Each week brings an ever increas­ing flood of data. For this rea­son, today is the last weekly ver­sion of Reëlec­tion Watch. Start­ing next week, Reëlec­tion Watch and Sen­ate Watch will be com­bined in a for­mat that will appear for the last three slots prior to the election.

What’s the lat­est news? Let’s dive in and see.

National Polls

Since Octo­ber 15, in the national pop­u­lar vote matchup of Pres­i­dent Barack Obama ver­sus Repub­li­can nom­i­nee Mitt Rom­ney, the Repub­li­can has led in more polls than he has trailed, though by a 2:1 ratio, which has con­verged to a national lead of just under a point. As I’ve noted before, and will note again, the national polls are to the right of the state polls, and state polls have his­tor­i­cally been more accu­rate than their national coun­ter­parts. Nonethe­less, the President’s national poll posi­tion is now 2.5 points behind Pres­i­dent George W. Bush’s on this date eight years ago, and 9.5 points below his posi­tion four years ago.

Since the first Pres­i­den­tial debate, Obama and Rom­ney have been effec­tively tied in their favor­a­bil­ity polls.

For the third week, the national polls sug­gest that Obama might lose the upcom­ing election.

As of yes­ter­day, Intrade had Obama at 63, up a point from last week.

Over­all, things on the national level remain sketchy for the President.

Early Vot­ing

Early vot­ing is now avail­able in all states. The green states are accept­ing absen­tee bal­lots, but not early in-​​person bal­lots, while the gold states are accept­ing both.

Six­teen states allow in-​​person vot­ing today; of the bat­tle­ground states, this includes Col­orado, Florida, Iowa, Nevada, North Car­olina, Ohio, and Wis­con­sin. Michi­gan and Vir­ginia don’t offer early in-​​person voting.

North Carolina’s mail vot­ing has favored Repub­li­cans by a two-​​to-​​one ratio, but their early in-​​voting has favored Democ­rats by about the same ratio. Both forms have had roughly the same num­ber of votes, 150,000 each. That’s about eight per­cent of the 2008 total.

Florida’s mail vot­ing has been closer, with Repub­li­cans lead­ing Democ­rats by four points.

The Elec­toral College

Here’s what the Elec­toral Col­lege looks like, based on cur­rent polling data:

Here are the key states, from red­dest to bluest:

  • North Car­olina was polled three times in the past week. Ras­mussen has Rom­ney up by six, Pub­lic Pol­icy Polling sees a tie, and Grove found Obama up by three. Because this is indica­tive of a momen­tum shift, but it’s prob­a­bly too lit­tle, too late, North Car­olina moves to “Leans Rom­ney”. Intraders give Rom­ney an 80 per­cent chance of car­ry­ing the Tarheel State, unchanged from last week. Nate Sil­ver gives Rom­ney an 81 per­cent chance here.
  • Florida got polled four times in the past week. Angus Reid and Gravis found Rom­ney ahead by five points and one point, respec­tively, while Mell­man sees the race as tied and Grove saw a two point lead for the Pres­i­dent. Account­ing for house bias, Rom­ney still has a small polling lead. Florida stays “Likely Rom­ney”, based on the con­sis­tency of pro-​​Romney polls in the Sun­shine State, the trend in Romney’s direc­tion, and the sys­temic advan­tage Rom­ney already enjoys. Florid­i­ans have been able to vote for nearly a month. Intraders give Rom­ney a 72 per­cent chance of pick­ing up Florida’s 29 elec­toral votes, up eight from last week. Nate’s model is less con­fi­dent, giv­ing Rom­ney “only” a 65 per­cent chance. Were it not for the sys­temic advan­tage, I’d agree with Nate’s num­ber. With that advan­tage, the Intrade mar­ket looks closer to the truth to me.
  • Vir­ginia was polled six times this past week. Wen­zel Strate­gies, Ras­mussen, and Fox News all find Rom­ney ahead by two points. Mell­man sees Obama ahead by one, while Pub­lic Pol­icy Polling’s two polls came away with Obama up by two and five points. The more recent has the big­ger lead. All in all, it’s still close enough to keep Vir­ginia a “Tossup”, though it looks a lit­tle bet­ter for Obama this week than it did last week. Vir­gini­ans have been able to vote for quite a while now. The Intrade mar­ket has Rom­ney at 58 per­cent, up seven from a week ago. Nate’s model dis­agrees, giv­ing Obama a 54 per­cent chance. I think Nate’s is the more likely number.
  • Col­orado had five polls pub­lished this week. Ras­mussen was the only one to see a Rom­ney lead, at four points. NBC/​Marist saw the race tied. Keat­ing Research, Grove, and Pub­lic Pol­icy Polling all found Obama up by three. Col­orado stays a “Tossup”, though shaded bluer than last week. Intraders agree with me; the Col­orado mar­ket has Obama at 51 per­cent, up a point from last week. Nate’s model is more bull­ish on the Pres­i­dent, peg­ging him at 57 per­cent. I think real­ity lies some­where between the two num­bers, at 54.
  • New Hamp­shire had four polls pub­lished this week. Amer­i­can Research Group and Ras­mussen, both Republican-​​leaning, show Rom­ney ahead by two points. UNH and Lake Research Part­ners have Obama up by nine and three points, respec­tively. UNH looks like an out­lier, so I dis­count it some­what. Account­ing for house bias, ARG and Ras­mussen end up at about the breakeven point. On bal­ance, it’s just enough to push the Gran­ite State one col­umn to the left. New Hamp­shire now “Leans Obama”. On Intrade, Obama has the lead at 57 per­cent, down five from last week. Nate’s model has Obama at 69 per­cent. I’m not quite as bull­ish as Nate is, but I’d put New Hamp­shire at 60 for Obama.
  • Iowa was polled twice this week. Ras­mussen saw the race tied, while Pub­lic Pol­icy Polling saw Obama up by two. Look­ing at this week alone would put Iowa in an Obama +1 posi­tion, but the big­ger pic­ture sug­gests this is a lit­tle too low. There’s not enough data in two polls to poll Iowa into “Tossup”, so Iowa remains “Leans Obama”, a posi­tion it has held dur­ing the entire time bal­lots have been able to be cast. That said, this is one to watch over the next sev­eral days. On Intrade, Obama has the lead here, with a 65 per­cent chance, up two from last week. Nate’s model is very opti­mistic for Obama, putting him at 68 per­cent. I think they’re both too opti­mistic; I have Iowa at 60.
  • Ohio was polled nine times this past week. Nobody saw a Rom­ney lead. Gravis, Angus Reid, Suf­folk, and Ras­mussen all saw the race tied. Pub­lic Pol­icy Polling, Quin­nip­iac, Sur­veyUSA, Time/​SRBI, and Lake Research Part­ners found Obama ahead by one, five, three, five, and two points, respec­tively. Account­ing for house bias Ohio is a lit­tle bluer than last week, and is more com­fort­ably in “Leans Obama” ter­ri­tory. Buck­eyes have been able to vote for a month now. The Intrade mar­ket gives the edge to Obama, at 64 per­cent, up two from a week ago. Nate’s model says Obama has a 75 per­cent chance. Mine says 66, right on the bor­der of “Likely Obama”.
  • Nevada got polled four times this week. Amer­i­can Research Group, Ras­mussen, NBC/​Marist, and Pub­lic Pol­icy Polling saw Obama ahead by two, two, three, and two points, respec­tively. Unlike last week, this week’s num­bers all point in one direc­tion. The Sil­ver State moves to “Likely Obama”. Intraders are really bull­ish for the Pres­i­dent this week; he is trad­ing at 80 per­cent, up 11 points from a week ago. Nate is almost as bull­ish, at 79 per­cent. I’m not quite that san­guine, but I have Nevada at 70.
  • Wis­con­sin got polled twice this week. Angus Reid and Pub­lic Pol­icy Polling show Obama up by five and six points, respec­tively. Even account­ing for house bias, this is still a com­fort­able lead for the Pres­i­dent at this stage of the elec­tion cycle. Last time I said we needed to keep an eye out, because polls were some­what sug­ges­tive of a tight­en­ing race. With the newest round of polls, the pic­ture looks clearer. I’m mov­ing Wis­con­sin to “Likely Obama”, given the lack of move­ment toward Rom­ney and the short time remain­ing for Rom­ney to pull off a vic­tory here. Intraders remain con­fi­dent in Obama here, trad­ing him at 69 per­cent, up one from a week ago. Nate con­sid­ers Wis­con­sin likely Obama, at 86 per­cent. I split the two, hav­ing Obama at 75.
  • Penn­syl­va­nia was polled four times this week. Angus Reid, Gravis, Muh­len­berg, and Ras­mussen have Obama up by nine, three, five, and five points, respec­tively. With so lit­tle time remain­ing in the race, and Rom­ney hav­ing trailed in all polls not con­ducted by Susque­hanna for the entire year, the Repub­li­cans’ brief hope here has fully faded. Penn­syl­va­nia drops off the list after this install­ment. The mar­kets at Intrade agree with this, putting the Pres­i­dent at 79 per­cent, down four from a week ago. Nate’s model has Obama at 94 per­cent, which is only a hair more opti­mistic than my 90 percent.
  • Michi­gan was polled twice this week. Angus Reid showed Obama up by nine, while FMW/​Baydoun had Obama up by only a half point. Bay­doun has con­sis­tently been sev­eral points to the right of the con­sen­sus, despite being con­tracted by the Demo­c­ra­tic Party; there’s no rea­son to believe oth­er­wise this time. Rom­ney hasn’t led in a poll since August (Bay­doun), and the recent con­sen­sus has had Obama ahead by a con­sis­tent five points. Like Penn­syl­va­nia, Michi­gan drops off the list after this install­ment. At Intrade, Obama is trad­ing at 91 per­cent here, up six from last week. Nate’s model has Obama at 98 per­cent, slightly higher than my 95 percent.

Three states shifted one col­umn to the left this week. North Car­olina moved back from “Likely Rom­ney” to “Leans Rom­ney”, which makes sense when you con­sider that it has been on the cusp between the two, and the model is more sen­si­tive this close to elec­tion day. New Hamp­shire moved from “Tossup” back to “Leans Obama”, giv­ing the Pres­i­dent four more elec­toral votes. Nevada moved one more col­umn to the left. Based on the model, Obama has a prob­a­ble 281 elec­toral votes, up four from last week. It’s been nearly four con­sec­u­tive months in which Obama could lose all tossups and still stay in the White House.

Con­clu­sion

The national polls still show the race effec­tively tied, while the state tal­lies con­tinue to indi­cate a com­fort­able Obama lead. His­tor­i­cally, the state-​​by-​​state tal­lies have been more accu­rate than the national ones, which leads me to believe that Obama still has the edge.

What if I’m wrong, and the national num­bers are right? If so, then we should shift the states about two points to the right. Rom­ney would pick up Vir­ginia, Col­orado, New Hamp­shire, and Iowa, giv­ing him 267 elec­toral votes. Rom­ney would then need three more elec­toral votes. In this sce­nario, Ohio would still lean to Obama based on the polls, but would be Romney’s most likely pickup. While the sys­temic forces in Ohio aren’t as strong as they are in Florida, they could be enough to push Ohio to Rom­ney if the national num­bers are right. This sce­nario is less likely today than it was a week ago. Alter­na­tively, Rom­ney could pick up Nevada or Wis­con­sin, but those appear to be out of reach as well.

If I had to pre­dict an Elec­toral Col­lege result based on the model, which puts more empha­sis on the state polls, I’d shift things a hair from last week. Both tossups, which last week were in Romney’s col­umn, would now be Obama’s. This would give Obama 303, and Rom­ney 235.

How do you feel about these pre­dic­tions? Do you dif­fer on them? If so, how, where, and why?

Related arti­cles