Comments are closed for this entry
View All
Recency  | 
Page:  « First  ‹ Previous  1 2 3 4  Next ›  Last »  (4 total)
11:02 PM on 11/01/2012
Republicans and Democrats are very different, Democrats tend to rely on data facts and things they can see even if it is something they don't like case in point democrats reaction to the first the debate, republicans on the other hand have faith the BELIEVE. Example if the polls say you are loosing then the polls are oversampling, If something you say isn't true don't back down because you believe it then it must be true that is the basic nature of this true tribes democrats want evidence republicans have faith.
09:57 AM on 11/02/2012
The polls are over sampling Democrats it is a FACT. No one anywhere believes that Obama will have the same turn out as 08, yet that is the model most polls are using esp. in Ohio. With Romney getting Independents in double digits in every poll how in the world do Democrats expect Obama to win? So who is going on Faith and koolaide now?
02:59 PM on 11/02/2012
What is the anecdotal explanation for the aggregation of pollsters UNDER sampling the GOP? Do Republicans just not answer their phones for polls in the same numbers as self-described Independents and Democrats?

Why, over a sample size that, in the aggregate must now be in the 100s of thousands, is there a continual need to tweak the sample b/c Republicans are under represented? If I randomly called 100,000 people, I would think that the sample would be pretty well representative of the population at large, no?

This is something that I've never understood.
03:30 PM on 11/02/2012
So 96% of the polls (Those that have Obama Tied or in the lead) just have somehow magically called (by percentage) more democrats then are represented in the public. These calls are made at random, not from some voter list. I know its hard to accept that you are going to lose, but really, blame Acorn or something, this oversampling buisiness just makes you all look pretty rediculous.
Socially Liberal, fiscal Conservative
10:04 AM on 11/02/2012
Democrats won last time by believing in HOPE, not facts. The country won't make the same mistake twice.
Liberal,conservative with my money
06:10 PM on 11/02/2012
Yup, thats right we won't! No more Bush wars and no more republican oligarchy for the rich.
10:48 PM on 11/01/2012
Obama could lose Ohio and win CO, WI, NV, IA, and NH and she still wins the election. Many recent polls show him leading in ALL of those states. OH is still the safest bet with the most consistent polling. I see a DOUBLE firewall.
10:00 AM on 11/02/2012
He is not winning in CO, IA or NH WI is tied and he is winning NV. I live in CO and I am telling you Romney has all the enthusiasm, this is nothing like 08. He will not carry the suburbs around Denver in Jefferson or Douglas Counties like last election. I would bet on it.
11:49 AM on 11/02/2012
What PLANET do you live on? He's up over 2% in all of the states you mentioned, Colorado up just 1 %. You have to take the average of ALL polling for accuracy. Quit depending on your Rasmussen polls. BTW, Rasmussen tracker today Romney's lost two points..going the wrong direction wouldn't you say? hehee
12:48 PM on 11/03/2012
Really goldilocks? Living in that fantasy bubble? Again? Go see the most recent polls. Colorado 5 latest polls Obama up by 2 to 4 points in four of the five polls. Tied in one. IA last 6 polls show Obama winning very last poll shows him up by 4.
New Hamsphire last seven polls show Obama up. Nevada last 9 polls have Obama up , and the latest poll shows a 6 point lead. I suggest you push the hair away from your face and wake up from your stupor, because right now, I dont see how Romney can win this thing.
10:42 AM on 11/02/2012
And to go along with goldilockspryor, I'm in Ohio and it is NOT a safe bet for Obama by any stretch of the imagination! The only polls showing him up are massively oversampling Dems with a D+7/D+8... when Ohio only went with Obama with half that! (And went R+1 in 2010 with Congress!)
11:50 AM on 11/02/2012
Good luck with your polls are bad prediction. Rasmussen was WAY Off in 2008 predicting an Ohio tie. He was only off 3%. No worries right:) Obama is up over 2% in Ohio and historically when a candidate leads by 2% or more it's an 80% win rate. You have a massive understanding of sample party ID;s, But you will see Nov 6th:)
03:04 PM on 11/02/2012
Why is it that Republicans systematically avoid answering pollsters? In other words, over 100 or so polls, each having a sample size of about 1,000, why is it that in such a wide sample, there is still a D+7 or D+8.

If the REALITY is R+1, when I randomly call 100,000 people, why don't I see a sample that is VERY CLOSE to R+1?

Are Republicans just not picking up the phones?
10:30 PM on 11/01/2012
the numbers are pretty clear; Romney has a strong lead amongst Republicans and unaffiliated voters.
I think that means he is going to win.
Also, Ohio is a fairly red state. Their governor is Republican and usually is. Their state legislature is Republican and usually is. Their congressional delegation is mostly Republicans and usually is.
12:54 PM on 11/03/2012
Really? I guess the last 10 Ohio polls that have Romney losing Ohio mean nothing. Also I guess the early voting where Obama is up by THIRTY points in Ohio also doesnt mean anything right? BTW that 1.3 million votes already cast in Ohio.
09:45 PM on 11/01/2012
Most incumbent presidents increase their margin of victory when they win a second term. Even the hated George Bush managed to do that. Well under 50%, around 47.4 on real clear politics average, Obama is not looking good. Pretty sad when George Bush gets more support for a second term than you do. Perhaps Obama should stop bashing and smearing Romney and dust off some of those hope and change buttons. No positive vision for the future = loss for team Obama.
Kelly ODonnell Griffin
The grass really isnt greener on the other side. I
03:27 AM on 11/02/2012
I am a Romney supporter but Acorn is still out there and all those dead voters in places like chicago Philly and Maryland will have a vote too. Also If Romney wins better be ready for the riots white man don't have a chance in this liberal bias bull. The excuse will be the voting made me do it. Never have more people been so involved and sadly it has nothing to do with the stare of the country but more of the color of the man skin America is still stuck after all these years. Obama not a bad guy he would make a great GQ model and a Co host on the "View" when Walters not in he does make for good eye candy
11:11 AM on 11/02/2012
Already voted for Obama and Sherrod Brown the instant my early ballot arrived in the mail, October 3rd or 4th. Sent it in very excitedly and promptly. It shows received by the board of elections on 10/17/2012. My partner, who I cannot marry, did exactly the same. There ya go!
11:50 AM on 11/02/2012
Philly voting is just an example of amazing patriotism when they kick out an aggregated vote that equates to 110% of their registered voters.....over achievers they are!
09:43 PM on 11/01/2012
Funny that the "trend show Romney win" article came out on same the day that the RCP polling average showed an Obama lead -- a change of almost a percentage point leftwards from just a few days ago. The national polls are indeed close, but the trend strongly favors Obama.
11:51 AM on 11/02/2012
Strongly huh?
03:32 PM on 11/02/2012
Yes, strongly. Going from a point down to half a point up in 3 or 4 days is a strong trend for Obama. (Not in any one poll, which could have a ton of statistical noise, but in the polling average.) This doesn't mean that the trend will continue. Lord only knows what people will think about over the weekend to help them decide, and Lord only knows who will actually show up on Tuesday.

Arguing based on short-term trends is always iffy. But if you're GOING to argue based on trends, as Lombardo does, then the advantage is all Obama's.
08:00 PM on 11/01/2012
Running a 26 data point regression model is an awful model. This guy is a CEO and he is coming with this? The predictive power of a regression model with 500 points is questionable at best. On top of that the small sample is even worse because the data points have a low probability being where they are placed because of large margins for error on each point. This guy is either bad at modeling; loading his model with favorable variables for Romney or being disingenuous. For the sake of his company I hope it is the latter.
09:55 PM on 11/01/2012
You don't seem to be taking the central limit theorem into account.
12:28 AM on 11/02/2012
Don't think 26 observations is near enough to invoke CLT. Assuming a strong enough trend to project what he is projecting using that data set is what provoked my response. If I am running a regression model with that small sample the confidence level would have to be ridiculously low so him saying Romney is going from even to +3.5 in 6 days is basically banking on a Karl Rove level of Democratic oversampling while assuming his numbers in the same polls for independents are accurate a.k.a. a miracle.
07:45 PM on 11/01/2012
At this late date, everyone has made up their mind. Independents my foot!
11:52 AM on 11/02/2012
I agree with is a turn out and enthusiasm question...
Hugo Marquez Soljancic
06:36 PM on 11/01/2012
"Please note that the author was an advisor to the Romney for President campaign in 2008, but is not affiliated with any campaign in 2012."

Romney winning the popular vote by 3% with 51% of the share? Simply ridiculous.....look at all of the other polls and then decide what to make of this analysis
09:59 PM on 11/01/2012
I'm think it's very easy for Romney to win the popular vote and Obama to win the election. Obama has a lot of paths to victory, Romney needs miracles. However Romney can easily draw up the popular vote, particularly because of the amount of support he's gotten in states that are strong for Obama. California and New Hampshire for example, and if he loses Ohio and Colorado will certainly have large numbers there.
Hugo Marquez Soljancic
12:19 AM on 11/02/2012
it could happen, the national numbers will be much closer than this column is implying, possibly to be decided by less than one point, but the swing states are Obama's to loose....I think that more than anything, it will be important for Obama to win the national voting to be seen with more legitimacy on the next 4 years, so in that sense, every vote counts
06:13 PM on 11/01/2012
It's not going to be as close as people think and not the way many think. Romney gets over 300 electoral votes. For Obama to actually win, he would be bucking virtually every historical trend considering the state of affairs with the economy. Americans vote with their wallets.
10:02 PM on 11/01/2012
There's one trend he wouldn't be bucking, and that's the ever increasing amount of loyalty to incumbents. Reagan, Clinton, and Bush all got far stronger in re-election. H.W. would have handily defeated Clinton but for Perot. What is it 98% re-election rate in Congress? America isn't even trying anymore.
09:23 AM on 11/02/2012
Well Congress is a different animal entirely because the whole country doesn't elect them only who they represent. If the whole country did there would be lots of changes. On the contrary I see support for Obama as very soft among many Democrats. They are trailing in the early voting in key states like Ohio and Democrats don't show up at the polls on election day itself like Republicans do. Turnout will be everything.
10:49 PM on 11/01/2012
For Romney to win there has to be a TOTAL STATE POLLING MELTDOWN. 2000-2010 state polls were accurate 90% of the time. Good luck with your "FEELING."
09:20 AM on 11/02/2012
When you say "Polling" you are talking about a wide swath. It's not a "feeling". An incumbent at less than 50% in almost every state is not in great shape.
11:56 AM on 11/02/2012
My "Feeling" is when a poll samples Dems in ohio that implies a greater turn by a factor of 2 from what turned out in 08 and a factor of 4 in not the one making estimations based off of "feelings"
05:27 PM on 11/01/2012
RE: "The trend line-based on 26 national polls conducted over the last 30 days --is both unmistakable and virtually unassailable."

Okay, I'll bite. If it's based on 26 polls why are there more than twice that many data points on the graph? What website can we go to to access the data (so we can all check the model)? Have all assumptions of the model been tested? Was the cubic trend significant? Does the model make the same prediction if you start it at Election-30 rather than Election-120? How about just publishing the regression model itself? I'd like to know Romney's margin if we waited to hold the election until Dec 1st. (Okay, that was a joke, but if you want to project beyond the data you have, you're kind of asking for this sort of thing.)
04:59 PM on 11/01/2012
Have you read this guys bio? It's clear who his horse is in this race.
04:42 PM on 11/01/2012
So if we continue your linear regression, if the election would be held 2 months from now, Romney would be at +100? I think not.
Always learning and leaning forward
04:31 PM on 11/01/2012
Please note that the author was an advisor to the Romney for President campaign in 2008, but is not affiliated with any campaign in 2012.

Really? Shouldn't you have put that up front in the article. In the front of my dissertation, before it was submitted for review, had to make a statement of bias. Not very professional of you I would say.
05:00 PM on 11/01/2012
You can't finish a short article? It's right at the end, what's the problem?
11:58 AM on 11/02/2012
Should we go down the litany of authors for the huff post and other major news outlets/media who have been contrinutors and active members of the DNC in the past and currently....

Then should we summarily discount their articles because of that?
Always learning and leaning forward
12:50 PM on 11/02/2012
It should be disclosed up front, either side. 
04:28 PM on 11/01/2012
The "nationwide" polls cited by the author may support his claim that Romney is likely to win the popular vote, but a state-by-state survey likely would lead to a different conclusion.

The RCP poll averages show, among other things, the states in which each candidate has a double-digit lead. The combined population of the states in which Obama has a double-digit lead, including California (the most populous state by far), is significantly greater than combined population of the states in which Romney has double digit leads, the most populous of which is Texas which has only about 2/3 the population of California.

Thus, the more reasonable conclusion is that Obama will win the overall popular vote, along with the electoral college vote.
10:50 PM on 11/01/2012
Obama will win both. Take it to the bank.
12:05 PM on 11/02/2012
I hope you already have. The economic boom that will happen after Romney pulls this off will need its burger-flippers....
03:45 PM on 11/01/2012
Dude come off it. There are "lies, damn lies and statistics". Statistics are at work here and they are being used to bolster weak arguments. For example, Lombardo states, "[B]oth camps have a strong "ground game". Really, when has Romney's "strong ground game" been tested? The answer is never. Obama's however has. In the 2011 landslide repealing the Ohio GOP championed union destroying law, Obama's ground game worked hand in hand with unions to turn out the vote for a 62-38% victory. The author apparently thinks you can rush in during the spring, throw up a few field offices, staff them with "paid volunteers" and expect them to actually compete against an organization in place for the past 6 years and tested successfully as recently as a year ago. Well, you can’t. Yes, there are lies, damned lies and statistics and they are all being used here by a Republican pollster to bolster the flimsiest of arguments.