Toomey’s message is sounding tired and cookie cutter. Anyone with a computer knows the Republican talking points: Pelosi, Obama, Reid and (fill in the blank) are all the same.

Sestak has the mo, and Onorato is picking up steam at the right time. Could be a Dem sweep of statewide races, even as we lose 2 or 3 Congressional seats. So long as we hold on to Governor we can combat the Republican gerrymandering they do every time they control the redistricting process.

rplinpa=
Sestak has the mo? (Like Lieberman’s Joe-mentum?)

I think your mind is still inhaling the fumes from last week, or you are looking in the wrong direction.

Muhlenberg Daily tracking:
10/19 Sestak +3
10/20 Tie
10/21 Tie
10/22 Toomey +3
10/23 Toomey +3
10/24 Toomey +5

I’m voting for Onorato, but I don’t believe he has a shot. Despite his recent uptick, he (like Sestak) has already begun to backslide as the Republicans are making their final push.
I think the recent tightening of the polls woke up the Republicans a bit who thought they had this in the bag.
Statewide stats show we are getting blown away on absentee balloting. At this point, it appears that 3% to 5% of the vote is already locked it, with Republicans beating out Dems by 20%. If that’s the case, then the Republicans are going into Tuesday with close to 1% advantage (maybe 30,000 votes) statewide.

I do agree that Sestak will do a few % better than Onorato, due to Philly. However, Onorato needs to be within 5 points for Sestak to even hope of winning.

The Toomey campaign is wising up and saying we don’t want Palin’s endorsement.

Toomey doesn’t want a Palin endorsement.

Sestak doesn’t want an Obama endorsement.

If candidates hate the leaders of their political parties so much, why don’t they just run as independents instead?

David – I don’t know, you’re only using one poll. We need to see other polls.

All I know is the five thirty eight was giving Toomey a 92% chance of winning just a few weeks ago. As of yesterday, they reduced it to a 63% chance he’ll win. Sure, he’s still the favorite but when it goes from 92% down to 63% quickly, it shows which way the momentum is heading.

Ok, disregard the above…the graph apparently depends on how your mouse rolls over the chart. At one point I saw 63%, then again at 76% and now 82%. It depends just the way your mouse rolls over the graph. At it’s peak Toomey was 94.6% and now 82.9%. Momentum shifting, but still not enough.

What is killing the Dems is not Obama but Ed Rendell. If he could just go away for a while. then Sestak might have a chance. Onorato is too closely tied to the corrupt Rendell self-dealers for the absence of Ed to be of any help.

Sam-
You have to look at the right end of 538 the chart. Those are the most recent dates. The probability is based upon whatever polls were available at that snapshot in time. The higher odds for Sestak last spring were long before Toomey started up his campaign against Sestak, and against polls that are 6 months out of date. Also, as you get closer to the election, a 5-point gap in the polls gets harder to change.

The 538 forecast uses the weighted average of all the available polls (even taking into account known biases). So, if Rasmussen has the race at +4 Toomey, then it’s probably closer to Toomey +2 (similar move in opposite direction for Dem leaning pollsters).

Sestak definitely got some bump the other week, and some very good press, however, Toomey and the GOP have pushed back. Also, after two debates that were largely viewed as ties, the election remains a referendum on Obama.

Even though Muhlenberg is only one poll, it does come out every day. The Sestak people were tripping over themselves rushing out to quote it as evidence of Joe’s viability and momentum shift. A few newspapers over the weekend quoted the first Sestak +3 number as a “new poll”, despite Toomey +3 having come out since then, and Toomey +5 today.
This is an 8-point shift within a single pollster, and it suggests a widening, not tightening, of the race.
When you look at the polls, Sestak is consistently in the lower 40′s and Toomey is in the upper 40′s.
Sestak may get this within a few points, but that’s not the same as actually winning.

GOPHawk-
Rendell is hurting Onorato, and Obama is hurting Sestak.
Beyond that, Sestak hurt both himself and the party be refusing to participate in a joint campaign. Do you see ANY signs with Sestak’s name and another candidate? In 2006, we had Rendell/Casey and Rendell/Casey/Sestak signs all over the place.
I’ve seen signs with Toomey and the local GOP Congressional candidate, and I’m sure there are other pairings.

David – it’s actually a good thing that Sestak is not trying a joint campaign this year. 2006 was a very good year for Dems (the public sentiment was anti-GOP) so it made sense to team up for a joint campaign that year.

However, in 2010, the public sentiment is very anti-Dem so it doesn’t make sense to team up this year. Sestak’s best approach is to say go ahead and vote for Corbett but split your vote and vote for me. (As this decade has shown, it’s not a good thing when one party controls all the chambers. People will listen to the idea of vote splitting to keep checks and balances on power). Rendell’s approval rating is very low, Onorato is getting smashed in the polls. Sestak is faring better than Rendell/Onorato/Obama. It’s probably still not enough to win, but his best chance in 2010 was to distance himself from a joint campain.

Jerry-
As a matter of tactics, I think you happen to be wrong. For example, in 2006, the Sestak campaign’s field director refused to do joint lit-drops with other Dems on the ticket (particularly in St. Leg districts where Sestak wanted Republican votes, and didn’t want to tie himself to local Dems). I (and others) felt this strategy was hurtful to the party and the ticket, as no attempt was made by Sestak to convince supporters that the other Dems on the ticket shared the same values.
As a result, Bryan Lentz was the only St. Leg gain we had, and we had NO gains in 2008 because Sestak wouldn’t help use his (20 point) advantage to help others and free up his volunteers to work on other campaigns (or share his lists when he didn’t need volunteers).

This year, the part infrastructure could be doing grassroots work at the committee/precinct level to support the entire ticket. However, the lack of a coordinated effort wastes resources and causes duplication of effort. Why have two calls: “Vote for Sestak” and “Vote for Onorato” when you could have one call “Vote for Sestak and Onorato” and tie in the problems with Toomey and Corbett.
Also, we are in danger of losing the State House along with the Governorship. This means that the Republicans will get to gerrymander the Congressional and other districts.

So, the “Sestak – me first”, “split-ticket” strategy that you advocate is of course hypocritical to Sestak’s argument during the Primary that Dems needed Joe at the top, leading the ticket? (Leading to betrayal is more like it.)
If Sestak put principle over politics, he would be touting the Democratic bonafides of his fellow candidates and how he needed them in charge to prioritize the money he’d bring in from Washington.
If Sestak actually got elected, there would be plenty of difference between him and the rest of the Democratic party (see: Joe Lieberman and Ben Nelson for a sneak peek).

David: Glad you’re keeping up with the polls. That first mcall poll was obviously an outlier and anyone paying attention to the polls at large should have known that. The only other poll to have Sestak up was by PPP which is notoriously unreliable. Of course this race has tightened from the 7-10 point leads that Toomey had 3 weeks ago, but it doesn’t seem to be tightening enough for Sestak to pull it out. His Santorum ad may have worked against Specter, but this isn’t a primary.

EK-
The first poll may have reflected the momentary bump Democrats were seeing nationwide as Obama was hitting the trail and we were seeing a bit of Democratic wake up.
I think the race may 2-4 points to Toomey’s advantage. Sestak has shrunk the gap, but not closed it.

Hey David,

The Mulhenberg poll is a lot like Gallup. They just take the results of the first 500 people that respond, without regards to what political party they are. That’s why the numbers in these two polling organizations fluctuate so much on a daily basis.

Check out today’s poll. They now have Toomey ahead 48-40, but when you read further it says that 48% of the poll participants are registered Republicans while 43% of the poll participants are registered Democrats. If you sample 5% more Republicans, of course it’s going to skew the overall results more to Toomey.

The best polling organizations out there try to a get a weighted cross segment that mirrors that actual percentage of each registered voters from each party (Dems, Reps and Inds) that’s actually going to vote.

If more registered Republicans than registered Democrats in PA, come out to vote on Tuesday then of course Toomey will win by 8 (and probably even more). However, given that the Democrats have a big registered voting edge in PA, it’s very unlikely that 5% more of registered Republicans than registered Democrats are going to come out to vote on Tuesday.

That’s what makes this poll bad. In reality, there’s not that much daily fluctation in this race at all. It’s the method of polling that’s producing these wild swings in the results.

Steven, but didn’t David say earlier that it’s really a 2-4 point race? I think even Rasmussen’s most recent poll has the race at 4 points. 2-4 points might not seem like much, but it’s still a lot to overcome. Corbett said on TV last week that when Republicans win in PA, they usually always win by 2-4 points. He said the only way they can win is if some Democrats vote for them because the Democrats have a huge registration advantage in the state and therefore, when Republicans do win, it’s usually only by 2-4 points (or less).

I don’t think anyone actually believes the Senate race is 8 points right now, but 2-4 points seems correct.

The former governor of Alaska has zero credibility. Toomey’s camp might as well said thy were being endorsed by “no one.” “No one” would have brought more credibility.

I think Steven is on to something. Check out the Newsweek poll on Obama’s current approval rating. They have it 54% approve and 40% disapprove. But when you read the poll data it says they surveyed 350 Democrats, 305 Republicans and 299 Independents.

It’s amazing that Real Clear Politics includes polls like Newsweek and Mulhenberg in their current batches when it’s clear these polls are not correctly weighted.

Chris, John Steven-
A few things about Muhlenberg and polls in general.

1) Muhlenberg calls about 100 people per day. Then, they use the 400 people in a 4-day window to compute their poll average.
2) For Toomey to jump 3 points in a day means that there was a big gap in his favor between the latest poll and the single poll from 4 days previous that dropped out of the average.
Think of it as your grade-point-average being the result of only your most recent 4 tests. Then you get a really good grade, at the same time a bad grade from 4 tests ago drops away.

3) The percentage of Dems/Rep in the sample gets weighted to computed the final results. For example, let’s say the ratio in the survey was 50% Rep and 40% Dem (but the real distribution is 50% Dem vs 40% Rep).
Then you would take each Dem response and count it 5/4ths (ie: multiply by 1.25) instead of one. Similarly, you take each Rep response and count it 4/5ths (0.8).
This is how pollsters normalize their data to un-skew it. They can do similar manipulations to accounts for age, region of the state, etc. to make them match the known distribution of voter registration.
So, what I “think” is going on is that Muhlenberg is showing the distribution of whom they contacted, but hiding the details of the normalization. If the data is not normalized, it is worthless.

4) What is of note here is the trend. So, even though I think it’s a 2-4 point race, it appears that all the shrinking is over, and it’s starting to widen a bit. So, the poll is relevant for showing the trend, even if the numbers are off by a few points.

I think people may have been waiting for the debates. Those that like/dislike Obama heard what they needed to hear from the candidates. Joe had his best debate performance ever in the Philly debate, but the debate was viewed as a tie, not a knockout by either. Toomey’s wrong on the issues, but the majority of people expected to turn out are wrong on the issues as well and are supporting Toomey. (Don’t forget, half the population of PA doesn’t even believe in evolution, but a lot of them believe in voting.)

Muhlenberg Daily tracking:
10/19 Sestak +3
10/20 Tie
10/21 Tie
10/22 Toomey +3
10/23 Toomey +3
10/24 Toomey +5
10/25 Toomey +8

The Toomey event with Rudy was held at a non-profit, volunteer fire company…just another example of GOP using non-profits illegally to further political agenda.

muhlenberg poll is now overpolling republicans, which accounts for the changes

I have to agree that the Muhlenberg poll seems somewhat dubious. To have an 11-point swing in less than a week even with a small sample size suggests their sampling method may be flawed. David, as much as we political junkies like to attribute shifts in polling to a debate performance, I wonder how many average folks were even aware that their were debates, let alone watched them. BTW, could the quality of these debates have been any worse? They might as well have shown dueling 30-second ads as it consisted of nothing but talking points and “gotcha” implications on both sides. I just wish they could’ve found a way to weave in the phrase “Aqua Buddha” somehow.

Pardon, their = there in third sentence

Muhlenberg was wrong. Reuters just released a new poll and shows that the race is tied at 46%.

http://blogs.reuters.com/frontrow/2010/10/26/reuters-ipsos-poll-pennsylvania-senate-race-tied-with-one-week-left/

Here is a link to the detailed info from the Reuters poll:

http://www.ipsos-na.com/download/pr.aspx?id=10100

Interestingly, when asked on a scale of 1-10, how likely are you to vote, 21% of Democrats responded with 5 or less, while only 6% of Republicans responded this way. However, Independents broke +8% for Sestak, which seems to make up for some of the “enthusiasm gap”.

Umm… “Tied”-

One poll doesn’t make the race “tied”. The Reuters poll is a sample of 400 people taken over 10/22 to 10/24 and has a +/- 5% margin of error.

Muhlenberg has 448 people taken from 10/22 to 10/25 and a similar margin.

Because the two polls are roughly the same size and over the same period, you could estimate the merger of the two polls. This would indicate Toomey +4 with a +/- 3% (the margin shrinks because now it’s 848 people polled). That’s ballpark, because the polls were not conducted identically and may deserve slightly different weights. The next update to 538 should show the weight he assigns.

Rasmussen should be coming out with a poll in a few days, and we can see if they show any movement away from last week’s Toomey +4, as an indicator if the races is widening or tightening.

Here’s how a Sestak supporter analyzes a poll.

a) If the poll has Joe in the lead, tied or “closing the gap” then we’ll praise it as the most credible poll of all time. Reading a poll like this may even result in an orgasm for a Sestak supporter

b) If the poll has Joe down then we’ll reach at all leaps, bounds, and stretches of the imagination to discredit that poll even if we had previously praised it. Because how dare anybody ever say anything negative about our Emperor Sestak.

c) If all of this makes us sound totally stupid then just quote Bill Healy and say “Sestak by 8″ even though there isn’t any credible evidence to suggest that anything like this will happen. Because if Bill Healy says it then it must be true.

And that is how a feeble minded Sestak follower gets through the day.

Speaking of GOD… Sestak not only thinks he GOD, he actually stated it in his first campaign speech in the winter of 2006. It was the very first time I met him and it was a creepy experience.

As for the behavior of Sestak supporters:
a) Pretty much true. Sestak campaign sends out emails and fundraising letters quoting old polls. Bill Healy probably had a Joe-gasm all over his keyboard when Muhlenberg first posted Sestak +3.
The Toomey +8 released Tues 10/26 becomes Toomey +5 for Wed 10/27. Sestak supporters should cover their keyboards in plastic first.

b) True. Rasmussen is now their “evil” poll, but most of the year prior to the Primary, it was their go-to poll in all their letters.

c) Bill is truly the Kool-aid drenched poster-child for the Sestak lemmings rushing off the cliff.

Being a bit feeble-minded seems to be a prerequisite so you don’t realize how much he’s taking advantage and damaging the party.

Muhlenberg Daily tracking:
ending date of poll
10/19 Sestak +3
10/20 Tie
10/21 Tie
10/22 Toomey +3
10/23 Toomey +3
10/24 Toomey +5
10/25 Toomey +8
10/26 Toomey +5

Yeah, you have to figure that the real… figure… is somewhere around Toomey +3, like David said. Of course the race has tightened, Toomey’s been on-air since the primary while Sestak only just got going a few weeks ago.

David – I’m surprised you keep posting & using Mulhenberg when you posted above that it’s basically worthless since they’re not normalizing the data.

For the past seven days, Muhlenberg has surveyed as many or more Republicans than Democrats in its data. When Sestak had the lead on 10/19, they probably surveyed well more Democrats than Republicans. It’s worthless data, but not only do you keep posting it, you use it to support your position. That doesn’t make any sense, especially since you yourself said it was worthless.

Now if you average Reuters with Franklin & Marshall (which would give you the same result as you did above since one shows the race even and the other shows the race Toomey +7), then I’d agree with you since those two polling organizations are far more reputable than Muhlenberg.

Sestak by 8 Nov 2 the only poll that counts. Keep bloviating

David Diano is GOD,

You think it’s bad now, just wait until Sestak wins on Tuesday. They’re going to have a full fledged orgy.

I’m not at all surprised that both candidates approval ratings are currently more negative than positive according to Muhlenberg. With the amount of negative ads these two are launching against each other, I hate both of them. I am strongly considering writing in a third party candidate just as a protest vote. I am also considering voting against the candidate that runs the very last negative ad I hear. In other words, if I’m driving to the voting precinct on Tuesday and I hear a negative ad as I pull into a parking space (and I’m pretty sure I will), then I’m voting against that candidate that runs that ad. Sure that’s a 50/50 coin flip, but these two candidates are both horrible for Pennsylvania and their fate deserves to be based on something just as stupid like that.

Gerald-
I’m posting Muhlenberg only to show the trend (and because the Sestak campaign jumped for joy over their first Sestak +3 result). My point is that Sestak’s “momentum” from last week has stalled/evaporated and he’s as close as he’s going to get. Close enough to smell victory, but not close enough to taste it.

Nate Silver at 538 takes each pollsters history and bias, as well as sample size into account.
They gave the Muhlenberg poll a weight of 0.68 and the Ipsos poll 0.70 (which still averages out to my Toomey +4 estimate). Overall, they have Sestak with less than 20% chance of winning.

http://elections.nytimes.com/2010/forecasts/senate/pennsylvania

The F&M poll had Toomey +2 among registered voters, and +7 among likely voters (there’s that enthusiasm gap). Of course, you can count upon the Sestak campaign to quote polls for RV instead of LV.

I still think that Muhlenberg must be making some correction for the distribution of Dems/Rep. That’s basic polling 101.

Bill-
You really should heed those “Just Say No to Drugs” public service messages. In the past 8 days (since Muhlenberg +3), Sestak has not been ahead of Toomey in a single poll (including the past 7 Muhlenbergs). F&M has it at Toomey +7. Ras at +4 (with new poll due out soon). Face it, the campaign is heading South. This actually works out for Sestak, since he’ll be heading back south to his real home in Virginia.

Michael-
I don’t think there will be a win (or an orgy). I’m expecting something like those mass suicides that occur at extreme cults.
I’m not happy that we are getting stuck with Toomey. But, getting Sestak would be like getting Joe Lieberman or Ben Nelson.
The real tragedy is that the money poured into this race is going to cost us 2-4 senate seats in other states that we could have won. The party should have cut their losses and sacrificed Sestak for the good of the team.

David, on which races should they have spent the money? It looks like the closest races will likely be Colorado and Nevada, where a ton of money is already being thrown around. At what point does the diminishing returns principle come into play? After that, Pennsylvania & Illinois are the next closest races, where it would make sense to allocate that money. Manchin seems to be getting a good foothold in WV, and I don’t believe that Washington state will go to the R’s. California & Kentucky are widening races (in opposite directions), not that Fiorina needs the financial assistance anyway. I honestly don’t know if Wisconsin is in play. It’s bad news when an incumbent is hovering around 45. Well, we could parse this data all day, but as they say, the only poll that matters is the one held next Tuesday.

The reason Sestak will win is because of a turnout. It all boils down to Philadelphia. Democrats will only vote in good weather conditions. Republicans will vote no matter what. The forecast for Tuesday in Philadelphia is sunny and 62. That means everyone in Philly will come out to vote. That will be the difference and all the pollsters that have Toomey ahead will struggle to understand why they were wrong. I have made tons of money on wagering on elections just based on the weather conditions. Some clown gave me 10:1 odds on Sestak, I’m gonna clear an easy grand on Tuesday just because of the Philadephia weather. Use this to your advantage, I get an extra few bucks in my pocket every election day just knowing the trends.

Dave you have no idea,Joe Sestak by 8 pts. on Nov.2

Brent-
The diminishing returns principle comes into play when figures indicate they’ve spent more on Sestak than those other races combined.
My thinking on this expanded after the Al Franken race, when he was trying to raise a few hundred thousand dollars to fight the challenge. I realized that if the DSCC has spent a a few hundred thousand extra, the race wouldn’t have been close enough for a recount, and we wouldn’t have lost six-months of Franken’s votes.
Take the 4 million they are spending on Sestak and add a million to 4 other Senate races and you are going to get better bank for your buck in terms of total Dem seats. The DSCC screwed up by not sacrificing Sestak two months ago, and putting their resources where they could do more good. Wisconsin (and Russ Feingold) would be in GREAT shape if they got a small fraction of the money being pissed away on Sestak.

Carl-
1) Wagering on elections is probably illegal.
2) The correct odds are 5:1, Toomey favored, so you got a good deal.
3) Weather will definitely help, but it won’t be enough.
4) Sestak has refused to hand out street money, and is relying upon Brady and other uncontested Dems to spend from their coffers.
5) In at least one other county outside of SE PA, Sestak refused to help defer the costs of sample ballots for the polls, (so I wouldn’t be surprised if some of them wind up with Onorato below the torn top edge.)

Bill-
Headline for Nov 3rd:
“Hundreds Dead: Sestak Campaign Workers Commit Mass Suicide at Election Night Party by Drinking Poisoned Kool-Aid”
The medical examiner said their walnut-sized brains just couldn’t handle the reality of the loss, even though it’s been obvious for months.
One victim was found with the tag: “+8″ circled in blood. Detectives are trying to decipher the inter-workings of this strange cult and the meaning behind their strange symbols.

BTW Bill-
The only +8 I see in the polls in Toomey.

Oct. 18-24 Franklin & Marshall Toomey +8

Muhlenberg Daily tracking (ending date) of poll
10/19 Sestak +3
10/20 Tie
10/21 Tie
10/22 Toomey +3
10/23 Toomey +3
10/24 Toomey +5
10/25 Toomey +8
10/26 Toomey +5
10/27 Toomey +8

Hey David,

There you go again. Go back and re-read my post from Oct 26. Everything that I said in that one still applies today. On today’s Muhlenberg update, they surveyed 48% Republicans and 43% Democrats. Of course that’s going to swing results +3 to Toomey for today.

The only way Muhlenberg will be right is if more Republicans than Democrats turn out to the polls on Tuesday. If that’s what happens, it’ll be amazing since Pennsylvania has a big edge in registered Democrats over registered Republicans. Whether you call it 1.2 million edge or 400,000 edge, it’s still a pretty big edge — one that’s unlikely to see 5% more registered Republicans than Democrats show up at the polls on Tuesday.

Muhlenberg Daily tracking daily survey participants:

10/20: Republicans 46%, Democrats 46%
10/21: Republicans 48%, Democrats 45%
10/22: Republicans 49%, Democrats 46%
10/23: Republicans 49%, Democrats 47%
10/24: Republicans 47%, Democrats 47%
10/25: Republicans 48%, Democrats 43%
10/26: Republicans 48%, Democrats 45%
10/27: Republicans 48%, Democrats 43%

Steven-
I read your post. Did you understand my belief/assumption that they correct for the oversampling of Republicans vs Democrats, and are only showing the raw numbers they reached by phone before applying the adjustments?

Look at 10/20 and 10/24. In both cases, you are showing equal percentages Rep/Dem.
Yet 10/20 was Tie, but 10/24 was Toomey +5.

10/26 showed a 3% gap in Rep/Dem, but it was also Toomey +5

The registration edge doesn’t mean squat, because a lot of the Dems in the T and some western counties are going to vote for Toomey for guns, pro-life and anti-Obama. We had this same registration edge in 2009 and lost are statewide races.
Also, this year, the ratio of Dem/Rep new registrations is the lowest in a decade. The Republicans are already ahead on absentee ballots and have a solid 1% of the vote margin already locked in. So, even if it’s a tie among people turning out election day, Toomey still wins by tens of thousands of votes.

Steven, David…

Muhlenberg merely believes that more registered Republicans than registered Democrats will vote on Tuesday. Polls show that this is likely to be the case. More Republicans than more Democrats are enthusiastic about voting this year.

Voter turnout is one of the hardest things to predict. The best polling methods attempt to make an educated guess at voter turnout, but they’re still just guessing. If we knew the exact voter turnout ahead of time, polls would always be 100% accurate. But since exact voter turnout is impossible to predict, there will always be a magin of error for this variance. In a lopsided race, voter turnout means nothing. In a very close race, voter turnout means everything. I doubt that a better than expected Democrat voter turnout would still save Sestak, but who knows. Maybe we’ll find out on Tuesday, but I’m with Muhlenberg as I expect about 2-3% more registered Republicans than registered Democrats to show up on Tuesday.

Tom-
In raw numbers, I think more registered Dems than Republicans will show up, but….
In Nov2009
Dems: 941,625 Reps: 896,577
(despite phantom 1.2 million registration edge, Dems got crushed statewide in elections)

Lots of pro-life, pro-gun, anti-Obama Conservative Dems are going to vote for Toomey.

Forecast for Tuesday keeps flipping between Sun and Rain. If it rains a lot, add 2% to Toomey’s estimates.

Yes, Carl might want to think again about his bet. If it rains all across the state that day, then Sestak is completely doomed.

Actually David, many conservative Dems will probably vote Sestak. Sestak is a strong military man. Toomey is not. The people who like military guys will go for Sestak. For every conservative Dem he might lose to Toomey for pro gun, he might gain one in moderate Republican for the military connection. So on that issue(s), it’s probably a wash and it evens out across the board.

Don-
A conservative Dem friend of mine in Western PA told me that some of the vets in his committee are disgusted by how Sestak uses his military service as a prop. They think it’s disrespectful.
There was a poll a few months ago that actually had Toomey ahead among vets. (I doubt Toomey will wind up with more than half of the vets, but he’s going to get a big chunk of them.)
I think the pro-life voters are going to tip it for Toomey. They simply will not vote for a pro-choice candidate no matter the position on other issues. Casey took that issue off the table against Santorum.

Muhlenberg Daily tracking (ending date) of poll
10/19 Sestak +3
10/20 Tie
10/21 Tie
10/22 Toomey +3
10/23 Toomey +3
10/24 Toomey +5
10/25 Toomey +8
10/26 Toomey +5
10/27 Toomey +8
10/28 Toomey +5

All the “momentum” has stopped. The graph showing Sestak’s recent rise has leveled off. It looks like Sestak and Toomey are asymptotically approaching parallel lines about 4-points apart.

(Bill Healy’s glasses double this and view the world upside down.)

New Rasmussen reports:

10/28 Toomey +4 (50% to 46%)
Same gap as previous poll, but half as many undecideds (and no room for Sestak with Toomey at 50%)

10/21 Toomey +4 (48% to 44%)

In the words of Harry Potter’s Prof. Dumbledore: “Aresto Momentum“

David, did you see the five thirty eight did a story about how automated polls (like Rasmussen) account for a +4 difference to the Republicans. If that’s true, then the race is tied…as Reuters said. That’s why this one is still a tossup. I predict a 1 or 2 point win for either candidate. It’ll depend on turnout.

Mel-
There is some skewing of the data because more people (especially young people) are on cell phones and get missed. However, a lot of young people don’t vote.

I think 538 uses a R+2.1-point difference for Rasmussen, but they take all this into account for their weighted estimate of ALL the polls. The polls all show the race widening or stalled. Sestak’s momentum of last week has completely evaporated, just short of the goal line. Sestak is not going to be able to close the deal, especially with Onorato trailing by close to double-digits.

Muhlenberg is NOT automated and has shown significant slippage by Sestak.

From 538:
On average, the robopoll firms have a 2-point Republican-leaning house effect, whereas the live interviewer polls have a 0.7-point Democratic-leaning house effect. The difference between the two, then, is 2.7 points.

While one can argue for a single poll that Joe remains in the margin of error, that argument is much weaker when you look at the combined effect of ALL the polls. (F&M Toomey +8)
The odds of Joe being ahead but outside the margin (or having all the random poll fluctuations at odds with reality) get pretty thin as more polls come out. Also, Sestak has no time left to shift the momentum.
Neither candidate has nothing new to say.

Sestak by 8 on election night.

Bill-
I really wish it wasn’t illegal to wager on elections. I would love to take your money on that bet.