Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature currently requires accessing the site using the built-in Safari browser.
By 2 points according to RCP. Big deal.
And the RCP AVERAGE incorporates the errors implicit in the polling. So if one poll uses registered voters while another uses likely voters, the inaccuracy of the former still gets averaged in to the latter.
And if SOME polls still persist in going with respondents weighted with a Democrat voter overload (unintentionally or intentionally doesn't matter) then the skewing of those polls' results will ALSO be "incorporated into the "average" outcome.
Ohio and PA and Florida and VA are going to go with Romney.
Watch.
Ohio, quite possibly.
Florida, I think that's a lock at this point.
PA... as much as I'd like to see that, since I live in PA, I am not sure Romney will take PA. RCp has Obama by 4.8 right now, and I don't think that is a number that can be overcome here in PA.
PA will go Obama, Florida will go Romney, and OH will decide the election for Romney.
Q9 If you are a Democrat, press 1. If a Republican,
press 2. If you are an independent or identify
with another party, press 3.
Democrat ........................................................ 48%
Republican...................................................... 37%
Independent/Other.......................................... 15%
It's tied.
It's tied.
Not really...
Romney will take OH... Probably by 2-3 points.. Maybe more...
Turnout in early voting is WAY down from 2008... Early voting is what gave 0bozo OH in 2008, as McCain won the election day vote totals...
People just aren't excited about the empty chair anymore and will likely sit this one out...
Nice analysis, L. The question, I don't think now, is whether Romney can win: I believe he will. The worry is a 53 or 54 Democratic majority in the Senate. If that happens, watch the lame duck season be a wild one as Obama and the Dems push the GOP for everything they can get, with the threat the GOP will get nothing after Romney is inaugurated if they don't deal during the lame duck.
It turns out that (possibly) there is no intentional or inadvertent Dim oversampling. The poll may have no Dim oversampling one way or the other at all (and looks like it is not guilty in this case).
So, to the extent the polling is otherwise accurate, it DOES appear that PA is leaning to President Obama. If the Dim voters are unmotivated, though, the Election Day outcome could still be an unpleasant surprise for Team Obama.
http://www.intrade.com/v4/markets/contract/?contractId=745813
I love how all liberals can point to are media polls.
Intrade - Democratic nominee to win Ohio
I love how all liberals can point to are media polls.
Nice analysis, L. The question, I don't think now, is whether Romney can win: I believe he will. The worry is a 53 or 54 Democratic majority in the Senate. If that happens, watch the lame duck season be a wild one as Obama and the Dems push the GOP for everything they can get, with the threat the GOP will get nothing after Romney is inaugurated if they don't deal during the lame duck.
I have big worries about the Senate even though I am of the belief that it will be a GOP majority. I hadn't contemplated the lame duck session prospects, but that doesn't concern me too much, since the House can block the Senate.
As for NEXT term, I foresee a LOT of Dim manipulation including but not limited to procedural choking of the process and filibusters. They can all be addressed EXCEPT, perhaps, the misuse and overuse of the filibuster. That could be a problem. Dims only object to the use of filibusters when the Dims are in the majority.
So I expect them to try it. But the Dims will then discover that there IS still a price to pay for being the MINORITY Party in both Houses.
If they are going to play chicken, I'd rather do so from the majority position.
It turns out that (possibly) there is no intentional or inadvertent Dim oversampling. The poll may have no Dim oversampling one way or the other at all (and looks like it is not guilty in this case).
So, to the extent the polling is otherwise accurate, it DOES appear that PA is leaning to President Obama. If the Dim voters are unmotivated, though, the Election Day outcome could still be an unpleasant surprise for Team Obama.
Ahhhh, but you forget that is why RV polls tend to favor Democrats more than LV polls. There are a lot more registered Democrats in the entire nation (not just PA) than Republicans. BUT Democrats don't vote with the same consistency as Republicans. There are a lot of reasons for this. Age is one. Lots of those registered Democrats are young voters who have a bad habit of not showing up to cast a ballot while a lot of the registered Republicans are senior citizens who vote religiously. In the end that difference between the total numbers that are registered equal out because of the participation rates.
Now when they change to an LV poll one of the things they are doing is applying a weighting formula to reflect expected turnout. That will vary according to the state. It stands to reason that the ratio of Dem to GOP who show up in New York will be higher than in...say...Virginia. The trick then becomes the pollsters skill in correctly predicting the LV formula that represents each state. This is also why the closer in the pollster "zooms" the more difficult it is to predict.
So the Dem:GOP ratio PPP shows may or may not be accurate because it's a LV poll and that means it is predictive and not reflective. It just depends o whether PPP's LV formula is accurate. Historically.....they are not
It turns out that (possibly) there is no intentional or inadvertent Dim oversampling. The poll may have no Dim oversampling one way or the other at all (and looks like it is not guilty in this case).
So, to the extent the polling is otherwise accurate, it DOES appear that PA is leaning to President Obama. If the Dim voters are unmotivated, though, the Election Day outcome could still be an unpleasant surprise for Team Obama.
Ahhhh, but you forget that is why RV polls tend to favor Democrats more than LV polls. There are a lot more registered Democrats in the entire nation (not just PA) than Republicans. BUT Democrats don't vote with the same consistency as Republicans. There are a lot of reasons for this. Age is one. Lots of those registered Democrats are young voters who have a bad habit of not showing up to cast a ballot while a lot of the registered Republicans are senior citizens who vote religiously. In the end that difference between the total numbers that are registered equal out because of the participation rates.
Now when they change to an LV poll one of the things they are doing is applying a weighting formula to reflect expected turnout. That will vary according to the state. It stands to reason that the ratio of Dem to GOP who show up in New York will be higher than in...say...Virginia. The trick then becomes the pollsters skill in correctly predicting the LV formula that represents each state. This is also why the closer in the pollster "zooms" the more difficult it is to predict.
So the Dem:GOP ratio PPP shows may or may not be accurate because it's a LV poll and that means it is predictive and not reflective. It just depends o whether PPP's LV formula is accurate. Historically.....they are not
I agree that a poll of likely voters is preferable and logically destined to be a better predictor of the outcome.
But the poll I was referencing USED Likely Voters.
Ahhhh, but you forget that is why RV polls tend to favor Democrats more than LV polls. There are a lot more registered Democrats in the entire nation (not just PA) than Republicans. BUT Democrats don't vote with the same consistency as Republicans. There are a lot of reasons for this. Age is one. Lots of those registered Democrats are young voters who have a bad habit of not showing up to cast a ballot while a lot of the registered Republicans are senior citizens who vote religiously. In the end that difference between the total numbers that are registered equal out because of the participation rates.
Now when they change to an LV poll one of the things they are doing is applying a weighting formula to reflect expected turnout. That will vary according to the state. It stands to reason that the ratio of Dem to GOP who show up in New York will be higher than in...say...Virginia. The trick then becomes the pollsters skill in correctly predicting the LV formula that represents each state. This is also why the closer in the pollster "zooms" the more difficult it is to predict.
So the Dem:GOP ratio PPP shows may or may not be accurate because it's a LV poll and that means it is predictive and not reflective. It just depends o whether PPP's LV formula is accurate. Historically.....they are not
I agree that a poll of likely voters is preferable and logically destined to be a better predictor of the outcome.
But the poll I was referencing USED Likely Voters.
Right...so the polling sample is somewhat irrelevant...at least it should be. Here's what I mean:
Consider a poll for state A of 100 people and according to your projections you expect a breakdown of 50% Dem and 50% showing up on election day (let's forget about Ind for now for simplicity)
So you get 100 people that meet your criteria of a likely voter depending on how they answer a series of questions and 60% of the were registered Dem and 40% are registered Rep. The Dems indicated 85% support for Obama and 15% support for Romney. The Reps indicated 90% support for Romney and 10% support for Obama.
Now if you just apply that to the 60/40 sample you got you end up with this:
60 Democrats * 85% Obama = 51 votes
60 Democrats * 15% Romney = 9 votes
50 Republicans * 90% Romney = 36 votes
50 Republicans * 10% Obama = 4 votes
Total Obama = 55
Total Romney = 45
Obama +10
BUT THAT'S NOT THE RATIO YOU EXPECT TO SHOW UP ON ELECTION DAY
So instead of applying those to a 60/40 split, you apply it to a 50/50 split that you believe will show up on election day as follows:
50 Democrats * 85% Obama = 42.5 votes
50 Democrats * 15% Romney = 7.5 votes
50 Republicans * 90% Romney = 45 votes
50 Republicans * 10% Obama = 5 votes
Total Obama = 47.5 votes
Total Romney = 52.5 votes
Despite the raw numbers showing Obama ahead based on reaching a poll sample that you believe is not reflective of who will show up on election day, your final projection is R+5
When pollsters skip that final step off applying the weighting what they really have is a modified RV poll pretending to be an LV poll. However they can achieve a different result simply by changing the weighting. PPP, for example, on average is about 4 points more in Obama's favor than other professional pollsters. My guess is that they shift the weighting more in Obama's favor than the others do
I agree that a poll of likely voters is preferable and logically destined to be a better predictor of the outcome.
But the poll I was referencing USED Likely Voters.
Right...so the polling sample is somewhat irrelevant...at least it should be. Here's what I mean:
Consider a poll for state A of 100 people and according to your projections you expect a breakdown of 50% Dem and 50% showing up on election day (let's forget about Ind for now for simplicity)
So you get 100 people that meet your criteria of a likely voter depending on how they answer a series of questions and 60% of the were registered Dem and 40% are registered Rep. The Dems indicated 85% support for Obama and 15% support for Romney. The Reps indicated 90% support for Romney and 10% support for Obama.
Now if you just apply that to the 60/40 sample you got you end up with this:
60 Democrats * 85% Obama = 51 votes
60 Democrats * 15% Romney = 9 votes
50 Republicans * 90% Romney = 36 votes
50 Republicans * 10% Obama = 4 votes
Total Obama = 55
Total Romney = 45
Obama +10
BUT THAT'S NOT THE RATIO YOU EXPECT TO SHOW UP ON ELECTION DAY
So instead of applying those to a 60/40 split, you apply it to a 50/50 split that you believe will show up on election day as follows:
50 Democrats * 85% Obama = 42.5 votes
50 Democrats * 15% Romney = 7.5 votes
50 Republicans * 90% Romney = 45 votes
50 Republicans * 10% Obama = 5 votes
Total Obama = 47.5 votes
Total Romney = 52.5 votes
Despite the raw numbers showing Obama ahead based on reaching a poll sample that you believe is not reflective of who will show up on election day, your final projection is R+5
When pollsters skip that final step off applying the weighting what they really have is a modified RV poll pretending to be an LV poll. However they can achieve a different result simply by changing the weighting. PPP, for example, on average is about 4 points more in Obama's favor than other professional pollsters. My guess is that they shift the weighting more in Obama's favor than the others do
I remember now why I never excelled in math!
Actually, that was pretty informative.
I suppose all the pollsters would be well advised to determine:
(a) whether their methodology has led to an oversampling of Dims (or Republicans, although for some reason the latter never seems to be a problem),
(b) identify the actual percentage (as well as can be determined) of Dims vs. Republicans vs. Indys in the relevant electoral population,
(c) Whether the voters sampled are or are not "Likely voters," and
(d) Crunch the numbers Properly.
I remember now why I never excelled in math!
Actually, that was pretty informative.
I suppose all the pollsters would be well advised to determine:
(a) whether their methodology has led to an oversampling of Dims (or Republicans, although for some reason the latter never seems to be a problem),
(b) identify the actual percentage (as well as can be determined) of Dims vs. Republicans vs. Indys in the relevant electoral population,
(c) Whether the voters sampled are or are not "Likely voters," and
(d) Crunch the numbers Properly.
I hope your writing is going well. And that you and yours are thriving. Glad to see you. We're doing we'll. Busy!
The location is NYC. I think we have a 100% chance of going blue.
HaHaa.. Everybody remembers that. That was earlier than 2008, it was when we were at that other place.
No, it was 2008. Fred started out with a little bit of potential, but eventually fell asleep one day in a park in Iowa, and when he awoke,
the note with his name, address, and who to call that was pinned to his jacket had fallen off, no one, including Fred, knew what to do with him,
and he was adrift in limbo for several years until the reverse mortgage people found him and nursed him back to usable condition.
After that, he made his TV Drama comeback on The Good Wife.
