Reëlection Watch: May 19, 2012

Today marks the sec­ond semi-​​monthly edi­tion of Reëlec­tion Watch. This time, I’m going to show another model on hand­i­cap­ping the race, using the Cook Par­ti­san Vot­ing Index (PVI).

So, how are things going for the Pres­i­dent lately?

National Polls

Obama’s approval/​disapproval rat­ing in the Real Clear Pol­i­tics aver­age con­tin­ues to sit mighty close to even. While it was a +0.6 point spread two weeks ago, and +0.7 now, these changes are suf­fi­ciently small as to be attrib­ut­able to ran­dom noise. If you look at the graph, you’ll see declin­ing oscil­la­tions over time, sug­gest­ing that we’re pretty much at sta­sis now. As I said last time, we should be look­ing for long-​​term trend sig­nals in the noise, but those need at least two months of data to be con­sid­ered sig­nal, rather than noise. The Right Track/​Wrong Track polls also remain pretty much the same. Last time it grew a lit­tle, and this time it shrank a lit­tle. The mid– to upper-​​twenties of neg­a­tive spread seems to be the con­ver­gence zone here.

Even Congress’s spread, which tight­ened for a few months, seems to have sta­bi­lized around −64 points. The generic Con­gres­sional bal­lot has had more volatil­ity than the other poll com­pi­la­tions, but I’ve come to believe that it’s more a func­tion of the low fre­quency of those polls than any par­tic­u­lar trends. That is, there’s enough noise that it’s essen­tially drown­ing out any signal.

As of yes­ter­day, Intrade had Obama at a 57.4 per­cent chance of reëlec­tion, down 2.5 points from last time. It appears that this is a slow trend of deflat­ing odds, but it’s not clear what’s behind this trend.

Things are still mostly sta­ble in a some­what pro-​​Obama state, but with a trend mov­ing toward a closer race.

The Econ­omy

Unem­ploy­ment data are essen­tially unchanged over the past two weeks. We’ll get more data in a cou­ple more weeks.

In terms of oil, West Texas Inter­me­di­ate crude is trad­ing at about $91, down 13 per­cent in the last two weeks, while Brent crude (rep­re­sent­ing the Euro­pean mar­ket) has fallen almost 11 per­cent to about $107. The drop in oil prices is unusual for this time of year, and it’s hard to say what’s behind it. It’s good news, how­ever, in terms of sup­port­ing more eco­nomic activ­ity as we head into summer…and that is good news for Obama head­ing into the fall. I had expected gaso­line prices to rise in response to last month’s oil price jump, but it appears that the recent sharp drop in crude prices has held gaso­line to a gen­tle decrease instead. The national aver­age is $3.71 per gal­lon, more than 20 cents below the same week last year.

Nat­ural gas prices have risen from their bot­tom at two dol­lars per mil­lion British Ther­mal Units to about $2.50. This is still half the price of nat­ural gas this time last year.

In gen­eral, things have been mov­ing in a pos­i­tive direc­tion over the past two weeks. But we still need to wait until our next install­ment to see how unem­ploy­ment looks.

The Elec­toral College

Last time, I showed you the mar­gins from the 2008 elec­tion, and com­pared them to the recent polls in the same states, with the intent of illus­trat­ing the dif­fer­ences between the national trends and the state trends. This time, I’m tak­ing the same notion from a slightly dif­fer­ent perspective.

The Cook Par­ti­san Vot­ing Index (PVI) for states is cal­cu­lated based on aver­ag­ing the results of the last two Pres­i­den­tial elec­tions. In par­tic­u­lar, it aver­ages the mar­gins of the Demo­c­ra­tic and Repub­li­can can­di­dates in each state, rel­a­tive to the national pop­u­lar vote mar­gin. So, if the Demo­c­ra­tic can­di­date received three per­cent more votes in 2004 in a par­tic­u­lar state than the national aver­age, and five per­cent more than the national aver­age in 2008, the state’s PVI would be D+4.

Since Pres­i­dent George W. Bush received two per­cent more votes nation­ally than did Sen­a­tor John Kerry (D-​​MA), this would have meant that the hypo­thet­i­cal state described above would have had Kerry tak­ing the state with a four-​​point mar­gin. This is because Kerry would have had three more per­cent­age points of votes, which would have meant that Bush would have had three fewer per­cent­age points (this pre­sumes an insignif­i­cant impact of the other par­ties; nat­u­rally, the cal­cu­la­tions are a bit dif­fer­ent if other can­di­dates receive sig­nif­i­cant num­bers of votes). That totals a six-​​point shift from the national aver­age, total­ing a four-​​point mar­gin for Kerry.

To reverse the process for a pre­dic­tion of the indi­vid­ual state results, take the national poll aver­age, and add the PVI to it. Since cur­rent national polling runs even, this essen­tially means look­ing strictly at PVI.

I sep­a­rated the two elec­tions, and cre­ated PVIs for 2004 and 2008. Then I cal­cu­lated the dif­fer­ence between the two, and ran a lin­ear extrap­o­la­tion based on that difference.

I read­ily admit that this is a crude analy­sis. The fac­tors that led a par­tic­u­lar state to move on the PVI scale are numer­ous, and may not have con­tin­ued. For exam­ple, Hawaii almost cer­tainly increased so much because Obama was born there. The above table is merely meant to be a rough guide to add some perspective.

That said, there are some inter­est­ing pos­si­bil­i­ties that come from this pro­jec­tion. Min­nesota, which has long sup­ported Demo­c­ra­tic Pres­i­den­tial can­di­dates, has been shift­ing right­ward. Could it be closer to a tossup than we’ve been assum­ing? Mon­tana has had a sim­i­lar left­ward shift.

If we take this PVI pro­jec­tion as a pre­dic­tion, the out­come looks like this:

Obama’s mar­gin of 1.7 points would mean that (assum­ing the pro­jected PVI is accu­rate), he’d take all states from North Car­olina down, giv­ing him 299 elec­toral votes.

For Rom­ney to win in that break­down, he’d have to take all of the states up to, and includ­ing, Penn­syl­va­nia, which would give him 276 elec­toral votes. This would require a national mar­gin of 1.2 points (my spread­sheet has the addi­tional sig­nif­i­cant dig­its). It’s within reach, but cer­tainly not a gimme.

With the PVI trend in hand, I’ll soon be com­par­ing it to the states with appre­cia­ble polling. This should give a view to the cred­i­bil­ity of PVI trend­ing, which we can use to apply to some of the less heav­ily polled states as well.

In the mean­time, let’s see what the cur­rent Elec­toral Col­lege looks like, based on cur­rent polling data:

Things have moved a bit left in the past two weeks. Here’s what’s changed since last time, from most likely to go Romney’s way to most likely to go Obama’s way:

• South Car­olina and Geor­gia are con­sid­ered “Leans Rom­ney” by RCP. I think that’s too gen­er­ous, par­tic­u­larly for South Car­olina, whose PVI moved to the right from 2004 to 2008. Even Geor­gia, which moved one PVI point to the left from 2004 to 2008, would still be at R+6 if it moved another point this time. I just don’t see either of these states being in play at all.
• Mon­tana is now con­sid­ered by RCP to be a “Leans Rom­ney” instead of “Likely Rom­ney” state. With the most recent Ras­mussen poll giv­ing Rom­ney a seven-​​point lead, I’m inclined to agree, par­tic­u­larly since Montana’s PVI shifted over four points to the left from 2004 to 2008.
• Ari­zona is still called a tossup by RCP. The home state effect makes Arizona’s three-​​point PVI shift to the right from 2004 to 2008 sus­pect as a trend this time around. But it’s still a pretty con­ser­v­a­tive state. We’ll have to wait and see what num­bers show up later.
• Mis­souri got another new Ras­mussen poll, giv­ing Rom­ney a mere three point lead. The polls would put the Show Me State into tossup ter­ri­tory for sure. On the other hand, the pro­jected PVI is R+5, which sug­gests that per­haps it’s near the bor­der between “Tossup” and “Leans Romney”.
• North Car­olina remains solidly in tossup ter­ri­tory. The Pub­lic Pol­icy Polling and Ras­mussen polls each show Obama and Rom­ney, respec­tively, as the leader. Ras­mussen shows a big­ger lead for Rom­ney than last time, but it’s hard to con­clude that any move­ment over the past two weeks is sig­nal, rather than noise. We could be see­ing some impact from the after­math of Amend­ment 1, though. North Car­olina has also been mov­ing to the left over the past sev­eral years, giv­ing a pro­jected PVI of R+2. So it’s a bit on the Rom­ney side, but not strongly so.
• Florida, had two new polls, one from Quin­nip­iac and one from Suf­folk. They strad­dle the even mark, leav­ing Florida a pure tossup from a polling per­spec­tive. On the PVI side, Florida has been inch­ing to the right, with a pro­jected PVI of R+3. This gives Rom­ney a slight edge.
• Ohio had three new polls since last time. Pur­ple Strate­gies, Pub­lic Pol­icy Polling, and Quin­nip­iac all give the edge to Obama, by five, seven, and one point, respec­tively. That’s more Obama than in Florida, but not quite enough to leave “Tossup” land for “Leans Obama”. It’s pur­ple, par­tic­u­larly when fac­tor­ing in a pro­jected PVI of R+3.
• Nevada hasn’t had any polls since my last install­ment. The PVI pro­jec­tion is EVEN. That said, Rasmussen’s most recent poll gives Obama an eight-​​point edge. It’s just not enough to make me push the (Nate?) Sil­ver State into “Leans Obama”.
• Vir­ginia got a new poll from the Wash­ing­ton Post; the seven-​​point Obama lead lines up with the recent Pub­lic Pol­icy Polling result. Virginia’s pro­jected PVI is R+4, which sug­gests to me that it’s prob­a­bly going to go pur­ple. I’m putting Vir­ginia barely on the “Tossup” side of “Leans Obama”.
• Penn­syl­va­nia had one new Quin­nip­iac poll, which shows no real change over the past cou­ple of months. With the pro­jected PVI of D+1, Penn­syl­va­nia remains (barely) “Leans Obama”.
• Col­orado got a new poll from Pur­ple Strate­gies, show­ing the two can­di­dates tied, though the pro­jected PVI is D+3. I’m leav­ing it “Leans Obama”, and wait­ing for another poll to see if Pur­ple Strate­gies is right to color the state purple.
• New Hamp­shire has a recent poll from Pub­lic Pol­icy Polling that gives Obama a 12-​​point lead. The pro­jected PVI is EVEN, but for now I’m putting New Hamp­shire in the “Leans Obama” column.
• Min­nesota has one recent poll from Sur­veyUSA, show­ing a 14-​​point lead for Obama. The pro­jected PVI is EVEN, though. For now, I’m leav­ing Min­nesota at “Leans Obama”, since it’s pretty much the same story as New Hampshire.
• New Mex­ico is solidly in “Leans Obama” ter­ri­tory, with polls lin­ing up with the pro­jected PVI of D+7.

Based on this month’s polling, Obama needs 20 elec­toral votes (down from 34 last time) out of the 107 in the tossup group. Nearly any pair of tossup states, or Florida on its own, would do the trick.*

Con­clu­sion

The more data I col­lect, the closer the elec­tion seems. That feels right to me intu­itively. If you’ve been read­ing my Reëlec­tion Watches over the past year, you know I believe we have a rel­a­tively weak incum­bent run­ning against a rel­a­tively weak chal­lenger. This is a sit­u­a­tion for which we have few exam­ples in his­tory upon which to draw.

For what it’s worth…on this day four years ago, Obama had a three-​​point lead over McCain, though that quickly nar­rowed to a sta­tis­ti­cal tie.

Obama remains in a good posi­tion rel­a­tive to Rom­ney, but there’s still plenty of oppor­tu­nity for that to change between now and November.

*Note: in an ear­lier ver­sion of this arti­cle, I erro­neously left Wis­con­sin in the “Leans Obama” cat­e­gory in the bar graph, which led to the incor­rect con­clu­sion that Obama needs only ten elec­toral votes. Wis­con­sin is still a “Tossup”.

Related arti­cles

• Jere­miah Wright and Kenya birthers could reëlect Obama and bring back Speaker Pelosi (the​hill​.com)
• Six Months Out, Pres­i­dent Obama’s Cam­paign Still Faces Stiff Head­winds (swamp​land​.time​.com)