Showing posts with label electoral college adjustments. Show all posts
Showing posts with label electoral college adjustments. Show all posts

Tuesday, August 19, 2008

Is Rasmussen's Inclusion of "Leaners" Affecting the Electoral College Outlook Now? An Update

It has been about three weeks since FHQ last checked in on the effect the early July decision by Rasmussen to report "leaners" in their polling releases had on our electoral college projections. At the time, only Ohio switched from favoring Obama to favoring McCain based on which version (with leaners or without) was used. In the period since though, there have been 27 new Rasmussen polls in 27 states. For 18 of those, that means the second (or third in the case of Minnesota) poll since the switch. The point of the comparison between the two types of numbers initially was to account for the discrepancy in the comparison of with leaners polls to previous without leaners polls. It just wasn't an apples to apples comparison.

With multiple "leaners" polls now out in 18 states, that really isn't the concern anymore. There are still nine states that had their first "with leaners" results reported since the beginning of August (and 13 others that have yet to have had their first with leaners poll conducted), but this is now primarily an exercise in examining how the electoral college projection would change -- and not a critique of the wholesale switch over to using the "leaners" numbers that took place in the aftermath of the change. We have the data, why not look at it? The more information we have from a "with leaners"/"without leaners" perspective, the more likely we are to begin seeing differences in the electoral college projections.

What we saw initially was that the leaners were breaking for McCain nearly across the board, but that around the time of Obama's trip abroad, that began to switch. And since that point, McCain's advantage in the "with leaners" numbers has almost been cut in half (from +1.08 to just +0.64 now). [Since the table is getting so large, I've decided to just simply append it to the end of the post instead of breaking up the text with such a mammoth figure. The new data are below the re-labelling of the columns.] Regardless, McCain continues to maintain an advantage over Obama with the leaners in these polls. And in an election that looks like it could come down to who is swinging the most independents, that's an important distinction.

But let's look at that more closely, focusing on the more recent polls (the new additions this time). Of those 27 states, eight are red states, eleven are blue states and the remaining eight are toss ups. In nine of the eleven blue states, the leaners are going for Obama (Though, it should be noted that in five of those nine, the leaners margin is exactly the same as the without leaners margin.), and in five of the red states, the leaners move in McCain's direction. So McCain in red states and Obama in blue. No real surprise there.

In the toss up states though, there are some differences. McCain holds slight advantages in 5 of those eight toss up states. The leaners broke for him in 60% of those cases (Missouri and Montana being the exceptions.). The picture for Obama was different. Of the three toss ups that favored the Illinois senator, the leaners moved toward him in one (Colorado), against him in another (Michigan) and broke even in the third (Nevada). That list of states was the very same as the one I discussed yesterday in the post about the benefits McCain could gain from tapping Mitt Romney as his running mate. Those three plus all the states in various shades of red add up to 271 electoral votes. And it is a very small consolation that Obama gains leaners in Colorado yet has the overall lead in the poll go to McCain. Nor is it beneficial to break even in Nevada when the overall poll favors McCain -- a switch from the month prior. In other words, having leaners break for him in Missouri and Montana is about all Obama can hang his hat on in this instance.

But what about the electoral college projection? Well, no matter which measure you use -- with or without leaners -- the projection (as a function of our weighted average) remains the same: 298-240 in favor of Obama. In fact, it is interesting that some of these leaner gains are cancelling each other out when subsequent polls are released. That has happened in both Minnesota and Wisconsin. Also, Nevada and Oregon have shown the same margins in both versions of the polls across two releases now.

One additional pattern that we can glean from these Rasmussen polls is the line up of states that are likely to have new polling data released in the next week. If everything holds to form, there should be new data in Arkansas, California, Florida, New Hampshire, New Mexico, Ohio and Pennsylvania in the near future. The last five of those are certainly worth keeping an eye on. All are toss ups with the exception of New Mexico. Given how other western states favoring Obama (Colorado and Nevada) have fared lately, it will be interesting to see if New Mexico follows suit.

Rasmussen Polls Since w/Leaners Distinction was Added (7/9/08)*
StateDatew/o Leaners
w/Leaners
Change
Undecideds Drop
Missouri
7/7+5
+50
-10
New Jersey
7/7+5
+3+2
-4
Illinois
7/8+13
+11+2
-6**
North Dakota
7/80
+1+1
-7**
Wisconsin
7/8+13
+10+3-6
Louisiana
7/9+20
+19+1
-2
South Dakota
7/9+4
+40
-4
Washington7/9+9
+8+1-6
Iowa
7/10+10
+100
-9
Michigan
7/10+8
+80
-5
Minnesota
7/10+18
+17+1
0
Kansas
7/14+20
+23+3
-9
North Carolina
7/15+3
+30
-5
Oregon
7/15+9
+90
-5
Nevada
7/16+2
+20
-5
Virginia
7/160
+1
+1
-6
Alaska7/17+5+5
0
-7
Arkansas7/17+10
+13+3-4
Georgia7/17+9
+11+2
+1
Maine
7/17+10
+8+2
-2
Colorado
7/21+7
+3+4-13
Ohio
7/21+6
+10+4
-7
Florida
7/22+1
+2+1-7
Minnesota
7/22+12
+13+1--***
New Hampshire
7/23+6
+4+2-5
Pennsylvania
7/23+5
+6+1-8
New Mexico
7/24+5
+6+1-6
California
7/24+12
+10+2-6
StateDatew/o Leaners
w/Leaners
Change
Undecideds Drop
Mississippi
7/28+11
+12+1-2
Nebraska
7/28+18
+19+1-3
Kentucky
7/29+10
+9+1-9
Montana
7/29+1
0+1-3
Alaska
7/30+5
+6+1--***
Arizona
7/30+16
+19+3-6
Texas
7/30+9
+8+1-7
Alabama
7/31+18
+20+2-9
Connecticut
7/31+15
+13+2-6
New Jersey
8/4+8
+10+2--***
New York
8/4+20
+19+1-2
Massachusetts
8/5+15
+16+1-6
Wisconsin
8/5+4
+7+3--***
Washington
8/6+12
+120
--***
Iowa
8/7+5
+50--***
Michigan
8/7+7
+4+3--***
Missouri
8/7+7
+6+1--***
Oregon
8/7+10
+100
--***
Kansas
8/11+15
+14+1--***
Nevada
8/11
+3
+30
--***
Illinois
8/12+15
+150
--***
Maine
8/12+13
+14+1--***
Virginia
8/12+1
+1+2--***
Colorado
8/13
+2
+1+1--***
Minnesota
8/13
+4
+40
--***
North Carolina
8/13+4
+6+2--***
Georgia
8/14+7
+9+2--***
Avg. Change+0.64-5.57
*The "with leaners" distinction was added to reports that were released beginning on 7/9/08. The date on which these polls were conducted (The ones that these releases were based on) stretches back to 7/7/08.
**Rasmussen has only conducted one poll in these states. Therefore, the difference was taken from between the with and without leaner numbers within the same poll in these cases.

***Previous poll had been taken after "with leaners" change had been made.



Recent Posts:
The New Ohio Poll and McCain's VP Choice

The Electoral College Map (8/17/08)

Which States are Underpolled in the Presidential Race?

Thursday, August 7, 2008

What Would Happen If...

...FHQ dropped some of the older state polls that may not be as reflective of the current (or if not current, recent) situation in the presidential race?

I've been asked several times by folks commenting here at FHQ and colleagues here at UGA if I had any plans to phase out any of the Super Tuesday period polls. My standard answer has been no, but once we hit the conventions and get past Labor Day, the traditional general election campaign will get underway and the dynamics of the race will likely change as well. I would expect increased polling and greater total attention from the public/electorate. Whether the latter means "better" responses from survey respondents is up for debate, but in any event we will likely have more representative polling emerging in September, vis a vis the vote in November, than we would with the polls that came out in February or March. Am I permanently dropping those polls from these analyses? No, this is simply a speculative look at how the landscape would appear in their absence.

Fine, what are you dropping then? This, as it turns out, was not an easy decision, though, it may look that way once I've explained. Some electoral college projections use just the most recent poll or the most recent several polls (two or three). In the case of FiveThirtyEight.com, they have used a decay function (a half life equation) to gradually phase out older polls. This is perhaps a good middle ground approach, but I don't want to be accused of aping what is being done over there. Despite that, I looked into where that cut off was on average. In most states, the oldest poll used ranges from mid-April to early May. I considered doing a similar cut off, but opted instead to cut things off at the beginning of the general election campaign, which I placed at the beginning of June. Yeah, Clinton wasn't officially out until June 3-7, but the Rules and Bylaws Committee's decision all but sealed her fate on the final day of May with their Florida/Michigan ruling.

The point here is to accomplish a couple of things. For starters, removing past and potentially inaccurate (at least outdated) polling may prove beneficial. Secondly, this removes the Clinton factor from the equation by focusing on polls conducted just during the general election race. The main criticism I'll get for this is that it is going to skew things in the favor of Obama because the polling cut off point sets the beginning at a point that encompasses his "bounce" during June.

Well, we've set the rules, so how does the map look? I should note now that this is not an official FHQ update of its electoral college analysis (Actually, the color scheme of the map is different so as to serve as a preventative measure on that front. Democratic blues will now be shades of purple and Republican reds will be altered to shades of orange.). This is merely speculative, but at the same time is something of a trial balloon. As I said, the dynamics of the race will change following the conventions, and I plan on officially revisiting our methodology at that time to see if there need to be any changes made. The method used here may be that change, if there is one, or it may not. With that said, to the map!
[Click Map to Enlarge]

If we compare this map to the most recent electoral college map I posted just yesterday, you'll see that there are twelve states that change categories (ie: from McCain lean to strong McCain) and a third of those change sides from Republican to Democratic or vice versa. The result is that Obama has a net gain of 22 electoral votes. Indiana, Montana and Virginia (27 electoral votes) move into Obama's column while Nevada (5 electoral votes) switches to favoring McCain. When each candidate's prospective electoral votes are broken down into categories, the distribution isn't all that different from before. The strong Obama category still holds -- by a large margin -- the most electoral votes, but lean (74D - 81R) and toss up (56D - 64R) electoral vote tallies are largely the same across the two parties.

Looking at the individual states, we see that Michigan, New Hampshire and Pennsylvania move out of the toss up distinction and become more solid for Obama. The movement of those 42 electoral votes is as important as the movement of those formerly Republican toss ups to the Democratic side. Aside from the states that switch from one partisan side to the other, these three states moving further into Obama's side are among the most important shifts. This is a good point to look at this in terms of the rankings we started doing here a few electoral college posts ago. The first thing to note is that Ohio is the new (and this is the last time this will be in quotation marks) "victory line." The Buckeye state is state where each candidate passes (or would pass) the 270 electoral vote threshold to claim victory in the election. The partisan line -- where the states shift from being Democratic to being Republican -- has now, with just the state polls from June on, been pushed further away from the victory line. That is indicative of the gains Obama makes when the cut off point for the polls is shifted. One additional thing to note about the victory line is that Ohio is the first toss up state for Obama. In other words, Obama is nearly to 270 electoral votes with just his lean and strong states. He needs just one of those toss up states to surpass that barrier (with the exception of Montana which would put him just three votes short). Colorado, Indiana or Virginia could be substituted for Ohio if the Buckeye state and the other three toss ups slipped into McCain territory.

The Electoral College State Rankings
HI-4
WA-11
NH-4
ND-3
TN-11
VT-3
MN-10
OH-20*
SD-3
KY-8
RI-4
NJ-15
CO-9
AK-3
WY-3
MD-10
WI-10
VA-13
GA-15
LA-9
NY-31
DE-3
MT-3
TX-34
AL-9
MA-12
OR-7
IN-11
WV-5
ID-4
CT-7
PA-21
NV-5
SC-8
KS-6
CA-55
NM-5
FL-27
AR-6
NE-5
ME-4
IA-7
MO-11
AZ-10
OK-7
IL-21
MI-17
NC-15
MS-6
UT-5
*Ohio is the state where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election.

As I've said, this is, for the time being, simply a glimpse into what things would look like if I decided to drop some of these "older" polls. The shift has been overwhelmingly toward Obama, but as we head into August this alternate version of reality will be something to keep tabs on. Does a shift toward McCain or away from Obama move the needle any more in this than it would in the normal version? The polling between now and the Democratic convention will help us to answer that question not to mention give us an idea of whether a change in methodology is even prudent. I will be keeping up with this now that I have the infrastructure in place, so that it will be an easier transition should we opt to go in that direction following the conventions (and their bounces).


Recent Posts:
The Electoral College Map (8/6/08)

Did Obama or McCain Win July?

About Those Rules: What Obama's New Florida/Michigan Stance Means for 2012 and Beyond

Sunday, July 27, 2008

An Update on the Rasmussen "Leaners" and a Look at How They Affect the Electoral College

We documented earlier this week that there may be a problem with Rasmussen's decision to report two different polling results with each release and have everyone use the newer and different, "with leaners" data when reporting it secondhand or using them in an electoral college analysis. Comparing a poll from Georgia in June that had no inclusion of leaners with a poll from July that did is problematic at the very least. Below is an updated look at the polls that Rasmussen has released since the "with leaners" distinction began appearing in press releases following July 4.

Rasmussen Polls Since w/Leaners Distinction was Added (7/9/08)*
StateDatew/o Leaners
w/Leaners
Change
Undecideds Drop
Missouri
7/7+5
+50
-10
New Jersey
7/7+5
+3+2
-4
Illinois
7/8+13
+11+2
-6**
North Dakota
7/80
+1+1
-7**
Wisconsin
7/8+13
+10+3-6
Louisiana
7/9+20
+19+1
-2
South Dakota
7/9+4
+40
-4
Washington7/9+9
+8+1-6
Iowa
7/10+10
+100
-9
Michigan
7/10+8
+80
-5
Minnesota
7/10+18
+17+1
0
Kansas
7/14+20
+23+3
-9
North Carolina
7/15+3
+30
-5
Oregon
7/15+9
+90
-5
Nevada
7/16+2
+20
-5
Virginia
7/160
+1
+1
-6
Alaska7/17+5+5
0
-7
Arkansas7/17+10
+13+3-4
Georgia7/17+9
+11+2
+1
Maine
7/17+10
+8+2
-2
Colorado
7/21+7
+3+4-13
Ohio
7/21+6
+10+4
-7
Florida
7/22+1
+2+1-7
Minnesota
7/22+12
+13+1--***
New Hampshire
7/23+6
+4+2-5
Pennsylvania
7/23+5
+6+1-8
New Mexico
7/24+5
+6+1-6
California
7/24+12
+10+2-6
Avg. Change+1.00-5.7
*The "with leaners" distinction was added to reports that were released beginning on 7/9/08. The date on which these polls were conducted (The ones that these releases were based on) stretches back to 7/7/08.
**Rasmussen has only conducted one poll in these states. Therefore, the difference was taken from between the with and without leaner numbers within the same poll in these cases.

***Previous poll had been taken after "with leaners" change had been made.

Overall then, there is still a bias toward McCain in cases where "with leaners" data is replacing w the data that does not include them. As I've said, this isn't going to hae major ramifications at the outset, but over time, once a series of polls have been conducted, that collective difference could have a huge impact on what FHQ and others are doing with their electoral college analyses. In the only instance where a state has been polled by Rasmussen more than once since the switch, the results cancelled each other out. The first poll gave McCain a one point edge with leaners while the poll that followed two weeks later saw Obama with the same advantage. Such a result likely won't be the norm, however. In Ohio, for example, using one version or the other shifted the Buckeye state's 20 electoral votes from Obama to McCain and back again. And though that is the only case of a change thus far, that list is likely to grow in the future when new polling data is released.
[Click Map to Enlarge]

So looking at the map, we see that using the without leaners data -- the type that was used prior to July 9 -- shifts the map in Obama's direction with the electoral vote tally reverting to the 298-240 margin that had been in place for most of July until the new Ohio poll surfaced last week. That four point change in Ohio triggered just a 0.34 point change in our weighted average, but that is all that is necessary to move 20 vital electoral votes from one candidate's column to the other's.

It could be said that all were arguing over is whether Ohio is a toss up favoring McCain or Obama when the real point is that the Buckeye state is a toss up. Period. It doesn't really favor anyone. That's a valid point. However, there are differences in other states -- obviously -- but one poll has not been enough to affect change in those cases. Further polling, though, may bring about that change.

Note: I don't know how regularly I'm going to update these figures, but I have built them into my spreadsheet and I'm tracking the differences. When and if any changes occur -- major or minor -- I will bring them to the attention to our readers.


Recent Posts:
The Electoral College Map (7/27/08)

Guam: Oh Well, So Much for Frontloading the General Election

A Game of Vice Presidential Chicken