones standing behind her, paid to hold up pre-printed signs at less than well attended rallies. How many Hillary supporters do you know who REALLY like her and all that she's about?

Two YUGE caveats...unless Clinton gets indicted or significantly damaged by the investigation, or an indie candidate throws their hat in the ring...The Donald will be our next president.

You heard it here first!

KalmNet - The Most Trusted Name in Fair and Balanced Leaning Forward News.

Oddly enough I agree with you. It's still a long shot but not outside the realm of the possible. And "Clinton Fatigue" may well be the most important factor. That woman may be the most unlikeable candidate put up by either party since I first voted in 1968. Either way Costa Rica will be looking pretty good in 2017. Yo hablo español.

I'm just back from a week in Puerto Vallarta and I see why there are more Americans moving to Mexico than there are Mexicans moving to the US. You have not had huevos rancheros or carnitas or chile verde until you have had them from Señorita Olivia. Who assures me that in 60 days my limited espanol will sound just like the locals. Mighty tempting.

[houndawg]

Oddly enough I agree with you. It's still a long shot but not outside the realm of the possible. And "Clinton Fatigue" may well be the most important factor. That woman may be the most unlikeable candidate put up by either party since I first voted in 1968. Either way Costa Rica will be looking pretty good in 2017. Yo hablo español.

I'm just back from a week in Puerto Vallarta and I see why there are more Americans moving to Mexico than there are Mexicans moving to the US. You have not had huevos rancheros or carnitas or chile verde until you have had them from Señorita Olivia. Who assures me that in 60 days my limited espanol will sound just like the locals. Mighty tempting.

And I still think it is too early to count Bernie out. Once the donk establishment comes to its senses it will figure out that its better to win with Bernie than to lose with whats her name.

[JohnStOnge]

How accurate have the polls been during the present campaign? I seem to remember Bernie exceeding them a few times.

The only race I think was a surprise as far as what the polls where saying vs. who won was Sanders winning Michigan.

Remember this: When you see that "margin of error" with polls that's usually a 95 percent confidence interval. If all things are done perfectly the truth will be outside of the 95 percent confidence interval 1 in 20 times. And things can't be done perfectly. Bottom line is that you should EXPECT results outside of the margin of error to occur from time to time.

I think pollsters do a remarkable job. I do consider what they produce to provide useful insight. But you can't do stuff like act like you KNOW there's been some kind of real change because one poll this week says Candidate B is up by 4 points while another polls last week said Candidate A was up 4 points. You have to bear the "fuzziness" of polling in mind. There's a limit to the resolution.

[JohnStOnge]

How accurate have the polls been during the present campaign? I seem to remember Bernie exceeding them a few times.

A good one to watch is today's Oregon primary. I've been seeing talking heads on the news opining that Sanders that it's made to order for Sanders. They seem to expect Sanders to win. But the latest Oregon poll at Real Clear Politics, which was taken May 6 through 9, estimates Clinton up by 15. Marin of error reported at 5.6. As I noted I think those margins of error understate things because they only represent random sampling error and there are other sources of error.

Even though that's just one poll and it's a little over a week old I'd look at it and expect Clinton to win. But, as noted, the pundits I've been seeing on TV seem to expect Sanders to win. So maybe it'll be another Michigan situation.

Couldn't find a poll on Kentucky at the Real Clear Politics site.

[kalm]

How accurate have the polls been during the present campaign? I seem to remember Bernie exceeding them a few times.

The only race I think was a surprise as far as what the polls where saying vs. who won was Sanders winning Michigan.

Remember this: When you see that "margin of error" with polls that's usually a 95 percent confidence interval. If all things are done perfectly the truth will be outside of the 95 percent confidence interval 1 in 20 times. And things can't be done perfectly. Bottom line is that you should EXPECT results outside of the margin of error to occur from time to time.

I think pollsters do a remarkable job. I do consider what they produce to provide useful insight. But you can't do stuff like act like you KNOW there's been some kind of real change because one poll this week says Candidate B is up by 4 points while another polls last week said Candidate A was up 4 points. You have to bear the "fuzziness" of polling in mind. There's a limit to the resolution.

I just looked up Wisconsin, and both Sanders and Cruz significantly out-performed polls. I think this happened elsewhere.

It's a wacky year, but even historically there have been big swings. IIRC, both Carter and Dukakis had large leads at one point.

[Ibanez]

Well, I hope you're wrong Kalm. Hopefully the Democrats will be successful in exploiting the target rich environment that is Trump's background of behavior and statements.

Otherwise: I don't think there is a "significant" falling trend for Clinton in polling at this point. You've had stuff like Trump up by 2 then Clinton up by 13 in two consecutive polls in the RCP average. There's a lot of variation. I did a trend thing for the poll results up right now and there's no significant trend. But you can see that by just noting that Clinton was +3 in the earliest poll up right now....conducted 4/17 through 4/20...and is +3 in the latest one...conducted entirely on 5/10. In between there were a couple of Clinton spikes to 11.. in a poll conducted 4/20 through 4/24...and to 13... in a poll conducted 4/28 through 5/1.

To list it out, it's like this for Clinton in the set of polls listed right now for Clinton:

+3, +7, +11, -2, +13, +6, +2.

Things could change but right now there's not a lot of evidence of any trend one way or another. Just an overall impression that Clinton is ahead by a modest margin and a lot of variability in the results of different polls.

One thing that gives me hope, though:

The polls that gave Clinton +11 and +13 (USA Today and CNN) are both rated A- by the Poll ratings at http://fivethirtyeight.com/interactives ... r-ratings/. And those ratings were constructed well before the current situation. The other four polls are rated C+, B, C, B-, and C.

In other words, the two most highly rated polls according to that site are the two that showed Clinton up by double digits. At least right now:

What's the margin of error? +/-3?

[89Hen]

Nope. Trump needs to flip 62 electoral votes from 2012 and win every state Romney did to win. Do you realize even if Trump somehow manages to flip Ohio, Pennsylvania, and Florida he will still need Arizona (where Hillary is winning in current polls) to win? And Arizona could very well go blue as Arizona has quite a large number of Hispanic voters. Not to mention (at least for now) that Georgia is in reach.

What does 2012 have to do with this? All bets are off this year. Look at the maps from 1988 >1992 and again from 1996 > 2000.

[JohnStOnge]

One thing that gives me hope, though:

The polls that gave Clinton +11 and +13 (USA Today and CNN) are both rated A- by the Poll ratings at http://fivethirtyeight.com/interactives ... r-ratings/. And those ratings were constructed well before the current situation. The other four polls are rated C+, B, C, B-, and C.

In other words, the two most highly rated polls according to that site are the two that showed Clinton up by double digits. At least right now:

What's the margin of error? +/-3?

You can see the "margin of error" for each poll in the "MoE" column of the Real Clear Politics table. However, I do think the error in terms of the spread is understated. Using the Gravis poll from the table above as an example:

As I wrote in another thread, the Gravis poll is reported as having a +/- 2.5 percentage points margin of error. But when you talk about the margin between Clinton and Trump, there is additional uncertainty because there are really three responses. Clinton, Sanders, and Other. The actual margin of error for Clinton - Trump is about +/- 5 percentage points. So you'd say you're 95 percent confident it's somewhere in the range Clinton +7 through Trump +3.

And THAT'S just counting random sampling error. That's what the 95% confidence interval would be in an ideal situation where a perfect, theoretically sound probability sample was collected. That is not possible in polling. Pollsters make assumptions and craft models to make up for that. And each polling company has its own thing. End result is that you see stuff like a Rasumussen poll conducted 4/27 - 4/28 yield Trump +2 then have a CNN/ORC poll conducted 4/28 - 5/1 yield Clinton +13. I think anybody with common sense realizes there is no way the true situation flipped by 15 percentage points or anything close to that between 4/27 - 4/28 and 4/28 - 5/1. Rasumussen polled likely voters while CNN/ORC polled registered voters but, still, there's no way that kind of difference was the reality.

[Pwns]

Nope. Trump needs to flip 62 electoral votes from 2012 and win every state Romney did to win. Do you realize even if Trump somehow manages to flip Ohio, Pennsylvania, and Florida he will still need Arizona (where Hillary is winning in current polls) to win? And Arizona could very well go blue as Arizona has quite a large number of Hispanic voters. Not to mention (at least for now) that Georgia is in reach.

What does 2012 have to do with this? All bets are off this year. Look at the maps from 1988 >1992 and again from 1996 > 2000.

I'm just using 2012 as a baseline. Most of the red states from the last few elections will be red and most of the blue ones blue.

[Ibanez]

What's the margin of error? +/-3?

You can see the "margin of error" for each poll in the "MoE" column of the Real Clear Politics table. However, I do think the error in terms of the spread is understated. Using the Gravis poll from the table above as an example:

As I wrote in another thread, the Gravis poll is reported as having a +/- 2.5 percentage points margin of error. But when you talk about the margin between Clinton and Trump, there is additional uncertainty because there are really three responses. Clinton, Sanders, and Other. The actual margin of error for Clinton - Trump is about +/- 5 percentage points. So you'd say you're 95 percent confident it's somewhere in the range Clinton +7 through Trump +3.

And THAT'S just counting random sampling error. That's what the 95% confidence interval would be in an ideal situation where a perfect, theoretically sound probability sample was collected. That is not possible in polling. Pollsters make assumptions and craft models to make up for that. And each polling company has its own thing. End result is that you see stuff like a Rasumussen poll conducted 4/27 - 4/28 yield Trump +2 then have a CNN/ORC poll conducted 4/28 - 5/1 yield Clinton +13. I think anybody with common sense realizes there is no way the true situation flipped by 15 percentage points or anything close to that between 4/27 - 4/28 and 4/28 - 5/1. Rasumussen polled likely voters while CNN/ORC polled registered voters but, still, there's no way that kind of difference was the reality.

I couldn't see that on my phone.


So, with the MoE taken into account, the polls tell me that these two are neck and neck.

[houndawg]

You can see the "margin of error" for each poll in the "MoE" column of the Real Clear Politics table. However, I do think the error in terms of the spread is understated. Using the Gravis poll from the table above as an example:

As I wrote in another thread, the Gravis poll is reported as having a +/- 2.5 percentage points margin of error. But when you talk about the margin between Clinton and Trump, there is additional uncertainty because there are really three responses. Clinton, Sanders, and Other. The actual margin of error for Clinton - Trump is about +/- 5 percentage points. So you'd say you're 95 percent confident it's somewhere in the range Clinton +7 through Trump +3.

And THAT'S just counting random sampling error. That's what the 95% confidence interval would be in an ideal situation where a perfect, theoretically sound probability sample was collected. That is not possible in polling. Pollsters make assumptions and craft models to make up for that. And each polling company has its own thing. End result is that you see stuff like a Rasumussen poll conducted 4/27 - 4/28 yield Trump +2 then have a CNN/ORC poll conducted 4/28 - 5/1 yield Clinton +13. I think anybody with common sense realizes there is no way the true situation flipped by 15 percentage points or anything close to that between 4/27 - 4/28 and 4/28 - 5/1. Rasumussen polled likely voters while CNN/ORC polled registered voters but, still, there's no way that kind of difference was the reality.

I couldn't see that on my phone.


So, with the MoE taken into account, the polls tell me that these two are neck and neck.

Yes but you're spoiling John's schtick when you say it that way.

[JohnStOnge]

The only race I think was a surprise as far as what the polls where saying vs. who won was Sanders winning Michigan.

Remember this: When you see that "margin of error" with polls that's usually a 95 percent confidence interval. If all things are done perfectly the truth will be outside of the 95 percent confidence interval 1 in 20 times. And things can't be done perfectly. Bottom line is that you should EXPECT results outside of the margin of error to occur from time to time.

I think pollsters do a remarkable job. I do consider what they produce to provide useful insight. But you can't do stuff like act like you KNOW there's been some kind of real change because one poll this week says Candidate B is up by 4 points while another polls last week said Candidate A was up 4 points. You have to bear the "fuzziness" of polling in mind. There's a limit to the resolution.

I just looked up Wisconsin, and both Sanders and Cruz significantly out-performed polls.

Yes but what I'm saying is that Sanders winning Michigan is the only one that was a surprise in terms of what the polls were saying and who won. In other words you're sitting there watching the polls and feeling confident that one candidate is going to win then the OTHER one does. The polls did under estimate the margins for Sanders and Cruz in Wisconsin but Sanders and Cruz WERE both ahead in the polls so the people who were ahead in the polls did win. Of course in both cases the polls were close enough so that it wouldn't have been a big surprise to see both races go the other way as well.

[JohnStOnge]

Here's an interesting article at FiveThirtyEight:

http://fivethirtyeight.com/features/wha ... nd-oregon/

When it comes to Oregon they are putting more stock in demographic assessment than they are in polls:

Oregon is different. Nate’s demographic model gives Sanders an edge of about 15 percentage points. That’s because whites made up about 90 percent of Obama voters in the 2008 general election. Keep in mind too that Sanders won next door in the Washington caucuses in March by about 45 percentage points. Clinton is expected to do better in Oregon because, unlike Washington, Oregon is a primary and is closed to non-Democrats. I should note that the only two polls taken this year, including one taken this month, have shown Clinton ahead, so it’s possible that she’ll pull it out.

So basically you have a demographic model saying Sanders by 15 vs. a poll taken during a period spanning 8 through 11 days ago saying Clinton by 15 and they're going with the Demographic model. And I must say that I'm watching CNN right now and all the talking heads are talking about Sanders winning Oregon as though it's a done deal.

[89Hen]

What does 2012 have to do with this? All bets are off this year. Look at the maps from 1988 >1992 and again from 1996 > 2000.

I'm just using 2012 as a baseline. Most of the red states from the last few elections will be red and most of the blue ones blue.

Point being, Trump could flip some states that haven't flipped in a while.

[∞∞∞]

I'm just using 2012 as a baseline. Most of the red states from the last few elections will be red and most of the blue ones blue.

Point being, Trump could flip some states that haven't flipped in a while.

I think the same can be said about whoever the Democratic nominee is.

[89Hen]

Point being, Trump could flip some states that haven't flipped in a while.

I think the same can be said about whoever the Democratic nominee is.

Sure, but pwns was pretty certain in one direction only.