Battlegrounds

[AUDub 11,146]

2 hours ago, McLoofus said:

Saw that. Seems like a perfectly executed hedging of bets- he even managed to comment on Trump's trumpiness without calling it a good thing or a bad thing- but there is some good background info.

I'm just not sure about his methodology. Polling is still the best way to predict an election, and insofar his model doesn't jive with them at all.

[McLoofus 35,182]

31 minutes ago, AUDub said:

I'm just not sure about his methodology. Polling is still the best way to predict an election, and insofar his model doesn't jive with them at all.

Yeah, no idea what his methodology is. I just know he was right last time, and early. And, evidently, he's almost always been right. Which in most cycles might not be hard, but the combination suggests he was neither just a blind squirrel in 2016 nor a blind follower before then.

Look, throwing my chips in with this guy means either I get to be right or something far better than that. Reverse jinxes usually treat me pretty well when I really commit to them. Like, Andy Kaufman commit.

Helmut Norpoth promise a trump win this November and Helmut Norpoth doesn't miss.

[AUDub 11,146]

Just now, McLoofus said:

Yeah, no idea what his methodology is. I just know he was right last time, and early. And, evidently, he's almost always been right. Which in most cycles might not be hard, but the combination suggests he was neither just a blind squirrel in 2016 nor a blind follower before then.

Look, throwing my chips in with this guy means either I get to be right or something far better than that. Reverse jinxes usually treat me pretty well when I really commit to them. Like, Andy Kaufman commit.

Helmut Norpoth promise a trump win this November and Helmut Norpoth doesn't miss.

We're not talking about a large sample size here, though. Probably dozens of these poli-sci guys out there with a "model" and he's simply one of the few left standing. I understand he hindcast his model and gotten everything correct back to 1912 but that's not really indicative of, well, anything (and automatically stinks of bull****, given the sparseness of reliable data before the modern era) given we're talking about probabilistic modeling. 

[Gowebb11 9,313]

Great discussion, but if you’re Biden you should campaign like you’re 10 points down in every state, from now until November. Recent elections suggest trusting the polls is risky. 

[IronMan70 3,277]

On 5/20/2020 at 6:30 PM, AUDub said:

Take it for what it is and remember we're still 6 months out. 

 

I was really curious about this QU poll so I dug down on the methodology. It is a perfect example of what I was saying earlier about finding out the sampling before drawing any conclusions. This particular poll is a poll of registered voters (not likely voters) but the worst part is the sampling. This poll has a Democrat +10 sampling. That makes this poll practically useless to draw valid conclusions. The worst polls are of adults (A), slighty better is registered voters (RV), with the best being likely voters (LV). Your 538 site breaks it out for you. But even with LVs you have to double check the sampling. The skewing effect in the results of a poll start to show up at a +5. A double digit skew of +10 like this one is up there.   

 https://poll.qu.edu/images/polling/us/us05202020_demos_ugjm33.pdf 

"This RDD telephone survey was conducted from May14–18, 2020 throughout the nation. Responses are reported for 1,323 self-identified registered voters with a margin of sampling error of +/-2.7percentage points. Margins of sampling error for subgroups are available upon request. Surveys were conducted with live interviewers calling landlines and cell phones. Data collection support provided by Dynata. All data was managed and tabulated by the Quinnipiac University Poll."

PARTY IDENTIFICATION QUESTION WORDING -Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?

REGISTERED VOTERS PARTY IDENTIFICATION

Republican 26%

Democrat 36%

Independent 31%

Other/DK/NA 7%

[AUDub 11,146]

10 minutes ago, IronMan70 said:

I was really curious about this QU poll so I dug down on the methodology. It is a perfect example of what I was saying earlier about finding out the sampling before drawing any conclusions. This particular poll is a poll of registered voters (not likely voters) but the worse part is the sampling for this poll is Democrat +10. That makes this poll practically useless to draw valid conclusions from. The worst polls are groupings of adults (A), then registered voters (RV), with the best being likely voters (LV) but even with LVs you have to double check the sampling. The skewing effect in the results of a poll start to show at a +5. A double digit skew like this one of +10 is up there.   

 https://poll.qu.edu/images/polling/us/us05202020_demos_ugjm33.pdf 

"This RDD telephone survey was conducted from May14–18, 2020 throughout the nation. Responses are reported for 1,323 self-identified registered voters with a margin of sampling error of +/-2.7percentage points. Margins of sampling error for subgroups are available upon request. Surveys were conducted with live interviewers calling landlines and cell phones. Data collection support provided by Dynata. All data was managed and tabulated by the Quinnipiac University Poll."

PARTY IDENTIFICATION QUESTION WORDING -Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?

REGISTERED VOTERS PARTY IDENTIFICATION

Republican 26%

Democrat 36%

Independent 31%

Other/DK/NA 7%

Not really familiar with polling methodology, are you? Those numbers are actually pretty pedestrian.

[AUDub 11,146]

Just now, IronMan70 said:

I am. 

Then you'd be aware that Quinnipiac isn't necessarily trash. They are a reputable firm. 

Ds always outnumber Rs on these. Don't know what the final result will be as yet, but 11 points is a reliable indicator that Biden is polling extremely well from them. 

[IronMan70 3,277]

6 minutes ago, AUDub said:

Not really familiar with polling methodology, are you? Those numbers are actually pretty pedestrian.

I am actually, a double digit bias is up there. The worst I have ever seen was a D+13 and that was in a R+5 actual voting district. Polls like that are push polls. Pollsters do stuff like this all the time.

[AUDub 11,146]

9 minutes ago, IronMan70 said:

I am actually, a double digit bias is up there. The worst I have ever seen was a D+13 and that was in a R+5 actual voting district. Polls like that are push polls. Pollsters do stuff like this all the time.

No that's not true. Their mean reverted bias is pro-Dem, mind you.

 

 

 

 

 

 

 

 

 

 

 

 

Pro-Dem +.02

[IronMan70 3,277]

2 minutes ago, AUDub said:

Then you'd be aware that Quinnipiac isn't necessarily trash. They are a reputable firm. 

Ds always outnumber Rs on these. Don't know what the final result will be as yet, but 11 points is a reliable indicator that Biden is polling extremely well from them. 

I wouldn't say they are trash. I would say that this poll is a bad one. If you a get a Biden +11 but you have to use a D+10 to get there what does that tell you. Other than you have a 10 point bias going in, it doesn't tell you much.

[IronMan70 3,277]

5 minutes ago, AUDub said:

Pro-Dem +.02

Gotchya, lol.

[AUDub 11,146]

1 minute ago, IronMan70 said:

HAHA, ok.

That's what their respndents are telling them, and it's a methodologically sound poll. Quinnipiac isn't known for posting trash polls. Granted, its only one poll very early in the game, so, again,  take it with a grain of salt, but don't question their methods. 

[AUDub 11,146]

6 minutes ago, IronMan70 said:

I wouldn't say they are trash. I would say that this poll is a bad one. If you a get a Biden +11 but you have to use a D+10 to get there what does that tell you. Other than you have a 10 point bias going in, it doesn't tell you much.

It could be. Anomalous results aren't rare.

But you'll find a lot of these polls, even those that favor Trump,  will have a similar distribution. Nature of the beast. 

Again, grain of salt. 

[IronMan70 3,277]

1 minute ago, AUDub said:

That's what their respndents are telling them, and it's a methodologically sound poll. Quinnipiac isn't known for posting trash polls. Granted, its only one poll very early in the game, so, again,  take it with a grain of salt, but don't question their methods. 

I am questioning a D+10 sampling methodology and you should too. Take the same poll and give it a sampling of your D+2 example, then make it a "likely voter" poll and see what you would get. Pollsters make a lot of money from their customers to publish their polls. Some of the esoteric sampling you will see out there is amazing. It's hard to get too excited about them.

[AUDub 11,146]

2 hours ago, IronMan70 said:

I am questioning a D+10 sampling methodology and you should too.

Again, not uncommon. Even in polls that favor Trump,  there will be a Democrat lean. 

2 hours ago, IronMan70 said:

Take the same poll and give it a sampling of your D+2 example, then make it a "likely voter" poll and see what you would get. Pollsters make a lot of money from their customers to publish their polls. Some of the esoteric sampling you will see out there is amazing. It's hard to get too excited about them.

Reputable posters have a stake in the game in not ******* it up, either. Reputation drops with obvious bias. It's something they seek to avoid. 

[AuburnNTexas 7,139]

At this point the polls are not very reliable because of Covid-19 people are afraid and the economy is in the tank. As we move through the summer and well into the fall if Covid-19 comes back big and the economy is not showing any signs of returning Trump loses big As would any incumbent President. If in the same time frame we have better treatments for Covid-19, numbers of deaths have become much smaller per month, and the economy is picking up steam Trump will have a very good chance of winning.

Covid-19 has turned the whole world on edge . Countries that the far left has said we should take after on health care in Europe have on the whole done worse than the US when you look at Death's per million of population. Countries with low population densities have done well and countries with high population densities have not done well.  (China and India are exceptions on density but their numbers have both been questioned for accuracy). We see a Microcosm of this in the US when looking at our states. Even in the states you see a microcosm of this when you look at Upstate New York versus New York City. 

One thing that I seriously doubt you will see in the Media is the difference in deaths per million in Red States versus Blue States part of that is due to population density but some of it is due to preparedness at state and local level. Of the top 12 states including District of Columbia death's per million 11 are Blue states and one is a Red State Louisiana which currently has a Democratic Governor and Democratic Mayor of New Orleans. To be honest the Democratic Governor of Louisiana would be a Republican in any other state. The fact that the National media glosses over this helps the Democrats and hurts the Republican's. Before anybody brings up the mayor of New Orleans blames Trump for him keeping Mardi Gras open during the pandemic as he believed Trump where previously he has said he doesn't believe anything Trump says. Florida, Texas, and Georgia all have larger populations than most of the Red states. To be fair it is not across the board some Blue States like California and Washington have done a very good job.

Polls starting around middle of September and after will be a much better barometer of who will win the Presidency. It will also be interesting to see if some of the traditional Blue States who have not handled the wellness side of Covid-19 or the economic side of Covid-19 very well will impact elections at the local and state levels.  It would be hard to think that people would think Cuomo and De Blasio have done a good job during this pandemic with Cuomo mandating the people with Covid-19 go into nursing homes and De Blasio initially saying Covid-19 was over blown even after Trump had started to change his tune.

[McLoofus 35,182]

16 minutes ago, AuburnNTexas said:

Countries with low population densities have done well and countries with high population densities have not done well.  (China and India are exceptions on density but their numbers have both been questioned for accuracy). We see a Microcosm of this in the US when looking at our states. Even in the states you see a microcosm of this when you look at Upstate New York versus New York City. 

One thing that I seriously doubt you will see in the Media is the difference in deaths per million in Red States versus Blue States part of that is due to population density but some of it is due to preparedness at state and local level. Of the top 12 states including District of Columbia death's per million 11 are Blue states and one is a Red State Louisiana which currently has a Democratic Governor and Democratic Mayor of New Orleans. To be honest the Democratic Governor of Louisiana would be a Republican in any other state. The fact that the National media glosses over this helps the Democrats and hurts the Republican's.

Your 1st paragraph submarines your 2nd.  (I mostly agree with the 1st.)

Also, why cut it off at 12? Why not count every state in your graphic? Maybe because that adds 4 GOP governors and only 1 Dem, with 3 of the GOP govs (and the Dem) falling outside the top 20 states by population density?

 

[AuburnNTexas 7,139]

8 minutes ago, McLoofus said:

Your 1st paragraph submarines your 2nd.  (I mostly agree with the 1st.)

Also, why cut it off at 12? Why not count every state in your graphic? Maybe because that adds 4 GOP governors and only 1 Dem, with 3 of the GOP govs (and the Dem) falling outside the top 20 states by population density?

 

As to why I stopped at 12 it was nothing nefarious when I did cut and paste that is how many fit on the screen.  For those who want to see the entire picture here is the actual link.

 https://www.worldometers.info/coronavirus/country/us/

Please sort by deaths by million.

As to my second statement undermining my first to am extent you are correct but the reality is like all statistics you have to look at the whole picture in all the states it is large urban areas that are impacted. All the Red states that have done well have large urban areas in Texas you have DFW, Houston, Austin, San Antonio, Georgia has Atlanta,  Florida has multiple large cities Miami and whole South Florida, Tamps, Orlando so they have heavily populated areas without the same impact. In my previous statement I mentioned the difference between New York City and Upstate New York. It is true Texas has low Population Density because so many areas in Texas with almost no people but if you have ever driven in DFW, Austin or Houston during rush hour their are a lot of people in those areas.

It will take years and lots of analyses for us to truly understand The W's (Who, What, Why, Where) of the Covid-19 in the US and the world. Some things are very clear nursing/assisted living homes have been impacted as you have people confined together in tight spaces with many pre-existing conditions. The impact has been different from State to State and a lot of that is because of policies put in place by the Governors of each state. In Texas our Governor Abbott said anybody testing positive with Covid-19 would not be allowed in theses homes in New York their Governor Cuomo required the homes to take people who tested positive. New York just recently changed that and at the same time to slow the firestorm of this inept policy that changed the way they counted death's in nursing homes. Prior to the change If a person contracted Covid-19 in the nursing home but when they became very sick and were moved to a hospital they counted it as a nursing home death they now count is as dying in a hospital Technically correct as they did die in a hospital but it hides the fact where they contracted it. Another key factor is the age of populations in each State and City as overwhelmingly I believe I read 80% or 85% of the deaths in the US were people over 65. I don't honestly know if the Blue States that have not done well have a higher concentration of people in my age bracket, It would be hard for me to believe many States have a higher concentration of old people than Florida which over the years has been the retirement destination of large numbers of people from the North East to escape the cold.

Before anybody says it I will concede as cities go in the US New York City probably is the densest as millions of people in a small area as they go very vertical. Most of the other large cities in the US are similar and less like New York with a few exceptions.

[McLoofus 35,182]

1 hour ago, AuburnNTexas said:

As to why I stopped at 12 it was nothing nefarious when I did cut and paste that is how many fit on the screen.  For those who want to see the entire picture here is the actual link.

 https://www.worldometers.info/coronavirus/country/us/

Please sort by deaths by million.

As to my second statement undermining my first to am extent you are correct but the reality is like all statistics you have to look at the whole picture in all the states it is large urban areas that are impacted. All the Red states that have done well have large urban areas in Texas you have DFW, Houston, Austin, San Antonio, Georgia has Atlanta,  Florida has multiple large cities Miami and whole South Florida, Tamps, Orlando so they have heavily populated areas without the same impact. In my previous statement I mentioned the difference between New York City and Upstate New York. It is true Texas has low Population Density because so many areas in Texas with almost no people but if you have ever driven in DFW, Austin or Houston during rush hour their are a lot of people in those areas.

It will take years and lots of analyses for us to truly understand The W's (Who, What, Why, Where) of the Covid-19 in the US and the world. Some things are very clear nursing/assisted living homes have been impacted as you have people confined together in tight spaces with many pre-existing conditions. The impact has been different from State to State and a lot of that is because of policies put in place by the Governors of each state. In Texas our Governor Abbott said anybody testing positive with Covid-19 would not be allowed in theses homes in New York their Governor Cuomo required the homes to take people who tested positive. New York just recently changed that and at the same time to slow the firestorm of this inept policy that changed the way they counted death's in nursing homes. Prior to the change If a person contracted Covid-19 in the nursing home but when they became very sick and were moved to a hospital they counted it as a nursing home death they now count is as dying in a hospital Technically correct as they did die in a hospital but it hides the fact where they contracted it. Another key factor is the age of populations in each State and City as overwhelmingly I believe I read 80% or 85% of the deaths in the US were people over 65. I don't honestly know if the Blue States that have not done well have a higher concentration of people in my age bracket, It would be hard for me to believe many States have a higher concentration of old people than Florida which over the years has been the retirement destination of large numbers of people from the North East to escape the cold.

Before anybody says it I will concede as cities go in the US New York City probably is the densest as millions of people in a small area as they go very vertical. Most of the other large cities in the US are similar and less like New York with a few exceptions.

You mentioned blue and red governors. At the state level, red governors mostly preside over states with lower density. The fact that they have large individual metros doesn't change that. And no, Atlanta and Houston are not even remotely like New York in terms of density.

Also, a certain group keep bringing up Cuomo as though he represents every blue state. Y'all should probably find more examples of bad policy.

[McLoofus 35,182]

Oh, and as far as we know, heat and humidity dampen the virus significantly. So yeah, Florida, the deep South and Texas might have hit the trifecta of having the virus show up later in less densely populated states in warmer weather.

[AuburnNTexas 7,139]

17 hours ago, McLoofus said:

You mentioned blue and red governors. At the state level, red governors mostly preside over states with lower density. The fact that they have large individual metros doesn't change that. And no, Atlanta and Houston are not even remotely like New York in terms of density.

Also, a certain group keep bringing up Cuomo as though he represents every blue state. Y'all should probably find more examples of bad policy.

Please read what I said when you respond I conceded that New York was the exception. Here is what I actually said in the Post you responded to.

"Before anybody says it I will concede as cities go in the US New York City probably is the densest as millions of people in a small area as they go very vertical. Most of the other large cities in the US are similar and less like New York with a few exceptions.."

I was comparing the Southern cities to Red State cities like Detroit with less than 700,000 people, Boston less than 700,000 greater Boston 4.8 million when comparing to red state cities Dallas with over 1.3 million people the DFW Metro area population 7.5 million, Houston 2.3 million, Houston Metroplex almost 7 Million. I mention New York Governor as the liberal media seems to think he can do no wrong despite some really poor decisions he has made. The Governor of New Jersey had a similar plan for nursing homes as New York and similar results. Governor Whitmer up until last few days thinks that one size fits all when making decisions on opening state. A lot of Northern Michigan has had minimal impact from Covid-19 because of low density but she treated them the same as Detroit.

I try to be open minded in the very first post I mentioned how well California and Washington State have done. I stand by my original statement about density as the Wyoming and Dakota type states have been doing social distancing as a way of life as lots of land and few people. I also think you can't just look at density per state but really must look at it where States have large population centers to give a truly accurate view.

[AuburnNTexas 7,139]

17 hours ago, McLoofus said:

Oh, and as far as we know, heat and humidity dampen the virus significantly. So yeah, Florida, the deep South and Texas might have hit the trifecta of having the virus show up later in less densely populated states in warmer weather.

It is funny when Trump mentioned heat and humidity many on this board scoffed at it because Trump said it.  I don't believe you were one of them. There is solid science behind the theory that heat and humidity might ameliorate the affects based on the fact that it has been seen to ameliorate the impact of other Corona type viruses like the flu, common cold, SARS, and MERS. That said with major busy International Airports in those Red states they would have had similar contacts with people bringing in Covid-19 as the Blue States when this started in February and March well before these states had high temperatures. Also heat and humidity would not explain New Orleans which is hotter and more humid than Dallas. 

While I am partially discounting the heat and humidity as the reason the Southern states were not hit as hard in the beginning as I said in an earlier post we will be collecting and analyzing data for years as there are multiple factors that could affect the intensity of the Covid-19 pandemic both worldwide, statewide, and locally. Italy and Spain were particularly hard hit and one reason is cultural there still are large number of extended families (Multi-generational) living in same house or very close and that increases the odds of older people catching it and dying.

[AUDub 11,146]

Fox News most recent poll. Don't cringe at the word Fox. Their polling is editorially independent and generally good.

Biden +8. 

https://static.foxnews.com/foxnews.com/content/uploads/2020/05/7ab2e761-Fox_May-17-20-2020_Complete_National_Topline_May-21-Release.pdf

[creed 1,635]

All these polls that I see being discussed here, unless I’m mistaken, are popular vote polls. Can someone show me an an electoral college based poll then I’ll pay more attention. 

[AUDub 11,146]

17 hours ago, creed said:

All these polls that I see being discussed here, unless I’m mistaken, are popular vote polls. Can someone show me an an electoral college based poll then I’ll pay more attention. 

That would be state level polling.

But there is a correlation between the popular vote and the EC. If the margin ends up being what polling indicates right now (RCP average sits at 5.5), we're looking at a pretty comfortable Biden win.

But, again, super early. Everything with a grain of salt.