The Mills campaign released a poll this morning showing (surprise!) their guy ahead.

The poll is by a new public opinion research firm called the Maine Center for Public Opinion and has a high sample size of 915 respondents.

Here are the results: Mills 22.1%, Otten 17%, Abbott 13.9%, LePage 10.7%, Beardsley 5.9%, Poliquin 4.9%, Jacobson 1.4%, unsure 24%.

In addition to the normal limitations of primary polls, polls paid for by campaigns and polls with large survey windows (this one is more than a week), there are a couple of decisions made by the pollsters here that may have affected the results:

First, they used a very tight voter screen - only registered Republicans with a strong voting history dating back two or three years were called. (Contrast this with Pan Atlantic's approach of apparently allowing respondents to self-identify as "likely" voters.) This screen may have missed some voters who are newly registered or newly energized to vote in this particularly primary.

Second, the question about candidate preference comes at the end of the poll, after a series of questions about Governor Baldacci, the economy, school district spending and the Tea Party. Any of these questions could influence how people think about the governor's race. Most pollsters looking for a clean response on the horse race question ask it first. For instance, people who have just been asked to consider whether or not the Tea Party is hurting the GOP might be more likely to choose a moderate like Mills.

Other parts of the process, from the question wording to the pre-weighting they did by geographic area to how their "virtual phone bank" works may also have affected the results.

The press release from the campaign implies that the firm has done other polls for them, including some earlier ones showing Mills behind. I'd be interested in seeing if their methodology has changed at all from poll to poll.

Here's the full memo:

I'm very excited about how well MPA's text message based straw poll of the gubernatorial race is going.

I was excited to try out some new technology but was worried that our membership and the general population of Maine, both of which skew a bit old, might not be into the whole "texting" thing.

It seems I was wrong. We've had hundreds of responses, coverage in the Keenebec Journal/Morning Sentinel, and it was great to see volunteers for all the candidates at the recent Democratic convention wearing stickers asking delegates to text in a vote for their candidate.

If you haven't yet voted, just send a text message to 978-242-6207 with the last name of any of the 11 candidates in the gubernatorial primaries. If you'd like to round up more votes for your preferred candidate, feel free to invite your friends to this Facebook event.

Results will be announced this Thursday.

Libby Mitchell's campaign today released an internal poll showing the Senate President far ahead of her competitors in the Democratic gubernatorial primary. According to the survey of "587 registered and likely June Democratic primary voters" interviewed April 11-14 by Greenberg Quinlan Rosner Research, Mitchell is in the lead with 36% followed by Rowe with 16%, McGowan with 13%, Richardson with 5% and Scarcelli with 4%.

This is the Mitchell campaign's own poll, so take it with a handful of salt.

If accurate, it shows that Mitchell likely has a significant name recognition advantage going into the television advertising wars that are about to begin. Who ends that fight with the advantage and wins the primary is still anyone's guess.

Full polling memo here, including name ID and favorability ratings.

Kiley and Company responds with information on the methodology of the poll showing Libby Mitchell ahead of some of the other Democratic gubernatorial candidates:

301 likely Democratic primary voters in the state of Maine were interviewed. Trained professionals, working from a central, monitored location, conducted the interviews by telephone during the evening hours of July 21 and 22, 2009. Respondents were randomly selected from an up-to-date file of Maine voters who have cast ballots in at least one recent Democratic primary election. They were screened to confirm that they are likely to vote in the Democratic primary in June 2010. County and gender quotas were imposed to ensure the sample population is representative of the Democratic primary electorate.

The margin of error for a dichotomous question for a sample of this size is approximately ±7 percentage points.

I assume the margin is 95 times out of 100.

Elizabeth Mitchell's campaign has released a poll showing the Maine Senate President trouncing three potential Democratic opponents in the June 2010 primary election.

When asked who they would vote for if the election were held tomorrow between Mitchell, Steve Rowe, Patrick McGowan and Dawn Hill, 37% chose Mitchell, 20% Rowe, 10% McGowan, and 3% Hill.

In a head-to-head match up, Mitchell beats Rowe 46% to 24%.

According to campaign volunteer Jodi Quintero, the poll is independent and was conducted before Mitchell entered the race by research firm Kiley and Company. Mitchell was then approached by the firm and chose to purchase certain questions from the survey for release.

The firm interviewed 301 Maine Democratic Primary voters (a relatively small sample size) on July 21-22, 2009. No other methodological information is given. Rosa Scarcelli, John Richardson and other current and potential Democratic candidates were not included in the poll.

Daily Kos/Research 2000 also released a Maine poll today showing every Democratic candidate polled beating every Republican, except for a match up between Steve Rowe and Les Otten, where Otten edges Rowe by 1%.

Full results after the jump...

This week at Down East I take a look at which pollsters came closest to predicting the election results in Maine.

SurveyUSA has announced some results from its final poll in Maine.

President: Obama 58%, McCain 38%
US Senate: Collins 56%, Allen 42%

(674 likely voters - Oct. 29-31 - ±3.8% MoE)

A new Market Decisions poll has the casino referendum statistically tied, but shows large margins in favor of Obama, Collins, and for repealing the beverage tax.

President: Obama 52%, McCain 33%
US Senate: Collins 54%, Allen 37%
Question 1: Yes 66%, No 28%
Question 2: Yes 49%, No 48%

The survey of 387 likely Maine voters was taken from October 13–26 with a margin of error of ±5%, 95 times out of 100.

This poll has a smaller sample size and was taken over a longer period than some of the other recent surveys. Polling even is good for CasinosNo!, as undecided voters are thought to be more likely to ultimately vote "No" (in favor of the status quo) on any given referendum.

Another poll, this one by Critical Insights, shows big leads for Barack Obama and Susan Collins. Obama leads by 20 points in the contested second district and Collins is ahead in her race by 12 points statewide.

The poll also shows Congressman Mike Michaud beating Republican challenger John Frary 64-23. If this result holds, Frary may garner a smaller percentage of the vote than Laurence D'Amboise, who lost to Michaud 70-30 two years ago.

The poll also has Chellie Pingree beating Charlie Summers in CD1 by 21 points and the beverage tax repeal passing 48-39. The survey has a sample size of 443 likely voters and a margin of error of ±4.7%, 95 times out of 100.

The AP has some interesting reaction from the Allen campaign:

Carol Andrews, spokeswoman for Allen's campaign, said recent independent polls show "wildly erratic results typical of a pre-election environment." Another independent poll released this week gave Collins a lead of nearly 21 percentage points.

"We are unconcerned and undeterred as we know that we are making the case with Maine voters that if they seek change they have to vote for it, and they won't get it by voting with Susan Collins," Andrews said.

Rather than being erratic, I'd say the polling in this race has been remarkably consistent. Every poll taken since the campaign began has given Collins a large lead.

(hat tip: TMR)

The Press Herald has some toplines from the Pan Atlantic poll. The survey, conducted October 13-16, has a sample size of 400 and a margin of error of ±4.9% 95 times out of 100. In past polls they have not weighted their results.

President: Obama 51%, McCain 39%
US Senate: Collins 57%, Allen 36%
CD1: Pingree 43%, Summers 32%
Question 1: Yes 57%, No 35%
Question 2: Yes 40%, No 52%

If these results hold, Mainers will be saving a few cents on soda but won't be able to bet that money on roulette.

A new WCSH-WLBZ/SurveyUSA poll of 642 likely or actual voters conducted on Sunday and Monday finds Allen gaining on Collins, but likely not quickly enough to win the election now less than two weeks away. Collins leads 54-43 in the survey, compared to 55-39 four weeks ago.

This survey has a higher sampling of Democrats than the previous poll (28% of whom say they're voting for Collins).

The one ray of sunlight for Allen is the 14% of respondents who report that they've already voted. Allen is winning this group by 5 points (with a higher margin of error).

In the same poll, Obama is winning the second CD by 6 and the first by a stunning 22 points. In the poll 4 weeks ago, Obama led by 5 points in both districts.

It looks like McCain's spending in CD2 and the Palin visit have kept it from being a blowout like southern Maine, but this result puts it squarely in the second tier of battleground states, which include Virginia, Colorado, New Mexico, New Hampshire, Minnesota, and Pennsylvania.

In a testament to his campaign's early voting operation, Obama leads by 27 points among those who have already voted. The poll has a ± 3.9% margin of error.

Pan Atlantic SMS will also be releasing a poll later today.

Liberal magazine The Nation weighs in on Collins' lukewarm opposition to McCain's robocalls:

If Susan Collins wants to be like Margaret Chase Smith, she needs to make her own "Declaration of Conscience" -- and then she needs to act upon it.

[...]doesn't she need to speak up, forcefully and in detail about why the McCain campaign's tactics are wrong? Doesn't she need to take her party's nominee to task and say, as Margaret Chase Smith did more than half a century ago: "I don't want to see the Republican Party win that way."

And what if McCain does not listen to Collins? What if he and his campaign refuse to halt not just the robo calls but related smear tactics against Obama, who happens to be a colleague of the Maine senator?

If McCain refuses to right the course of his McCarthy-like campaign, shouldn't Collins quit as his campaign co-chair for Maine?

Kevin Wack thinks the robocall issue helps the Collins campaign and represents a tactical mistake by the Democrats.

If the new numbers from Research200/DailyKos are right, it doesn't really matter. The poll has Collins up by 13 points. They also have Obama up by 11 in the second district.

More of the October 2nd Rasmussen poll has been released, and it puts Collins at 53% and Allen at 43%. This is the highest level of support Allen has received in Rasmussen polling, and it gives some credence to the recent DSCC poll, but the movement from the last Rasmussen poll is small, and within the margin of error.

10 points is closer than the 19, 13, and 16-point margins that the last three polls (from different companies) have shown, but it's still a big deficit with less than a month to go until the election.

The Allen campaign has a new ad up on the economy:

The attacks on Collins' economic record from the "worst" ad have also been slightly repackaged in a new, more accurate advertisement.

There's been a lot of talk about the DSCC poll showing an 8-point race, and some haven't noted the problems with reporting internal polls. I'll wait for independent results. The big firms should have some new polls out soon.

A survey USA poll taken on Monday and Tuesday of 675 likely voters found Obama ahead of McCain by 5 points and Allen trailing Collins by 16.

This is a fairly large survey, so one would think the internals might be somewhat reliable, but the poll contains some strange numbers. For instance, McCain is up 12 points over Obama among respondents 18-34 and there's virtually no difference between "North" and "South" (I assume they used the congressional district boundaries for this determinant) between McCain and Obama.

Keep in mind, this poll was taken before McCain gave up on the presidential race and wandered off somewhere yesterday.