Wednesday, September 10, 2008

The Electoral College Map (9/11/08)

Ugh, internet connectivity issues are the worst. [So are blog posts that start off by making excuses.] Sorry for the delay.

Anyway, the polling yesterday was interesting. Again, we saw the continuation of the post-convention trends into which we have settled...for the most part. [There are always exceptions.] Most importantly though, we saw the list of battleground states continue to contract. First, Alaska, then Montana and now North Dakota slipped out of the toss up McCain category and into a safer, and more traditionally Republican, position for McCain. If anything, the Palin selection seems to have taken some of those traditionally red states that have been surprisingly competitive in the trial-heat polls off the board. The biggest convention bounces have been in already Republican states.

And the result is that the electoral map has suddenly taken on a more traditional look. Now it's all about Florida and the typical collection of Midwestern states. Sure, you can add in Colorado and Virginia, but when this election was discussed early on as a "map-changing" election, those saying it had more than two states in mind. The more this election resembles a traditional election, the more that favors the GOP and McCain. First of all, it is an alignment of states under which Republican candidates (read: George W. Bush) have fared well recently. But also, it means that McCain won't have to play defense in states where, traditionally, he wouldn't have to. Now, it may come to pass that some of these states inch back toward competitiveness once the post-convention environment settles down, but I wouldn't necessarily count on that.

New Polls (Sept. 10)
StatePollMargin
(With Leaners/ Without Leaners)
Alaska
Rasmussen
+31
Michigan
CNN
+3
Missouri
CNN
+4
New Hampshire
CNN
+5
New Mexico
Rasmussen
+2
North Carolina
Public Policy Polling
+4
North DakotaRasmussen
+14
Pennsylvania
Strategic Vision
+2
Virginia
CNN
+6
West Virginia
Blankenship
+5

With that said, the ten new polls out in ten states have once again shaken things up in FHQ's weighted average, something that, as the race progressed through July and August, we just didn't see that much of. There just wasn't that much volatility. New polls would emerge but just did not break enough from the established pattern of polling in most states to make that much of a difference. That has changed now in the wake of the VP selections and conventions.

Changes (Sept. 10)
StateBeforeAfter
AlaskaMcCain leanStrong McCain
New MexicoObama leanToss Up Obama
North DakotaToss Up McCainMcCain lean
West VirginiaStrong McCainMcCain lean

That leaves us with a rather long list of changes -- at least comparatively. Alaska, we can understand. The Palin selection has taken the Last Frontier out of the mix. And North Dakota, along with Montana figures in to a similar calculation. Those states, then, are terribly surprising. New Mexico and West Virginia? Those results are worthy of a closer look.

Polling in New Mexico has been similar to what we have witnessed in Ohio: erratic. Now, Ohio has stretched out over an 18 point range from +10 for McCain to +8 for Obama, which on its face, appears to give a slight edge to McCain (...if you were to average those two results in isolation). Indeed, that's what we see on the map below. But New Mexico is not that different. Across all the polling conducted in the Land of Enchantment since Super Tuesday, the results have ranged from +16 for Obama to +6 for McCain. [Just in the last week we've seen a +14 for Obama -- pre-GOP convention -- and a +2 for McCain.] If you apply the same, simplistic methodology from the Ohio example to New Mexico, you find that you end up with a state that is competitive, but favors Obama all the same. True to what we've have seen over the course of the summer, New Mexico appears to be a state that is on the line between a toss up and a state leaning more heavily toward Obama. In each case, though -- in Ohio and New Mexico -- we still have a pretty good picture of what's going on given the existing polling data throughout the race. [As an aside, that's the beauty of not just looking at the most recent poll(s). You have this established idea of how the state is going to break in November. It isn't as volatile, yet it does change if trends are persistent. Our measure, then, is skeptical of change until that shift fundamentally alters the equilibrium in a state.]

West Virginia also offers an eye catching poll. First of all, there's a new poll in West Virginia! There are only a few states (Delaware, Hawaii and Vermont) that have had fewer surveys conducted than the Mountain state. So when we get one, you take notice, if only to see that John McCain is up by a ton...or not. The flip side is that you get a result that is surprising not only for its closeness, but that it is breaking with the trend seen in other similarly Republican states. [And I say that based on the other two presidential elections this decade. In West Virginia, you have a state that has switched from reliably Democratic to reliably Republican on the presidential level. And that shift has been rather rapid.] Granted, what we have here is a piece to a very incomplete puzzle; a 1000 piece job with 600 pieces missing. Still, in the limited polling are privy to, West Virginia is on a trajectory that moves it toward competitiveness. +18 for McCain in February turns into +8 in June and +5 in September. Does that mean the Mountain state is Obama's in November? No, but the fact that West Virginia is moving in a direction counter to other 2004 Republican states is worth noting. And if you're Obama and you're already sinking money into Ohio, Pennsylvania and Virginia, why not throw a little towards neighbor to all three, West Virginia. From an advertising standpoint, you get the northern part of the state by buying time in the Pittsburgh market. But that's only one aspect of it. Given the thrashing Obama took in the late May primary, there's likely less of an Obama field operation in the state which would potentially put him behind in the ground game. Then again, the McCain campaign likely thought it wouldn't have to defend West Virginia and may have to kickstart its own ground game there. We'll have to see. The closeness in this poll will probably trigger additional polling to determine whether this was an outlier or the continuation of a trend toward competitiveness.
[Click Map to Enlarge]

There is some shuffling then, but no electoral votes change sides. I'm not going to belabor the point here because I'd like to spend a bit of time discussing the movement on the Electoral College Spectrum. [And there's been a lot of it this week.] Suffice it to say, Obama still has a lead in the electoral college, but, as I've noted, the number of paths to victory are shrinking. The Kerry states with Colorado, Iowa and New Mexico are all the Illinois senator really needs to break 270, but bringing in Florida or Ohio is not totally out of the question. Nor is losing those two and the two western neighbors above to McCain.

The Electoral College Spectrum*
HI-4
(7)**
WA-11
(165)
CO-9***
(269/278)
ND-3
(160)
KS-6
(64)
VT-3
(10)
MN-10
(175)
NH-4***
(273/269)
SC-8
(157)
NE-5
(58)
RI-4
(14)
DE-3
(178)
NV-5
(278/265)
SD-3
(149)
AR-6
(53)
IL-21
(35)
OR-7
(185)
OH-20
(298/260)
TX-34
(146)
TN-11
(47)
CT-7
(42)
NJ-15
(200)
VA-13
(311/240)
WV-5
(112)
ID-4
(36)
MD-10
(52)
IA-7
(207)
IN-11
(322/227)
GA-15
(107)
KY-8
(32)
ME-4
(56)
WI-10
(217)
FL-27
(349/216)
AK-3
(92)
AL-9
(24)
NY-31
(87)
NM-5
(222/321)
MO-11
(360/189)
MS-6
(89)
OK-7
(15)
CA-55
(142)
MI-17
(239/316)
NC-15
(375/178)
AZ-10
(83)
WY-3
(8)
MA-12
(154)
PA-21
(260/299)
MT-3
(163)
LA-9
(73)
UT-5
(5)
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, McCain won all the states up to and including Pennsylvania (all Obama's toss up states, but Michigan and New Mexico), he would have 299 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Obama's number is on the left and McCain's is on the right in italics.

***The line between Colorado and New Hampshire is the where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

The striking thing about the spectrum is that the list of "safer" McCain states now stretches into the middle column. That 163 electoral votes looks pretty solid for McCain in this current environment. And we can talk about West Virginia all we want, but when you look at it, the Mountain state is way down the rankings for Obama at this point, barely within the bounds of the "lean" distinction. At this point, given the current momentum of the race, the election looks like it is being waged in about eight states; that group from Michigan through Virginia. That is the perception at least. Are those other pink states out of the question for Obama? No, but given the circumstances, those are states that are traditionally Republican. And traditional Republican states seem to be aligning behind McCain-Palin. We have additional polling out from Quinnipiac this morning in Ohio and Florida that could change things (tune in tonight to see how that affects things), but getting additional information out of Indiana and Nevada will help us to get a better idea of what is happening in the 2004 Bush states. If they continue the trend, then that solidifies the impression that those states are moving away from Obama and the the battle is over the eight states I referenced above.

The Watch List*
StateSwitch
Georgiafrom McCain leanto Strong McCain
Minnesotafrom Obama leanto Strong Obama
Mississippifrom Strong McCainto McCain lean
Montanafrom McCain leanto Toss Up McCain
Nevadafrom Tieto Toss Up McCain/Obama
New Mexicofrom Toss Up Obamato Obama lean
North Carolinafrom Toss Up McCain
to McCain lean
North Dakotafrom McCain lean
to Toss Up McCain
Ohiofrom Toss Up McCain
to Toss Up Obama
Washingtonfrom Strong Obamato Obama lean
Wisconsinfrom Obama leanto Toss Up Obama
*Weighted Average within a fraction of a point of changing categories.

This round of polling adds few more states to the Watch List. So, after bottoming out over the weekend -- likely due in part to the lack of polling during the conventions -- the shake ups on the electoral map have augmented the list. New Mexico is now a toss up but continues to hover around the line between the toss up and lean categories. Like Montana, North Dakota is gone, but not necessarily forgotten in terms of being competitive. More polling is warranted in both cases for the foreseeable future. Virginia becomes the latest state to slip off the list, though. The CNN poll pushed it off the Watch List and into a firm, yet not unreachable, McCain toss up state. The list of states on the Watch that could change sides imminently has now dropped to two: Ohio and Nevada. And we have a new poll out in Ohio today. Now, where are those Nevada polls?


Recent Posts:
The Electoral College Map (9/10/08)

The Links (9/9/08): Advertising Call and Response

The Electoral College Map (9/9/08)

Tuesday, September 9, 2008

The Electoral College Map (9/10/08)

Now here's the bounce for McCain. But the question I posed yesterday was, how is the post-convention bounce distributed across states? Among the battleground states that had polling released yesterday and today, not so much. Among traditionally red states like Montana and North Carolina, however, the bounce appears to be quite significant. With eight new polls out today, we're starting get a clearer picture of the state of play in the race for the White House at this point in time.

New Polls (Sept. 9)
StatePollMargin
(With Leaners/ Without Leaners)
Florida
Public Policy Polling
+5
Maryland
Gonzales
+14
Oklahoma
Survey USA
+33
Michigan
Strategic Vision
+1
Montana
Rasmussen
+11
New Jersey
Fairleigh-Dickinson
+6
North Carolina
Survey USA
+20
Wisconsin
Strategic Vision
+3

But why does Montana shift categories and North Carolina doesn't? Good question. And it has to do with the number of polls conducted since Super Tuesday in each state. North Carolina has nearly eclipsed the 30 poll mark while Montana has had just a handful of surveys done. The result is a small N problem. The state with fewer polls is more susceptible to outliers than the state where the average is more established. That is why Montana jumped North Carolina, trading in toss up pink for McCain lean red in the process. Even though that 20 point margin is larger than the margin in Montana, it is an extreme outlier among the other polls in the Tar Heel state. There has not been a double digit margin in either direction in the state since all the way back in February. North Carolina may end up a McCain state in November, but this poll is out of step with the host of polls conducted thus far in the state.

Changes (Sept. 9)
StateBeforeAfter
MontanaToss Up McCainMcCain lean

Montana, then, reverts to its roots, handing the Republican candidate a heavier share of support. In the process, McCain augments his "safer" electoral votes by three. Obama still has nearly as many electoral votes stored away in his strong category as McCain does in his strong and lean categories. That hasn't changed. What has changed is that some McCain toss ups are shifting into more comfortable positions for the Arizona senator and all the while -- especially with the addition of Ohio yesterday -- has increased his toss up total as well. McCain's advantage over Obama in projected toss up electoral votes is now almost 50.

[Click Map to Enlarge]

What we are witnessing is a subtle shift toward McCain. The Arizona senator and his running mate have shored up some of the traditional Republican states, but other than Ohio, Obama is not yielding any states that are shaded in blue.

The Electoral College Spectrum*
HI-4
(7)**
WA-11
(165)
CO-9***
(269/278)
MT-3
(160)
KS-6
(64)
VT-3
(10)
MN-10
(175)
NH-4***
(273/269)
SC-8
(157)
NE-5
(58)
RI-4
(14)
DE-3
(178)
NV-5
(278/265)
AK-3
(149)
AR-6
(53)
IL-21
(35)
OR-7
(185)
OH-20
(298/260)
SD-3
(146)
TN-11
(47)
CT-7
(42)
NJ-15
(200)
VA-13
(311/240)
TX-34
(143)
ID-4
(36)
MD-10
(52)
IA-7
(207)
ND-3
(314/227)
GA-15
(109)
KY-8
(32)
ME-4
(56)
NM-5
(212)
IN-11
(325/224)
MS-6
(94)
AL-9
(24)
NY-31
(87)
WI-10
(222)
FL-27
(352/213)
WV-5
(88)
OK-7
(15)
CA-55
(142)
MI-17
(239/316)
MO-11
(363/186)
AZ-10
(83)
WY-3
(8)
MA-12
(154)
PA-21
(260/299)
NC-15
(378/175)
LA-9
(73)
UT-5
(5)
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, McCain won all the states up to and including Pennsylvania (all Obama's toss up states, but Michigan), he would have 299 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Obama's number is on the left and McCain's is on the right in italics.

***The line between Colorado and New Hampshire is the where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

And when the focus shifts to the Electoral College Spectrum, the thing that is striking is that there just isn't that much light blue anymore. Ohio is pink and Nevada remains tied. The real test for Obama, then, appears to be keeping Colorado and New Hampshire on his side of the partisan line. With Montana's shift the Arizona senator has now consolidated the right two columns on the spectrum, but that gets him only to within 110 electoral votes of 270. There are still a number of states in pink, but one has to wonder how long that will be the case. If the new trend is for traditional Republican states to begin shifting toward McCain, then we might expect Indiana and North Dakota to shift toward McCain.

The Watch List*
StateSwitch
Georgiafrom McCain leanto Strong McCain
Minnesotafrom Obama leanto Strong Obama
Mississippifrom Strong McCainto McCain lean
Montanafrom McCain leanto Toss Up McCain
Nevadafrom Tieto Toss Up McCain/Obama
North Carolinafrom Toss Up McCain
to McCain lean
Ohiofrom Toss Up McCain
to Toss Up Obama
Virginiafrom Toss Up McCainto Toss Up Obama
Washingtonfrom Strong Obamato Obama lean
Wisconsinfrom Obama leanto Toss Up Obama
*Weighted Average within a fraction of a point of changing categories.

With new polling, Montana and North Carolina join the Watch List. I don't know that we've heard the last of Montana as a toss up, but, like Georgia, it could be jettisoned by the Obama campaign in order to focus on Colorado, Nevada and New Hampshire, for example. North Carolina is now right up against the line between being a toss up and a lean state. If 20 point margins are the new trend in the Tar Heel state, then it won't take too long for the state to move into McCain lean territory. Given the polling history there, though, that seems a stretch.

As was the case yesterday, though, we are going to have to wait and see how this all plays out for the next week or so.


Recent Posts:
The Links (9/9/08): Advertising Call and Response

The Electoral College Map (9/9/08)

2008 vs. 2004: Glass is Half Empty/Glass is Half Full

The Links (9/9/08): Advertising Call and Response

Here at UGA we've had a discussion group on this presidential campaign going for the better part of two years. Of late, some of the Obama partisans in the group have been calling for the Obama campaign and the Democratic Party to strike back on the McCain campaign's emphasis on its maverick credentials. Well, we have officially moved into Rovian "attack your opponent's strengths" territory.

The McCain-Palin ticket is playing up their ability to shake up Washington:


And the Obama-Biden campaign is just calling it "more of the same:"


This is likely to be one of the major debates before the debates. Well, that and the continued discussion of who is the real agent of change. But the two are related, aren't they?


Recent Posts:
The Electoral College Map (9/9/08)

2008 vs. 2004: Glass is Half Empty/Glass is Half Full

The Electoral College Map (9/7/08)

Monday, September 8, 2008

The Electoral College Map (9/9/08)

And so it begins. Here is the first wave of state-level polling since the two parties' conventions. It is an interesting mix of results in a series of hotly contested states. But is there a bounce (...for either candidate)? The national polls have been volatile over the course of the last week, peaking for Obama early last week and currently favoring McCain. But how does that translate to the state level, or more importantly, does it?

New Polls (Sept. 8)
StatePollMargin
(With Leaners/ Without Leaners)
Colorado
Rasmussen/FOX
+3
Florida
Rasmussen/FOX
0
Michigan
Public Policy Polling
+1
Ohio
Rasmussen/FOX
+7
Pennsylvania
Rasmussen/FOX+2
Virginia
Survey USA
+2
Virginia
Rasmussen/FOX+2
Washington
Survey USA+4

In looking at these eight polls across seven battleground states (Well, six if you exclude Washington, which has been on the line between a strong Obama and Obama lean state throughout the summer. However, more polls like this will quickly put the state on a more McCain-friendly trajectory.), nothing really stands out as that different from the pre-convention period. Only Florida, Ohio and Washington are more than three points outside of where our weighted averages had them positioned. Everything else is closely clustered around those averages. The Washington poll from Survey USA is not all that surprising compared to the past polls of the Evergreen state from that firm. Survey USA has been the only polling agency to find Washington to be in the single digits over roughly the last two months. The other polls have placed the state in the 10-12 point range in Obama's favor. So, while this result differs from the recent history of polling in the state, that discrepancy may be agency-specific and not necessarily indicative of a trend.

The case is somewhat similar for Florida. Rasmussen's polling of the Sunshine state has been less responsive to what appears to have been a summertime move toward Obama in other polling of the state. What we have seen is a steady move in the Rasmussen polls in Florida. What was a 7 or 8 point McCain advantage in the state in late June gradually tracked down two points in August and a tie now. [Yes, that glosses over the blip that was the one point lead Obama had in the late July Rasmussen poll, but bear with me here.] Over the late June to September period there have been 17 polls in Florida and the median and modal values for that series of polls is 3 points to McCain (The raw average is 2.35 points in the Arizona senator's direction.). That Rasmussen now finds the state to be tied is something of a coup for the Obama campaign, especially considering the perceived bounce McCain got out of his convention last week.

Changes (Sept. 8)
StateBeforeAfter
Ohio
Toss Up Obama
Toss Up McCain

Ohio is somewhat quirky as well. I don't want to sound as if I'm trying to explain away McCain's gains, but Rasmussen has really been the only firm to show McCain as far ahead as 7 points over the last couple of months. In the same period we examined Florida in above, there were 13 polls in Ohio and the range of results was rather wide. However, the raw average of those polls would give Obama a slight .08 lead (and the median value was a 1 point Obama edge). Rasmussen's polls in Ohio represented the three most McCain-favorable results (10, 7 and 5 points). So, while the most recent poll of the Buckeye state from Rasmussen sends Ohio into Toss Up McCain territory, it is not without caveats. And as is the case with Florida and Washington, we'll have to see some more post-convention results from each to get a sense of where each state actually is.

[Click Map to Enlarge]

With that shift of Ohio's electoral votes, though, things suddenly seem a lot tighter in this race. The truth is, it always was close. However, that cushion of electoral college votes I've discussed in relation to Obama isn't as cushiony anymore. In other words, you don't want to use all your excess electoral votes now. The good news for Obama partisans given this first round of post-convention polling is that the national bounce McCain is getting isn't necessarily extending into the swing states. And that isn't particularly good news for the McCain folks. Let's let this play out for a week or more though and see where this takes though.

The Electoral College Spectrum*
HI-4
(7)**
WA-11
(165)
CO-9***
(269/278)
MO-11
(381/168)
KS-6
(64)
VT-3
(10)
MN-10
(175)
NH-4***
(273/269)
SC-8
(157)
NE-5
(58)
RI-4
(14)
DE-3
(178)
NV-5
(278/265)
AK-3
(149)
AR-6
(53)
IL-21
(35)
OR-7
(185)
OH-20
(298/260)
SD-3
(146)
TN-11
(47)
CT-7
(42)
NJ-15
(200)
VA-13
(311/240)
TX-34
(143)
ID-4
(36)
ME-4
(46)
IA-7
(207)
ND-3
(314/227)
GA-15
(109)
KY-8
(32)
MD-10
(56)
NM-5
(212)
MT-3
(317/224)
MS-6
(94)
AL-9
(24)
NY-31
(87)
WI-10
(222)
IN-11
(328/221)
WV-5
(88)
OK-7
(15)
CA-55
(142)
MI-17
(239/316)
NC-15
(343/210)
AZ-10
(83)
WY-3
(8)
MA-12
(154)
PA-21
(260/299)
FL-27
(370/195)
LA-9
(73)
UT-5
(5)
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, McCain won all the states up to and including Pennsylvania (all Obama's toss up states, but Michigan), he would have 299 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Obama's number is on the left and McCain's is on the right in italics.

***The line between Colorado and New Hampshire is the where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

Regardless, with Ohio switching and Nevada's electoral votes stuck in limbo -- at least in terms of FHQ's weighted average -- the 13 electoral vote margin is the smallest we've witnessed in quite a while. It is certainly the smallest margin during the general election campaign (post-Obama nomination). The divide has not been this small since McCain held a 270-268 electoral vote edge in mid-May. Beyond that, we are beginning to see a convergence of the partisan and victory lines on the Electoral College Spectrum. Nevada now serves as the partisan line -- where Democratic states turn to Republican states and that is nearly right up against the victory line in between Colorado and New Hampshire. So, whereas we had seen two or three light blue states in between those two lines, that has now been shrunk to nearly nothing. And that is indicative of this cushion we've been discussing here. Obama had that cushion, and thus more paths to victory, but it has largely disappeared here of late.

The Watch List*
StateSwitch
Georgiafrom McCain leanto Strong McCain
Minnesotafrom Obama leanto Strong Obama
Mississippifrom Strong McCainto McCain lean
Nevadafrom Tieto Toss Up McCain/Obama
Ohiofrom Toss Up McCain
to Toss Up Obama
Virginiafrom Toss Up McCainto Toss Up Obama
Washingtonfrom Strong Obamato Obama lean
Wisconsinfrom Obama leanto Toss Up Obama
*Weighted Average within a fraction of a point of changing categories.

The states to watch are still the same as they were before we got this new polling. The only change is that Ohio is now in the verge of switching from McCain to Obama instead of vice versa. The interesting thing is to contrast this Watch List to the one that would have existed four years ago in the Bush-Kerry race. An overwhelming majority of those states on the Watch four years were blue and were moving in Bush's direction. The distribution of states near the various lines between distinctions here is much more even. There are actually some red states that could potentially move as well. The real difference is that the 2008 Watch List on this date has fewer states which seems to indicate that there is a settling in this race that was not present in 2004. Granted, that could change in the coming weeks, but the movement toward Bush that we saw down the stretch in 2004 has not reappeared in the 2008 context.


Recent Posts:
2008 vs. 2004: Glass is Half Empty/Glass is Half Full

The Electoral College Map (9/7/08)

On to the Debates! -- And a Note on Compression

2008 vs. 2004: Glass is Half Empty/Glass is Half Full

If you're like I've been today, you've been clicking refresh every so often over at Pollster to see if any new state-level polls have emerged in this post-convention period. [There were several late-day releases, but I'll get to that in a bit when I post the first of the daily electoral college updates. I think we'll have enough polling volume that we'll be able to pull it off. At the end of the election, we'll have a much richer picture of what was happening over these last fifty-seven days.] It is one thing to answer the "convention bounce" question with national polls, but I've always been inclined to discover how this is working out on the state level. Is that national bounce that we've witnessed for McCain uniformly distributed across all states, or is it the case that McCain/Palin just fired up partisans in already-red states with their convention last week? I suspect the answer is somewhere in the middle, but I'd like to take an opportunity to cast an eye to the past for a guide in all this. As you'll recall, about a month ago, I examined how, using FHQ methodology, 2004 would have looked at the same, mid-August point of the campaign. The middle of August in 2004 was roughly two weeks after the conclusion of the Democratic convention, so we would expect to have seen Kerry in a somewhat advantageous position in the electoral college projection relative to Bush. And that's exactly what we saw. Kerry had essentially the same projected lead in the electoral college that Obama held in the same mid-August period.

However, given that mid-August was in the time after Kerry's convention, we'd expect to see a bounce for him as well, right? Well, since Kerry got basically no bounce (see the graph and point D04) out of his convention that isn't necessarily the case. The Massachusetts senator didn't get anything out of the convention in the national polls, but did he in the state polls? The thing that marked 2004 was the tight equilibrium that we saw across much of the campaign cycle. The Democratic convention didn't do anything to shake up the steady state of the race. In fact, as you'll see below, that state was held through and even past the Republican convention in 2004. September 8, 2004 was roughly a week after the conclusion of the GOP's convention and at that point there had not been any change to the distribution in the electoral college. Kerry, then, didn't get a bounce out of his convention, nor does it appear that he was hurt in any way by the modest bounce Bush received from his convention -- at least not in the week following the conclusion of the Republican convention.
[Click Map to Enlarge]

After the conventions then, Kerry still held just enough of an advantage in enough states to provide him with a 44 electoral vote edge over the incumbent president. But as I'm sure President Kerry will attest, that lead did not last. While the margin stayed the same the fundamentals of the race were changing. The states that shifted over the last 100 days of the race had begun, in most cases, their moves toward President Bush. New Mexico shifted from a Kerry lean to a toss up favoring Kerry. And though Florida continued to ever so slightly favor Kerry, the Sunshine state was basically a tie and was slowing inching toward the Republican nominee. The only other state that shifted sides was Iowa and the Hawkeye state actually moved slightly toward Kerry in the mid-August to early September period in 2004. Still, the lead was under three points and Bush was obviously able to swing just enough votes to pull out a victory in the state in November.

Colorado was also an interesting case over this period. The Centennial state, like Iowa, moved toward Kerry over this period, in the process shifting from an Bush lean to a toss up state. Why is that interesting? Well, obviously the trend is counter to what we might expect of a Bush state in the post-convention period, but Colorado also had a special measure on the ballot that fall. The measure, if supported by voters, would have distributed the state's electoral votes by congressional district in the manner that Maine and Nebraska do and would have taken effect for the 2004 electoral college session in Washington in December following the election. The measure was voted down, but in the context of this tightening, is an interesting footnote to the election.

So how does this enhance our understanding of what is happening in the current race for the White House. For that let's start by comparing the map above to the map from yesterday's electoral college projection update.
[Click Map to Enlarge]

There is quite a bit of overlap between the toss up states in each cycle. But 2008 has brought several atypical states into the mix. Indiana, Montana, North Carolina and North Dakota are all much closer than they were just four years ago. All still favor the Republican candidate, but are closer. States like Iowa, Minnesota, Oregon and Wisconsin that were toss ups four years ago are all leaning more heavily toward the Democrat in 2008. That has helped to provide Obama with a cushion that neither Kerry nor Gore enjoyed in either of the last two elections. Obama is ahead and doesn't need Florida to cross the 270 electoral vote threshold. And though the Illinois senator is slightly ahead in Ohio as of now, he could cede the Buckeye state to McCain and still eke out an eight electoral vote victory assuming McCain also inches ahead in currently tied Nevada. In fairness, we don't have even a partial picture of how the conventions are playing on the state level. [Well, we do now, but I'll get to that in a little while.]

The Electoral College Spectrum*
MA-12
(15)**
HI-4
(168)
WI-10
(264/284)
NC-15
(167)
KS-6
(45)
RI-4
(19)
WA-11
(179)
FL-27***
(291/274)
AZ-10
(152)
AK-3
(39)
NY-31
(50)
ME-4
(183)
NV-5
(296/247)
SC-8
(142)
OK-7
(36)
CT-7
(57)
MN-10
(193/355)
OH-20
(316/242)
GA-15
(134)
MT-3
(29)
VT-3
(60)
NM-5
(198/345)
MO-11
(327/222)
SD-3
(119)
ND-3
(26)
IL-21
(81)
OR-7
(205/340)
AR-6
(333/211)
LA-9
(116)
NE-5
(23)
MD-10
(91)
MI-17
(222/333)
WV-5
(338/205)
KY-8
(107)
ID-4
(18)
CA-55
(146)
NH-4
(226/316)
CO-9
(347/200)
IN-11
(99)
MS-6
(14)
NJ-15
(161)
IA-7
(233/312)
VA-13
(360/191)
AL-9
(88)
WY-3
(8)
DE-3
(164)
PA-21
(254/305)
TN-11
(178)
TX-34
(79)
UT-5
(5)
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, Bush won all the states up to and including New Hampshire (all Kerry's toss up states, but Michigan), he would have 323 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Kerry's number is on the left and Bush's is on the right in italics.

***Florida is the state where Kerry crosses (or Bush would cross) the 270 electoral vote threshold to win the presidential election. That state is referred to as the victory line
.

Both the Electoral College Spectrum and the Watch List of this period in 2004 underscore the precarious position in which Senator Kerry was at the time. His hold on Florida was the only thing keeping him above 270 and then it was only by a fraction. Plus the number of light blue toss up states had snaked all the way into the the second column of Kerry states. This was compounded by the fact that all the states that were near moving -- those on the Watch List below -- were predominantly Kerry states. And while many were on the line between toss up and lean, most had already moved into the toss up category and would end up staying there or in the case of Iowa and New Mexico, would move into Bush's column.

The Watch List*
StateSwitch
Californiafrom Strong Kerryto Kerry lean
Floridafrom Toss Up Kerry
to Toss Up Bush
Mainefrom Kerry leanto Toss Up Kerry
Marylandfrom Strong Kerryto Kerry lean
Michiganfrom Toss Up Kerryto Kerry lean
Minnesotafrom Toss Up Kerryto Kerry lean
Nevadafrom Toss Up Bush
to Toss Up Kerry
New Hampshire
from Toss Up Kerryto Kerry lean
New Mexicofrom Toss Up Kerryto Kerry lean
Oregonfrom Toss Up Kerryto Kerry lean
Tennessee
from Bush lean
to Toss Up Bush
*Weighted Average within a fraction of a point of changing categories.

So, glass is half empty, Obama is in the same position Kerry was in (in terms of an aggregation of electoral votes). But, glass is half full, Obama has built-in advantages that neither Kerry nor Gore before him held in their respective runs for the White House. Where the 2008 race stands now, though, depends on how these states begin breaking with the information the conventions of the last two weeks have added to the discussion.

We'll get to those shortly.


Recent Posts:
The Electoral College Map (9/7/08)

On to the Debates! -- And a Note on Compression

Presidential Primary Reform: Still Alive with the GOP?

Sunday, September 7, 2008

The Electoral College Map (9/7/08)

We have had just enough of a trickle of polls released over the last two weeks to have kept FHQ from putting up the same ol' map with no means of change. [And during a change election, too!] I would expect a steady stream to start coming out on Monday -- despite the fact that the recent trend has been away from weekend polling and especially releases -- and by the middle of the week we'll begin seeing a vastly larger number of polls. The conventions are over and state polling, like all the other polling, will kick into overdrive as this presidential race enters the home stretch.

New Polls (Sept. 3-7)
StatePollMargin
(With Leaners/ Without Leaners)
Alaska
Moore
+19
Indiana
Howey-Gauge
+2
North Carolina
Democracy Corps
+3

The trickle of polls during the latter half of the week included releases from a trio of McCain states. And that adds a nice symmetry to the week's polls. The first half of the week saw three polls from states favoring Obama. We have to be careful about how we treat the trends we see in each due to the timing of the polls. In other words, Indiana, for example, is close -- closer than one might think -- but it was done in the two days after the Democratic convention. Yes, that is during the time when Sarah Palin's selection was rolled out, but still, the caveat should be added. In the case of the North Carolina poll, it was conducted over the course of a week, the end of which overlapped with the first two days of the Democratic convention. There may, then, be some respondents in the poll who heard Hillary Clinton's speech, but that certainly would have been a late hour for polling firms to have been making their calls. Regardless, this is around where the Old North state has been for much of the summer -- right around the 3 or 4 point range.

Changes (Sept. 3-7)
StateBeforeAfter
AlaskaToss Up McCainMcCain lean

And that leaves us with Alaska, the home of the GOP vice presidential selection. The Last Frontier has been much closer in the polls than history would otherwise tell us. With the selection of Governor Sarah Palin, though, that closeness -- real or just simply in the polls -- seems to be disappearing quickly. Granted, this is just one poll, but this trend will likely continue. What else could come out about Palin to shift things back toward Obama, and even if it did, it seems that most people have made up their minds about her. On average, only 17% of respondents in recent favorablility polls which have added in the Alaska governor failed to view her either favorably or not. That is the lowest of any vice presidential nominee over the last three cycles. Palin also has, again on average, the highest favorable and unfavorable ratings among that group of VP selections. Even with all the bombshells thus far, perceptions appear to formed.

[Click Map to Enlarge]

So Alaska moves into the McCain lean area and will likely continue to move even further to the right of the Electoral College Spectrum, becoming even more intensely red. This may change some in the coming weeks, but, as of now, the underlying distribution of electoral votes is the same. There is an equal distribution of electoral votes between the McCain and Obama toss up and lean categories, but the real difference is between the strong categories, where Obama still holds a 70 electoral vote advantage. That is built on California and New York being firmly on Obama's side. But that is a large part of Obama's -- and any Democrat's -- electoral math. The truth remains that Obama is ahead but not irreversibly so.

The Electoral College Spectrum*
HI-4
(7)**
WA-11
(165)
CO-9***
(269/278)
MO-11
(381/168)
KS-6
(64)
VT-3
(10)
MN-10
(175)
NH-4***
(273/269)
SC-8
(157)
NE-5
(58)
RI-4
(14)
DE-3
(178)
OH-20
(293/265)
AK-3
(149)
AR-6
(53)
IL-21
(35)
OR-7
(185)
NV-5
(298/245)
SD-3
(146)
TN-11
(47)
CT-7
(42)
NJ-15
(200)
VA-13
(311/240)
TX-34
(143)
ID-4
(36)
ME-4
(46)
IA-7
(207)
ND-3
(314/227)
GA-15
(109)
KY-8
(32)
MD-10
(56)
NM-5
(212)
MT-3
(317/224)
MS-6
(94)
AL-9
(24)
NY-31
(87)
WI-10
(222)
IN-11
(328/221)
WV-5
(88)
OK-7
(15)
CA-55
(142)
MI-17
(239/316)
NC-15
(343/210)
AZ-10
(83)
WY-3
(8)
MA-12
(154)
PA-21
(260/299)
FL-27
(370/195)
LA-9
(73)
UT-5
(5)
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, McCain won all the states up to and including Pennsylvania (all Obama's toss up states, but Michigan), he would have 299 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Obama's number is on the left and McCain's is on the right in italics.

***The line between Colorado and New Hampshire is the where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.

We can talk about this cushion that Obama has based on the comparison between his and McCain's strong category electoral votes, but this election is still based on what is going to happen in those toss up states and to a large extent independent voters in them. The top five states in the Spectrum's middle column above are still the states where much of the action will take place. It is no coincidence then that Obama/Biden and McCain/Palin have been in Virginia and Ohio and Colorado. But the race certainly stretches beyond those boundaries. Florida, Indiana, Pennsylvania and Wisconsin are also targets. All four are toss up states as well with the exception of Wisconsin, which is, as you can see both above and below in the Watch List, on the line between being a lean or toss up state favoring Obama.

The Watch List*
StateSwitch
Georgiafrom McCain leanto Strong McCain
Minnesotafrom Obama leanto Strong Obama
Mississippifrom Strong McCainto McCain lean
Nevadafrom Tieto Toss Up McCain/Obama
Ohiofrom Toss Up Obamato Toss Up McCain
Virginiafrom Toss Up McCainto Toss Up Obama
Washingtonfrom Strong Obamato Obama lean
Wisconsinfrom Obama leanto Toss Up Obama
*Weighted Average within a fraction of a point of changing categories.

Once again, the Watch List shrinks by one this week. With Alaska moving off, and more firmly into the McCain side, the list is down to just eight states. And only three of those states are on the verge of switching partisan sides and just one additional state -- Wisconsin -- is near moving toward being more competitive. The remaining four are either in the high single digits or low double digits. This is one indication that the list of competitive states for these final two months of the campaign are solidifying. What was talked about early on as a map-changing election, then, has narrowed, as can be expected to some extent approaching election day, to the regular group of swing states with a few exceptions (Colorado, Indiana and Virginia to name a few). We will have to see in the coming weeks whether either campaign expands or contracts its operations, as was the case with Georgia recently. That will provide an even clearer indication of where the fight will take place moving forward.


Recent Posts:
On to the Debates! -- And a Note on Compression

Presidential Primary Reform: Still Alive with the GOP?

Why Attack the Community Organizer?

Friday, September 5, 2008

On to the Debates! -- And a Note on Compression

Let's not put the cart before the horse here, but this next two months is going to fly by. [And I thought after Labor Day, it was going to take forever to get to UGA's fall break the weekend before the election.] Think about it. Next week will be silly season in the polls as they readjust to the post-convention state of play. And then you have just two weeks until the first debate on September 26. A week ago that seemed far off, but all of a sudden, it's just three weeks away.

This really has been an unusual presidential election cycle from a timing standpoint. Primary season kicked off just three days into the new year and all the talk then was about how 2008 would be the longest general election campaign in history. It just didn't work out that way with the Democratic contest stretching into June. But now that we have been into general election mode since June, things don't look like they did in January. People are just now starting to really tune into the race and now it's not about how long the general election campaign will be, but how compressed it will be. From Tuesday at the Democratic convention to November 4 is just ten weeks.

How is that 10 weeks divided?
Democratic convention = 1 week
Republican convention/VP announcement = 1 week
Debates = 2.5 weeks (September 26-October 15)

That leaves just 5.5 weeks of actual campaigning. Now, I understand that the debates don't cause the campaign to shut down completely, but preparation time will factor in and the media's focus will shift just as quickly. [Hey, isn't this post titled, "On to the Debates!"? Who is shifting the focus here?] There are these next three weeks, the debates and then that leaves just under three weeks until election day. The span is not that different from four years ago, but with the VP announcements and conventions happening so close together it has only fed the perception of compression.*

So what will we be hearing these next three weeks? I'd imagine refined versions of what we have heard over the last two weeks. The Democrats will attempt to keep things focused on the economy and the GOP will make the case for their version of change with reminders of the importance of having the right person in charge in regard to the wars in Iraq and against terrorism. These next three weeks will be crucial to both campaigns as they hone their campaign themes heading into the conventions.

A few other things:
What was the general impression of McCain's speech last night? Sure the consensus seems to have formed around the idea that it was solid if unspectacular. Anyone differ with that assessment?

Did anyone catch Obama with Bill O'Reilly last night? I'm without FOX News Channel and haven't read too much about it today.

Also -- and this I'm sure is a shocker if not a teaser -- the Ivan Moore poll out of Alaska has shifted the Last Frontier into McCain lean territory. I think we'll continue to see it inch closer and closer to McCain/Palin as we approach election day. I'll have more Sunday when the updated Electoral College Map is up.


*Yes, this is very similar to the Seinfeld episode where Jerry and Elaine turn five more days in Florida with Jerry's parents into half a day once sleeping, trips to the airport, etc. are taken into account. Perhaps not as extreme, but similar.


Recent Posts:
Presidential Primary Reform: Still Alive with the GOP?

Why Attack the Community Organizer?

The Electoral College Map (9/3/08)

Thursday, September 4, 2008

Presidential Primary Reform: Still Alive with the GOP?

In the on again, off again world of politics nothing is ever dead. If you wait long enough, something may actually happen you previously thought impossible. Or to steal a line from last night's speeches, "If you don't give up you can't be defeated." [FHQ's mind is failing right now to accurately attribute that quote. I've seen a speech or two these last two weeks. If I'm lucky, one of our ever-loyal readers will come to my assistance. If not, I'll be called Joe Biden and my political career will be over. Such is life. Back to frontloading.] That's true in this case as well. I spent last week and the weekend railing against the Republican Party for once again failing to do anything regarding presidential primary reform (See here, here, here and here). Ah, but the proponents of reform within the party were not yet ready to let the cause die before 2012. While they did pass a plan to basically maintain the status quo for 2012 (Iowa, Nevada, New Hampshire and South Carolina go first and everyone else can go no earlier than the first Tuesday in March.), embedded within the new rules describing the formation of a commission similar to the one the Democrats pushed through at their convention last week.

This is actually fairly monumental. The fact that the GOP is allowing for the rules governing the selection of delegates and thus presidential nomination to be altered outside of the convention setting is a big step toward dealing with the frontloading issue. Frontloading is a problem that requires some flexibility from a rule-making body tasked with dealing with it. That the GOP could only confront frontloading and the primary process at their convention, deprived them of the ability to adapt if need be to the changes on the ground (like in Michigan and Florida in 2008). They now have that flexibility and can wait and see how large a contingent of states attempts to move and/or violate the rules for 2012. That's a low threshold and doesn't really confront reform, but I think the GOP will be resistent to anything but the status quo unless something like Florida or Michigan repeats itself in the lead up to 2012. As I've said, with only one party likely active during primary season in four years, the number of states seeking to move their primaries and caucuses to advantageous dates is likely to go down. If the number of states attempting to move (or balking at the idea of having to move back to March after jumping to February in 2008) is small and/or within the rules, the Republicans are likely to ride it out and wait for 2016 if they have to.

Having said that, if McCain loses in November, he won't have the sway over the rules-making powers within the party as he did ahead of this current convention. Remember, Ohio plan supporter and Ohio GOP chair, Bob Bennett was pointing the finger at McCain for the plan's failure to pass the party's rules committee. With that intervention removed, they may actually be able to pull off something meaningful (...if they are so inclined). But that is a huge hurdle removed from the process.

I think that both parties have to work together to make that meaningful though. If both parties can go to the states with a unified plan, it is much stronger than if they do it separately. Mixed messages regarding reforms gives states the excuse to fall back on the status quo, one where they really hold all the cards (hold the relative freedom to decide when they hold their contests).

It is interesting that this should come to light (at least my light) today. The Caucus just this morning had a post up examining the Mitt Romney in 2012 question. Now, let's do a quick exercise here. Let's assume that McCain fails to top Obama in November. [I know. Sorry GOPers. I'm thinking of the future here, though.] Let's further assume that both Romney and Huckabee run again. Yes, this ignores the possibility of Sarah Palin throwing her hat in the ring, but let's focus on the two known quantities -- in terms of presidential primaries -- for the time being. We can add Palin into the mix if you like in the comments section. Anway, which of the two does better under which system?

First, the status quo system: All other things equal, we would expect Huckabee to win in Iowa again and in South Carolina. Romney, due to his roots in Utah and in Massachusetts would likely have advantages in Nevada and New Hampshire. So the two are "tied" heading into Super Tuesday on March 6, 2012. [Yeah, this is kind of silly, but bear with me here. Only one political scientist has a Crystal Ball.] Conventional wisdom tells us that the candidate with the most money would have the advantage as Super Tuesday approaches. I would argue that that favors Romney in large part because of his personal wealth. But hey, we could have a caucus sitation like we had among Democrats in 2008. Both Romney and Huckabee did well in caucuses this time around, so that could be considered a wash. Though, it should be pointed out that Romney nearly swept the caucuses on February 5, 2008.

And a reformed system? Let's assume that Bob Bennett and the other reformers get their way and the Ohio plan becomes guiding rule behind the 2012 process. The Favored Four go first still, and break the same way as under the status quo system. The two emerge tied going into the small states primary. On the one hand, the fact that the smallest states are up next so as to nurture retail politics is something that plays to Huckabee. On the other, the fact that you have campaign in a series of states that, while not large, are not lumped into one geographic area. It is unlikely then that either candidate would sweep those states. Like in 2008 then, each would have to pick and choose their spots. Let's look at the map:
[Click Map to Enlarge]

The states in green are the ones to look at here. And that happens to be a lot of Western states, states where Romney did very well in 2008. He was not in the race for several, but he won caucuses in Maine, Montana, North Dakota, Alaska and Wyoming. Among the small states, Huckabee only did won in West Virginia and the Romney folks would argue that the Huckabee and McCain campaigns were in cahoots to prevent Romney from winning there. On the surface then, it would appear that Romney would have a real good start under either system. If Huckabee and Palin (or any other potential candidate in 2012) begins railing against such a system pre-emptively, we'll have a pretty good early indication of who might do well (...or who the party appears to be coalescing behind.)

*A tip of the cap to Don Means at NationalCaucus.org for bringing the story to my attention.


Recent Posts:
Why Attack the Community Organizer?

The Electoral College Map (9/3/08)

A Follow Up on the 50% Mark: The View from 2004