Two new polls came out late last night that show Scott Brown rolling into Tuesday’s special election with momentum and possibly an insurmountable lead — depending on which poll one prefers.  The new PPP poll shows Brown with a five-point lead and a majority of the vote, 51-46, which this late may mean that majority will have cemented in the three-person race.  A larger majority (58%) believe that Brown made a case for his campaign, while only 41% believe the same about Coakley:

Scott Brown leads Martha Coakley 51-46 in our final Massachusetts Senate poll, an advantage that is within the margin of error for the poll.

Over the last week Brown has continued his dominance with independents and increased his ability to win over Obama voters as Coakley’s favorability numbers have declined into negative territory. At the same time Democratic leaning voters have started to take more interest in the election, a trend that if it continues in the final 36 hours of the campaign could put her over the finish line.

The PPP news isn’t all good:

-Brown is up 64-32 with independents and is winning 20% of the vote from people who supported Barack Obama in 2008 while Coakley is getting just 4% of the McCain vote.

-Brown’s voters continue to be much more enthusiastic than Coakley’s. 80% of his say they’re ‘very excited’ about voting Tuesday while only 60% of hers express that sentiment. But the likely electorate now reports having voted for Barack Obama by 19 points, up from 16 a week ago, and a much smaller drop from his 26 point victory in the state than was seen in Virginia.

On the plus side, the likely voters in the PPP poll still oppose ObamaCare 48/40, which is one of the main issues driving opposition to Coakley.

However, the sample of the PPP poll looks suspect.  Like the ARG poll, it likely overstates Democratic turnout, with a +22 partisan advantage (D-R-I sample 39-17-44).  It does show only a 56-37 split for voters in the 2008 presidential election, but that includes four percent who “don’t remember” which candidate got their vote.  This turnout model is unlikely to match the actual turnout in this special election, especially since Democrats only outnumber Republicans by a +17 on party registration.

The new Pajamas Media poll, also conducted with IVR response by Cross Target, looks more realistic.  It uses a turnout model of +14.6% for Democrats, which seems a lot more realistic than a +22, and gets the same lead that the Merriman Group found for InsideMedford:

A poll taken Sunday afternoon while President Obama was in Massachusetts campaigning for Democrat Martha Coakley against Republican Scott Brown for the open Senate seat in that state showed Brown leading his Democratic opponent by 9.6% (51.9% to 42.3% with 5.7% undecided).

The poll, conducted via telephone for Pajamas Media by CrossTarget, was of 574 Likely Massachusetts Voters and has a margin of error of +/-4.09%. CrossTarget used the exact method – Interactive Voice Technology (IVR) – it used in a similar poll for PJM on Friday. The previous poll showed Brown ahead by approximately 15%.

A poll released earlier Sunday from the Merrimam River Group shows Brown up by an identical 9.6%.

The big takeaway on this is the favorability ratings of both candidates.  Brown gets a 60/32, while Coakley sank to a 40/54 negative rating.  And that comes from a sample with a partisan advantage of almost 15 points for the Democrats, which explains why Democrats aren’t fired up and ready to go for Tuesday.

The PJM poll on its face suggests that the race tightened, with Brown dropping from 15 up to the 9.6 point lead from last night.  I’d guess that the difference comes from a better polling sample.  Even if the race really “tightened” to a 9.6-point lead, I think Brown will take it.