Research 2000 for Daily Kos (1/15-17, likely voters, 1/12-13 in parens):
Martha Coakley (D): 48 (49)
Scott Brown (R): 48 (41)
Joe Kennedy (L) 3 (5)
Undecided: 1 (6)
(MoE: ±2.8%)
First the bad news: R2K gives us a trendline pointed steeply down… an R2K poll conducted earlier in the week (paid for by Blue Mass Group rather than Daily Kos, but seeing as how that shouldn’t change the numbers, we’ll accept that as a trend) gave an eight-point lead to Democratic AG Martha Coakley, and this one sees a tie. On the other hand, that’s the best result that rolled in over the last few days: not only are there the Suffolk (-4) and PPP (-5) numbers, but also ARG (which sees a 48-45 Scott Brown lead)… and the Merriman River Group, whoever they are, who found a 51-41 lead for Brown in a poll that was apparently taken over the space of four hours and found no undecideds, so take that for what it’s worth.
So, should we be pleased or not? Does this mean that Coakley’s bungee-jump downward over the last week got arrested right before she hit the bottom of the canyon? There are a few other positive indications; the constantly-leaked Coakley internals, for what they’re worth, seem to have stabilized over the weekend (which saw Barack Obama and Bill Clinton appearances, and maybe some backlash over the “curling iron” incident), to the extent that they reportedly show Coakley up 2, according to the Boston Herald. (Nate Silver has a helpful graph of all poll trendlines that includes leaked Coakley internals, which brings a lot more datapoints to bear.)
One other indication is that state officials are suddenly looking at extremely high turnout, at near-presidential levels, with everyone suddenly focused on the election — with estimates of up to 70% turnout, based on absentee ballot requests. Turnout, as you know, most likely helps the Democrats here — and the pollsters that have been giving pro-Brown results may be basing their likely voter models on now-obsolete projections based on low-turnout, high-intensity-voters-only projections. One other good Nate Silver observation is that Obama’s approval is polling under 50% in most polls, which is lagging his national averages… in Massachusetts, one of the bluest of all states… suggesting their LV models are predicated on conservatives disproportionately showing up. (Of course, he also points out the possibility of what Rasmussen alone seems to be seeing: people approving of Obama, but still voting for Brown.) But if the state’s turnout predictions are to be believed, maybe some of those unlikely-voter Dems who were planning to sit this one out or weren’t aware of it have finally realized there’s a real race here and have gotten converted into LVs over the last few days, and pollsters are still playing catchup.
RaceTracker Wiki: MA-Sen
…I think she wins. It will be an ugly win, one way more expensive than it should have been, and for bad reasons that could easily have been prevented. But there’s now too much attention on the race, Obama being there and the focus of Coakley’s closing ad (http://bit.ly/69PxJ4)–evidence, btw, that the campaign thinks Rasumussen is wrong and that Obama is a good asset for Coakley–and the overwhelming Dem makeup of the electorate all suggest to me that she’ll eke out a win.
But this won’t be a win that people involved in the campaign prior to about a week ago will ever want to brag that they were a part of.
So, here’s to optimism that we’re probably going to run no worse than about 20 points below the Democratic base!
If there is 70% turnout, there’s no way Brown wins.
I won’t be watching the returns at all. I do not want to be one of the first people to know that MA is now a represented by John Kerry and Scott Brown if that teabagger wins.
Independent – 43%
Democrat – 37%
Republican – 20%
I D R
Brown – 70/17/98 = 56%
Coakley – 27/82/0 = 42%
Kennedy – 3/1/2 = 2%
We can do this if people just go out and vote!!!!! Nate Silver says its possible then its possible. Keep making phone calls!
But my gut says the opposite. Of the two things Crisitunity suggest might be positive, I don’t think you can hang your hat on internal polls when all other polls are pointing to the opposite direction, and the turnout idea sounds good, but it’s hard to say who is going to turnout or whether the woman quoted in the Herald is right about 70 percent turnout. I’ve never heard of election where turnout somehow caused a race to go against every poll that was taken.
I think you have to look at the public polls. Only this one even points to a tie, at this point. No one has shown a Coakley lead. (Rassmussen still has to weigh in – scary that we are relying on them to give some hope). So if I were a betting man, I’d bet on Brown.
I’ve lost a lot of bets in the past, though. I won’t be completely surprised if lose this one too, so I’m not saying anyone should give up yet.
the argument of “the polls were wrong for these primaries that Obama lost so they could be wrong here” is…. spurious. It’s a coping mechanism to deal with trauma.
Unless people were saying that they were for Scott Brown but secretly for Coakley. Considering that Scott Brown is a white male, I doubt that will happen.
There’s a shot Coakley wins, but don’t ask me to bet money on it.
is getting slaughtered by double digits in Gardner, Fitchburg, and Peabody counties according to Suffolk university.
http://www.politico.com/blogs/…
the R2K poll did screw up by IDing Joe Kennedy as “the libertarian candidate” when he’s listed on the ballot as an Independent. Which means that it undershoots the Dem votes for Kennedy (who will confuse him with the Democratic congressman) and overshoots the Rep votes for Kennedy (who are more friendly to libertarianism). Which makes Coakley look better than she really is.
Is it really that hard to identify a third party candidate by his actual ballot label as opposed to his party affiliation?
I think the 70 percent-presidential level TO stuff is just nonsense.
Besides, the key issue is not total TO but the DRI distribution.
I reckon we will see high GOP TO. We always do. Ds might be down a tick from their 3-1 ratio but the real key is Indies will not TO proportionately to their registration, allowing the Ds edge in registration to hold the seat.
And I reckon the Obama visit and the stakes will spur D TO a bit. The key is whether Coakley loses Indies 2-1 or holds the bleeding to 40/60.
What I don’t see is the Is actually turning out in numbers corresponding to registration. My gut is the Indies will end up being less than 40 percent of total votes cast, probably 35 percent or a little less. I think there will more total D voters than I.
MC% Coakley Brown COAK BROWN
D 48 75 36.00 12.00 D 44 33.00 11.00
R 17 10 1.70 15.30 R 16 1.60 14.40
I 35 40 14.00 21.00 I 40 16.00 24.00
51.70 48.30 50.60 49.40
Figure Kennedy keep two percent, mostly out of Brown’s indy edge, so take 40 percent of his vote from Coakley and go 60 from Brown.
End up with
Coakley 50.9
Brown 47.1
Kennedy 2.0
Now, if Coakley does loses 2-1 among Indies
MC% COAK BROWN COAK BROWN
D 48 75 36.00 12.00 D 44 33.00 11.00
R 17 10 1.70 15.30 R 16 1.60 14.40
I 35 33 11.70 23.30 I 40 13.33 26.67
49.40 50.60 47.93 52.07
Kennedy.81.2.8 -1.248.6 49.4 47.13 50.87
then we are toast unless we take the full Kennedy vote out of Brown’s column (a possibility given the dynamics of the race with Coakley as de facto incumbent)… in which case Coakley can still pull out a squeaker.
And all this is contingent on Coakley losing a quarter of registered Dems. If she holds 80 percent, the math gets tough for Brown.
Look at my worst case 40 Indy scenario if she holds 80 of Ds still losing 2-1 among Indies.
MC% COAK BROWN
D 44 80 35.20 8.80
R 16 10 1.60 14.40
I 40 33 13.33 26.67
50.13 49.87
– Kennedy -.8 -1.2
49.33 48.67
If Kennedy falls to one point,
49.73 49.27
Brown’s winning scenario involves holding 90 percent of Rs (likely), actually holding the 2-1 edge among Indies (think 60/40 more likely) and Indy TO being above 40 percent (possible but a real break from pattern, esp in a special election). A 37.5-62.5 Coakley loss among Indies could give us a Florida-style dead heat.
Brown is going to run close but come up short. The 3-1 edge Ds have among folks registered by party is just too much to overcome. Democrats should be spooked by the erosion among Indies but this is only a near-death experience… not metaphysical disaster.
Of course, that’s just my opinion. I could be wrong.
I’m looking at the internals of the R2k Connecticut poll, and it of course shows Blumenthal running far ahead and Obama with a healthy 63-30 fav/unfav. It even shows Ned Lamont winning the governor’s race with 46% of the vote, and he’s only really popular with the liberal base.
Now Connecticut is not Massachusetts, and in fact votes a bit more conservatively than MA ever had (look at one of our senators for proof, and that Kerry only won CT by 10% versus 25% in MA). I can’t figure why the Dems are in such good shape in CT and in the toilet our neighbor to the north.
Blumenthal is AG, Coakley is AG (though he is a bit more popular and has served longer). Blumenthal leads for a tainted seat (Dodd), but Coakley trails for Ted Kennedy’s seat? I just think something is off here, even if a crunched schedule for the special election has heightened caused by decision making by voters or exasperated a badly run Coakley campaign.
Massachusetts is by no means a stupid state and should know what it means to send a Republican to the Senate. I’m actually surprised no ads have been run to warn voters that they’ll have a republican for 4 years. CT voters are stuck with Lieberman now, MA should be smart enough to avoid that problem.