SUSA, Steele & Cardin tied.

[Moooooo]

No link just an image from the news.



From today.  I NEED GOV RESULTS!@#$!@#

[Sam Spade]

Once again, it's like SUSA is polling a different race.

[Moooooo]

Sam, quick question.  Is there a difference between a poll and a tracking poll?

According to Lowry, Steele's internals have him down 8% today.

Yesterday:

You must be logged in to read this quote.

Today:

You must be logged in to read this quote.

[WalterMitty]

nick, im a little worried about one thing...please correct me if im off base.

a steele victory or near victory might defeat ehrlich?  im thinking that the steele candidacy may draw a few more blacks to the polls to vote for steele...wh will also likely vote for o'malley.  could make the difference in a close race.

[Moooooo]

nick, im a little worried about one thing...please correct me if im off base.

a steele victory or near victory might defeat ehrlich?  im thinking that the steele candidacy may draw a few more blacks to the polls to vote for steele...wh will also likely vote for o'malley.  could make the difference in a close race.

Not off at all.  Ive heard that arguement before.

[Mike in Maryland]

I'm not Sam, but I think that tracking polls tend to have an average of a few nights rolled together.  Plus they're done consistently throughout campaign season (almost every day they do a similar number of interviews), while other polls are done here and there.

SUSA may be wrong, but they're at least consistent.  For since the nominations were determined they continue to show the race even while all other surveys have a Cardin lead of varying size.  Over the last week we've had an 11-point Cardin lead, a tie, and three polls with 5-6 point leads for the Dem, which seems right at midpoint.  Overall, I think Cardin has a decisive edge, but not an insurmountable one.  Not as sure about O'Malley, though I think he too is ahead by a bit.

Interestingly, Steele was on CNN yesterday (or was it Wed.?) claiming that his internals showed the race closer than the public polls.   In light of Lowry's observations, perhaps that was another of his untruths, or he was at least gilding the lily to make his prospects sound better.

[Sam Spade]

Sam, quick question.  Is there a difference between a poll and a tracking poll?

According to Lowry, Steele's internals have him down 8% today.

Yesterday:

You must be logged in to read this quote.

Today:

You must be logged in to read this quote.

Polls and tracking polls are essentially the same.  With a poll, you do one two-three day sample of a certain number of people and it ends, you release the results.

In a tracking poll, you do a sample of say 500 likely voters on Friday, then the same on Saturday and Sunday.  Then, you sample 500 likely voters on Monday and drop the Friday sample.  Repeat and rinse.  It's akin to what Rasmussen does on the Bush approval numbers and is thought to be better because you can catch day-to-day trends.  Of course, you get outliers sometimes. 

Look, with SUSA polling, it's obvious that they're either off or they're catching something we're missing.  The easiest thing is to believe they're off, but the Governor's results they have been pulling with the same samples (here it is O'Malley +1) are along the lines of the other polls.  I have sort of become convinced that the difference in numbers has to do with the way they ask the horse race question, with the addition of the lines, "if you were in the voting booth right now". 

It comes down to this: do you believe there's going to be a sizable portion of the population (black) who's going to lie to pollsters until they're "in the voting booth", and then pull the Republican lever for Senate?  If you do, then this race is a pure toss-up.  If not, then it's a five-six point lead.  That's the reason why I call this the most difficult race to predict right now.

One other thing:  About these public pronouncements of internals from inside sources, "Do Not Believe Them!"  In some races, these people will lie too liberally, in others they will lie too conservatively, in others they will simply just lie.  I cannot get this point across enough; we will have so much of this in the next few days, it's quite easy to let it all get to you and cloud your thinking.

[Mike in Maryland]

It comes down to this: do you believe there's going to be a sizable portion of the population (black) who's going to lie to pollsters until they're "in the voting booth", and then pull the Republican lever for Senate?  If you do, then this race is a pure toss-up.  If not, then it's a five-six point lead.  That's the reason why I call this the most difficult race to predict right now.

I've heard that theory tossed around, mostly by Republicans, that blacks who intend to vote for Steele are lying to pollsters.  I don't buy that to any appreciable extent--certainly not enough to explain a five or six point difference among the general population.  Why would they tell the truth to Survey USA and lie to all others?  Nor do I think that there are a whole lot of whites who tell pollsters they intend to vote for him and then won't on election day just because he's black.

[Sam Spade]

It comes down to this: do you believe there's going to be a sizable portion of the population (black) who's going to lie to pollsters until they're "in the voting booth", and then pull the Republican lever for Senate?  If you do, then this race is a pure toss-up.  If not, then it's a five-six point lead.  That's the reason why I call this the most difficult race to predict right now.

I've heard that theory tossed around, mostly by Republicans, that blacks who intend to vote for Steele are lying to pollsters.  I don't buy that to any appreciable extent--certainly not enough to explain a five or six point difference among the general population.  Why would they tell the truth to Survey USA and lie to all others?  Nor do I think that there are a whole lot of whites who tell pollsters they intend to vote for him and then won't on election day just because he's black.

Look, I shouldn't buy it, but it's the only explanation that I've got other than totally discounting it (which is, of course, another possible explanation).  As I have said before, I am not buying either explanation at this moment, but it is stuck there in the back of my mind.

The rationale is based on the way SurveyUSA asks the horse race question: 

"If the election for X were today, and you were standing in the voting booth right now, who would you vote for?" 

No other polling company asks the horserace question this way.  I can pretty much categorically prove it's the reason why SUSA has fewer undecideds than any other state polling company, and the reason why their results jump around like crazy.  But I cannot prove anything concerning other attributes this question raises.

The simple fact is that you believe that the race is what the other polls say it is:  a five-six point Cardin lead.  I have said that this is fine.  I am personally torn between the two because of the two counterveiling reasons (and I have other subjective reasons too), but in all of my predictions, I have held with what you would believe and will continue to do so.

But it will still be in the back of my mind, until election day.

[Eraserhead]

SUSA is getting everybody worked up over possibly nothing.