Polls

Posts filed under Polls

2010Thu21Oct

Senate Projection for October 21

Permalink | Shortlink | Comments (5) | TrackbackFiled in PoliticsTags: Democrats, Elections, Polls, Republicans, Senate

With the latest batch of Senate polling results, Neil Stevens of Likely voter has updated his Senate Projection. According to his latest model, he projects R+7, with a 3% chance of Republicans reclaiming the majority (R+10 or greater).

The problem I have with the Likely Voter projection model is two-fold: one, it factors in races that are, for all intents and purposes, already decided; and two, it seems to assume a normal distribution. If a projection were performed in which seemingly non-competitive races were removed, and the ANOVA based solely on the actually competitive races, intuition tells me that such a projection (especially with a right-leaning distribution) would have to center around R+8, if not +9. To wit:

Introducing Uncertainty Based on Non-competitive Races

Range of Possible Outcomes

First, let's set the table:

Seats not up for election: 63 (23R, 40D)
Seats up for election: 37 (18R, 19D)
Bondary of Possible Results: R-18 to R+19

Now, let's add in some realistic boundaries to those results.

Range of Realistic Outcomes

Lower Boundary of Realistic Outcomes

All 18 R-held seats are 100% likely R. (R+0)
The following D-held seats are 100% likely R: AR, IN, ND, PA (R+4)

It is logical to conclude that, at this point, any projection that shows anything less than R+4 is just not consistent with reality.

Upper Boundary of Realistic Outcomes

The following D-held seats are 100% likely D: HI, MD, NY, OR, VT (R+14)

It is logical to conclude that, at this point, any projection that shows anything more than R+14 is just not consistent with reality.

Range of Realistic Outcomes

So, at this point, the range of realistic outcomes is R+4 to R+14. Anything outside of these numbers should be considered 0% likely.

Range of Likely Outcomes

The following D-held seats are 90% likely R: CO, WI (R+6, lower)
The following D-held seats are 90% likely D: DE, NY (s) (R+12, upper)

So, at this point, the range of likely outcomes is R+6 to R+12. Anything outside of this range should be considered unlikely.

Likely Voter's current projection distribution curve has a mean of R+7, and R+5 - R+8 accounts for 77% of all outcomes (R+7 22.6%, R+6 22%, R+5/R+8 35.4%). If I assume that Likely voter's probability curve is normally distributed, then, IMHO, the mean simply must be shifted too far left. There is just no possible way that R+5 has 18% probability. I'd say, at the absolute upper end, it has 5-10% probability. Balancing the 90% Likely R pickups against the 90% Likely D holds lowers the probability even further.

So, just using back-of-mental-napkin calculations, I would say:

<R+5: 0% likely
R+5: 5% likely
R+6 - R+12: 90% likely
R+13: 5% likely
>R+13: 0% likely

Analysis of Actually Competitive Races

The eventual outcome will be determined entirely by the results of six races: CA, CT, IL, NV, WA, and WV.

Two or three weeks ago, I would have rated those races as follows:

Lean-R: The following D-held seats are 55% likely: IL, WV
Toss-Up: The following D-held seats are 50% likely: NV, WA
Lean-D: The following D-held seats are 45% likely: CA, CT

However, things have shifted a bit; I would now rate these races as follows:

Lean-Likely-R: The following D-held seats are 60% likely: WV
Lean-R: The following D-held seats are 55% likely: IL, NV
Toss-Up: The following D-held seats are 50% likely: CA, WA
Lean-Likely-D: The following D-held seats are 40% likely: CT

As you can see, aside from CT (which, to be honest, I am close to writing off as a potential Republican pick-up), all of the competitive races have shifted in the Republicans' favor. I put together a quick Monte Carlo simulation of my own, and here are the results:

So, my model projects a mean +9 seat gain for Republicans, and a 40.1% chance that Republicans will regain control of the Senate (a gain of +10 or more seats). Results:

n = 10,000
μ = 5.2
σ = 1.4
max = +13
min = +5
+8 - +10 = 73.1%
10+ = 40.1%

At first blush, these numbers appear to me to be more realistic, given the current state of the races in play (and not in play).

Evaluating the Normal Distribution Model

It seems that the Likely Voter projection model is based upon the assumption that the outcomes of competitive races will be normally distributed. I wouldn't expect a normal distribution for these outcomes, even in a "normal" election year - but especially not in a "wave" year.

Just as the outcome distribution of competitive races was biased toward the Democrats in 2006 and 2008, I fully expect the distribution to be biased toward the Republicans in 2010. This bias is due primarily to two factors that are not easily accounted for through pre-election polling: the enthusiasm gap and shifts in party affiliation.

In short, pollsters simply don't have a reliable means of estimating the breakdown of voter turnout, and it is entirely likely that they will tend to err on the side of a conservative estimation of the shift from 2006/2008 to 2010.

In a later post, I will examine some of these factors in each of the six competitive races.

2010Thu21Oct

Is Sestak Closing on Toomey in PA?

Permalink | Shortlink | Comments (1) | TrackbackFiled in PoliticsTags: Democrats, Elections, Media Bias, Polls, Republicans, Senate

The latest news out of the Pennsylvania Senate race between Pat Toomey and Joe Sestak is that a few recent polls have shown Toomey's 10-point lead evaporate. First, Quinnipiac showed Toomey up only 48-46. Then PPP showed Sestak up 45-46. Now Morning Call shows the race tied at 43-43. So what gives? Is this race, that Rasmussen still lists as "Solid GOP", actually in trouble?

In a word: no.

To explain why, we'll need to look more closely at the crosstabs of these polls. But first, so you don't have to take my word for it, read what Jim Geraghty has to say over at NRO's The Campaign Spot.

Quinnipiac

Now, onto the analysis of the polls. For the sake of expediency, and since their crosstabs are available in the poll report, I'll focus on Quinnipiac. Specifically, I'll look at the party-affiliation breakdown for the Senate race, and for Obama's job approval.

Poll Results

Topline

Toomey: 46%
Sestak: 48%
Don't Know/No Answer: 5%

Senate

	D	R	I	Total
Toomey	7%	88%	56%	48%
Sestak	89%	8%	35%	46%
DK/NA	4%	3%	9%	5%

Obama Job Approval

	D	R	I	Total
Approve	85%	11%	30%	44%
Disapprove	13%	87%	64%	53%
DK/NA	2%	2%	6%	3%

Analysis

The poll doesn't indicate its party-affiliation weight values, but based on the above polls, I calculate that this weighting is as follows:

D: 38.0%
R: 34.5%
I: 27.5%

And there's the problem with this poll: this party-affiliation weighting bears little resemblance to reality. Jim Geraghty's piece linked above does a great job of explaining how these numbers are completely inconsistent with the electorate. But to prove the point, I'll look at some comparisons (previous-election party affiliation breakdowns taken from the Geraghty post).

Obama Job Approval

First, I'll examine jjust one change from the previous Quinnipiac poll to this one:

President Obama gets a negative 44 – 53 percent job approval rating, compared to a negative 40 – 56 percent September 22.

Note that overall, Obama's approval numbers continue to decrease, not increase. Yet somehow, this poll (miraculously) discovered a seven-point swing in Obama's favor since the previous month. That, alone, is enough to raise questions about the validity of the poll's topline. Interestingly, adjusting the party-affiliation weighting from 38.0%D / 34.5%R / 27.5%I to 32%D / 38%R / 30%I returns Obama's Approval/Disapproval numbers to 40% Approve / 56% Disapprove, and result in a 52% - 42% Toomey lead over Sestak (which is essentially right where the race has been for some time).

Democrat Best-Case Scenario: 2008

Next, I'll adjust the party affiliation weighting from the above numbers to the 2008 election numbers. In 2008, the turnout was 44% Democrat, 37% Republican, and 17% Independent. These numbers, which clearly represent not only a best-case scenario, but also an absolute pipe dream, result in a 48% - 45% Sestak lead over Toomey.

Get that? In a pipe-dream scenario, the Democrat would only be leading this race by 3%.

Democrat Second-Best-Case Scenario: 2006

Since the 2008 results are clearly out of reach, let's examine the 2006 results, in which the turnout was 43% Democrat, 38% Republican, and 19% Independent. These numbers, which represent a huge Democrat midterm election (again, something that will not be repeated in 2010), result in a 48% - 47% Sestak lead over Toomey.

So, once again, a pipe-dream scenario results in the Democrat leading this race by only 1%.

Bad News for Democrats: 1994

Since 2010 is clearly a Republican wave year, let's examine the poll results adjusted for 1994 turnout in the state, which was 39% Democrat, 41% Republican, and 20% Independent. These numbers result in a 50% - 45% Toomey lead over Sestak.

The problem for Sestak is that even the clearly skewed Quinnipiac party-affiliation weighting shows a lower Democrat turnout in 2010.

The even bigger problem for Sestak is that not only is the Democrat vote suppressed, but also the Independent vote is breaking 2-to-1 in favor of Toomey, and the Independent vote is highly motivated (by similar 2-to-1 ratios, Independents disapprove of Obama's job performance, disapprove of Obama's handling of the economy, prefer their Senator to oppose Obama's agenda, would prefer the Senate to be controlled by Republicans, and believe that Toomey rather than Sestak shares their personal values; also, 82% of Independents are dissatisfied/angry with the way government works).

Charting The Results

For comparison, here are the results of the above analyses:

	D	R	I	Toomey	Sestak
Quinnipiac	38.0%	34.5%	27.5%	48%	46%
Obama 40% Approval	32%	38%	30%	52%	42%
2008 Turnout	44%	37%	17%	45%	48%
2006 Turnout	43%	38%	19%	47%	48%
1994 Turnout	39%	41%	20%	50%	45%

Pennsylvania Gubernatorial Race

Perhaps the worst news yet for the Sestak campaign is that the Pennsylvania Senate race isn't the only statewide election this year. Pennsylvania also has a gubernatorial race, and that race has exhibited a very steady, 10-point lead for the Republican candidate.

Pennsylvania voters are not likely to switch parties between Gubernatorial and Senate candidates, and vice versa. Thus, if the sudden tightening of the Senate race is real, it should translate into a similar tightening in the Gubernatorial race. Unfortunately for Sestak, no such tightening exists.

Conclusion

Much ado about nothing. Make of it what you will, but the conclusion that Sestak is leading Toomey - or that he has even closed the gap - simply doesn't withstand a reality check.

By all appearances, Republican Pat Toomey will win the Senate race by 5-10 points over Joe Sestak.