Sam Spade's Final (see p.5) Gubernatorial prediction thread...

[Sam Spade]

I have updated for the week of September 17.

My changes for this week are to move Oklahoma into Safe D, Arkansas and Massachusetts into Likely D.  I see no other moves as being necessary and I am not changing my prediction, yet.

Remember, for those of you who might bring X poll here or Y poll here for contention, I typically only move when two polls say something similar.  The only place where this is not the case is in the House, but different rules apply there.

[Moooooo]

Sam, a new Ras poll set to be released tomorrow will show O'Malley up 7.  I believe the last 3-4 polls have shown him with a similar lead and hovering right around 50%.  Do you believe this is his core base?  Im trying to put my figure on a base number for both Ehrlich and O'Malley, but given Ehrlich is a unique candidate and generally does very well with moderate/consertaive 'Reagan Democrats' its hard to judge what his core support # his.  Do you have any ideas?

Also, do you think it is a positive thing for O'Malley that Ehrlich has been running negative attack ads for 2 months straight now and he has yet to make a dent in O'Malley's lead according to Ras' numbers?

Give me your thoughts on this race you ex-Marylader, you....

[Sam Spade]

Sam, a new Ras poll set to be released tomorrow will show O'Malley up 7.  I believe the last 3-4 polls have shown him with a similar lead and hovering right around 50%.  Do you believe this is his core base?  Im trying to put my figure on a base number for both Ehrlich and O'Malley, but given Ehrlich is a unique candidate and generally does very well with moderate/consertaive 'Reagan Democrats' its hard to judge what his core support # his.  Do you have any ideas?

Also, do you think it is a positive thing for O'Malley that Ehrlich has been running negative attack ads for 2 months straight now and he has yet to make a dent in O'Malley's lead according to Ras' numbers?

Give me your thoughts on this race you ex-Marylader, you....

Well, first, no poll has showed O'Malley above 47% since early, early in the year, so if he's up towards 48% or 49%, that would be better for him.  Of course, no poll has shown Ehrlich above 42% other than the crappy AP poll.  And it is only one poll, a point which I can never stress enough.  Remember RI. 

If you want real honesty, I would say that O'Malley's base is probably in the 45% figure and Ehrlich's is probably in the 40% figure (Rasmussen pushes leaners too much for me).  Add in leaners, etc., that probably brings the figure up 2%-4% for both candidates, so you can do some math and figure that one out.

The real question is, who are the undecideds right now.  That would help determine how hard I think O'Malley's base is and how well he should do among the remaining voters.

Of course, I agree with you that ads lose their effectiveness after a while, unless something "new" comes out.  As I pointed out with Pederson, if you spend $5 million bashing someone and it doesn't change numbers, spending $5 million won't do anything, unless there is something really "new" and fresh in voters' minds.

[Moooooo]

Well, first, no poll has showed O'Malley above 47% since early, early in the year, so if he's up towards 48% or 49%, that would be better for him.

http://www.rasmussenreports.com/2006/State%20Polls/August%202006/marylandGovernor.htm

[Moooooo]

Fair analysis.  I actually had very similar numbers in mind.  Now the hard part - figuring out who the undecideds are, why they're undecided, and which way they will break.

O'Malley's strategy lately has been to tie Ehrlich to Bush and Corporate interests.    Given Bush's poor numbers in MD do you think this strategy will work or will it backfire because Bush has nothing to do with this election?

[Sam Spade]

Well, first, no poll has showed O'Malley above 47% since early, early in the year, so if he's up towards 48% or 49%, that would be better for him.

http://www.rasmussenreports.com/2006/State%20Polls/August%202006/marylandGovernor.htm

I must be getting lazy and not reading correctly.  Sorry.  Please ignore the previous post.

The base is probably then more like 41%-42% for Ehrlich and 47%-48% for O'Malley, but with leaners pushed, as Rasmussen does, you can expect that to be up 2% for both candidates.  I'd say that O'Malley's base is probably enough to be on the cusp of probably providing a winning margin, barring fundamental changes in the campaign, but I would have held that anyway.  The thing is that a change of a couple of percent here or there could have a large effect, hence my call on the race.

[Sam Spade]

Fair analysis.  I actually had very similar numbers in mind.  Now the hard part - figuring out who the undecideds are, why they're undecided, and which way they will break.

O'Malley's strategy lately has been to tie Ehrlich to Bush and Corporate interests.    Given Bush's poor numbers in MD do you think this strategy will work or will it backfire because Bush has nothing to do with this election?

It's good base solidification strategy, which should be what O'Malley strives to do from now on, IMO.  That and making sure blacks turn out.  Will it have any impact on others (undecideds)?  I doubt it.

[Sam Spade]

Because it's now the six-week mark before the November 7 elections, I am going to review all the races, take a few off the table and give detailed commentary where I think things are right now:

Governor

Safe D (nothing really much to say about these, folks)
Arizona
New Hampshire
New Mexico
New York (R)*
Ohio (R)* (moved from Likely D to Safe D) - The polling is not quite there yet, I just think it's going to be a very bad year for Republicans in Ohio overall.
Oklahoma
Pennsylvania (moved from Likely D to Safe D)
Tennessee
Wyoming

Likely D
Arkansas (R)*- Haven't seen a poll out of here in a while.
Colorado (R)*(from Lean D to Likely D) - I have my corrboration.
Kansas - Another poll like SUSA and I'll move, another poll like Rasmussen and I won't.
Massachusetts (R)* - I don't move based on one poll.

Lean D
Illinois - Still on the cusp of Likely.  I prefer to go with the more competitive ranking.  Besides, Blago is still polling under 50%.
Maryland (R) - Steady as she goes.
Oregon - It's just where I feel the race is right now.
Wisconsin - Same here.

Toss-up
Iowa* (D) - The generic wisdom is that when polls have shown either one candidate leading or at best tied, the candidate who is leading has a lead somewhere just right at the MOE.  I believe this is probably the case here, but the fact there appear to still be a lot of undecideds is what is making me opt for the more competitive designation.  When I post the Senate thread, a couple of races I have determined to be Lean because no poll has shown the opponent ahead and therefore I determine that the lead is somewhere just outside MOE.  So for me, possibly inside/outside MOE = Toss-up, definitely outside MOE = Lean.
Maine (D) - Baldacci is leading, but all of his leads have been inside MOE (in polls that aren't just utter crap), and his numbers are too low for an incumbent.  I would be close, as with Iowa, to putting this in Lean D, but I have those reservations.
Michigan (D) - Total toss-up.
Minnesota (R) (from Lean R to Toss-up) - Mere preservation right now.  The HHH poll and Star-Tribune are noted for being totally crappy, but the Mason-Dixon made me think I'd better play conservative on this one.  I suspect that Pawlenty has a lead slightly inside or outside MOE, but until I'm sure, I'll play it safe.
Rhode Island (R) - Same here.  Unless someone corroborates the uni poll.  But I don't see that happening.  There aren't any good polls out of RI anyway, so we're probably just guessing.

Lean R
California - On the cusp of Likely, but I opt for the more competitive designation.
Florida* - When Mason-Dixon polls here, I'll make some changes, probably.
Nevada* - Also Lean on the cusp of Likely.

Likely R
Alaska - Steady as she goes.
Alabama - If I see another poll consistent with SUSA's earlier findings, I'm putting this one in Safe R.
Georgia - Taylor is out of money.  I see this one as being close to over with, but I am waiting for a SV poll (they are an excellent polling firm for GA)
South Carolina (from Lean R to Likely R) - SUSA corroboration made this move.
Texas - I see no reason to make a change.
Vermont - Once again, another poll corroboration and I'll make the move.

Safe R
Connecticut
Hawaii (from Likely R to Safe R) - No polls here, but this seems to be the general consensus in the state seems to be such from what I've read.
Idaho*
Nebraska
South Dakota

Present prediction (As of September 24, 2006)

Democratic gains
Arkansas
Colorado
Maryland
Massachusetts
New York
Ohio

Republican gains
None

Total changes:
+6 Dems, 28D, 22R total

[Sam Spade]

October 1 update.

My general comment would be that with the way things are going, the probable best Republican result is going to be Dem +3 and the best Democratic result will be Dem +9.  Dem +4 to +8 is the best call, all told.

Governor

Safe D
Arizona
Massachusetts (R)* (from Likely D to Safe D) - Walter will probably not like me. 
New Hampshire
New Mexico
New York (R)*
Ohio (R)*
Oklahoma
Pennsylvania
Tennessee
Wyoming

Likely D
Arkansas (R)*
Colorado (R)*
Kansas - Unless Rasmussen's poll is validated by someone else, no change.

Lean D
Illinois - On the cusp of Likely D.  I'm a little more leery on this one since Blago is under 50%.
Maine - (from Toss-Up to Lean D) - With two polls showing Baldacci up 5, I'm making the move.  However, if a poll comes around showing Woodcock leading, I'm moving back to toss-up.
Maryland (R)
Oregon
Wisconsin

Toss-up
Iowa (D)*
Michigan (D) - EPIC/MRI says one thing, everyone else says something different.  When this happens, I typically go with everyone else.
Minnesota (R)
Rhode Island (R)

Lean R
Nevada* - On the cusp of Likely R.
South Carolina (from Likely R to Lean R) - I suspect I will be moving this one back soon, but I'd like to see another poll.  Really on the cusp, all told.

Likely R
Alaska
Alabama - Still on the cusp of Safe R
California (from Lean R to Likely R) - Making the move.
Florida* (from Lean R to Likely R) - Same here.  M-D solidifies.
Georgia
Texas
Vermont

Safe R
Connecticut
Hawaii
Idaho*
Nebraska
South Dakota

Present prediction (As of October 1, 2006)

Democratic gains
Arkansas
Colorado
Maryland
Massachusetts
New York
Ohio

Republican gains
None

Total changes:
+6 Dems, 28D, 22R total

[Moooooo]

When you make your predictions do you go strictly by polling or do you take into account political atmosphere in each state, party id advantage, GOTV effort, organization, etc...?

For example, in 2005 some of us were saying Kaine needed to be up in the polls 3, 4, even 5 points to win on election day.  Turned out he was up that much going into the eve of the election and ended up winning by a similar margin.

Im asking becuase Im trying to gauge for sure how the race in MD stands.  The last 3-4 polls show O'Malley +6-8 and right around 50 and Ehrlich in the mid 40's.

Ive also been looking at past polling data lfrom 2004 and 2002.  Almost every poll Ive looked into has underestimated the Democrats final result in MD.  I wonder if this is due to undecideds breaking towards the Democrat or the GOTV operation in MD or possibly even the increase in voter registration fo the dem's...?

[Moooooo]

One more thing, I know Garin Hart Yang is a partisan polling firm, but do you have any idea how strong their record is?  Do they tend to show overwhelming Democratic support or are they one of the more accurate partisan pollsters?

[Sam Spade]

When you make your predictions do you go strictly by polling or do you take into account political atmosphere in each state, party id advantage, GOTV effort, organization, etc...?

In the Governor's races, it's mainly polls combined with fundraising and my own personal opinion about the quality of the candidate.  The political atmosphere of the state does come into play, much more so in governor's races, and much more than the national atmosphere.  Party ID advantage will occasionally come into play, but not as much so because state races tend to be less partisan than national ones.  GOTV effort and organization are really tough to analyze properly and so I tend to ignore them.  Besides, I have gotten things wrong by even relying a little on those factors before.

You must be logged in to read this quote.

One's assumptions about a state are not always accurate.  I would say that typically polls tend to underestimate Republicans in VA slightly.  Notice I was wrong last year.  However, remember 1993 in New Jersey, a state which we always assume overestimates Republicans.  Most polls had Florio with a 5%-8% lead.  Whitman ended up winning by 1%.  You can't let yourself fall into laziness because you presume something that might not happen in this particular situation.

You must be logged in to read this quote.

That's where it probably is right now

You must be logged in to read this quote.

That could be due to any, all or none of those factors.  And although history says something, you can't let yourself be fooled by those things.  This year may be totally different.  You can make educated guesses based on that info, though. 

[Sam Spade]

One more thing, I know Garin Hart Yang is a partisan polling firm, but do you have any idea how strong their record is?  Do they tend to show overwhelming Democratic support or are they one of the more accurate partisan pollsters?

Trying to guess the accuracy of a partisan internal pollster is like trying to figure out the attractiveness of the girl you're hitting on when you're totally plastered.  Good luck, man.

[I spent the winter writing songs about getting better]

I'd make Iowa Lean D, otherwise I agree.

[Sam Spade]

I'd make Iowa Lean D, otherwise I agree.

Not when there are so many undecideds and the recent polls have it within the MOE.

[Amenhotep Bakari-Sellers]

RI goes to leans R, mason dixon solidifies the lead.

[Sam Spade]

With the rash of Mason-Dixon polls out today, the only major change I would make is to push Rhode Island to Lean R.

A lot of these governor's races are moving closer and closer to the Safe side of the equation rather than otherwise, in my opinion.  I think we will have few highly contested Governor's races on the whole.  Contrast that to the Senate, where the opposite seems to be happening.

[TheresNoMoney]

Sam, I think you did a great job with your predictions, and I have a hard time disagreeing with anything you said.

I would love to see Dina Titus make a comeback in Nevada, but I agree that it's definitely a Republican-favored seat.

[Moooooo]

With the rash of Mason-Dixon polls out today, the only major change I would make is to push Rhode Island to Lean R.

A lot of these governor's races are moving closer and closer to the Safe side of the equation rather than otherwise, in my opinion.  I think we will have few highly contested Governor's races on the whole.  Contrast that to the Senate, where the opposite seems to be happening.

You knew it was comming...  What about the M-D Maryland poll?  Change your mind about the race at all?

[Sam Spade]

With the rash of Mason-Dixon polls out today, the only major change I would make is to push Rhode Island to Lean R.

A lot of these governor's races are moving closer and closer to the Safe side of the equation rather than otherwise, in my opinion.  I think we will have few highly contested Governor's races on the whole.  Contrast that to the Senate, where the opposite seems to be happening.

You knew it was comming...  What about the M-D Maryland poll?  Change your mind about the race at all?

Not really.  Maryland's going to be one of the few good Gov. races out there.

[Amenhotep Bakari-Sellers]

When you have 50% approvals I don't think you can say you will be defeated that easily. He is at mediocre approvals  and he can still win.

[Sam Spade]

The most strongly noticed part of this update will be me putting a number of few of the Safe D races in Likely D.  Based on some polls this week, my choice would either be to take a couple of Likely R's off the table (Alabama, Georgia, Vermont) that look the same as a couple of races that I've had in Safe D.

Rest assured, none of these races will likely be competitive, it is merely a small note of change.

Governor

Safe D
Arizona
New Hampshire
New Mexico
New York (R)*
Oklahoma
Tennessee
Wyoming

Likely D
Arkansas (R)*
Colorado (R)*
Kansas - Unless Rasmussen's poll is validated by someone else, no change.
Massachusetts (R)* (from Safe D to Likely D) - Just a gut feeling.
Ohio (R)* (from Safe D to Likely D) - The polling says this, even though I think the race is over.
Pennsylvania (from Safe D to Likely D) - See above.

Lean D
Illinois
Maine
Maryland (R)
Michigan (Toss-up to Lean D) - Rasmussen validating EPIC/MRA made me make the move.  However, a move back could be made with different polling data.
Oregon
Wisconsin

Toss-up
Iowa (D)*
Minnesota (R)

Lean R
Nevada*
Rhode Island (Toss-up to Lean R) - Rhode Island polling is crap, but we haven't seen Fogarty lead in a while.  I don't know whether I believe M-D or Gallup, but my gut says M-D is more right than Gallup.
South Carolina

Likely R
Alaska
Alabama - Still on the cusp of Safe R
California (from Lean R to Likely R) - Making the move.
Florida* (from Lean R to Likely R) - Same here.  M-D solidifies.
Georgia
Texas
Vermont

Safe R
Connecticut
Hawaii
Idaho*
Nebraska
South Dakota

Present prediction (As of October 8, 2006)

Democratic gains
Arkansas
Colorado
Maryland
Massachusetts
New York
Ohio

Republican gains
None

Total changes:
+6 Dems, 28D, 22R total

[MAS117]

Has there been any polling the Alaska race?

[© tweed]

Has there been any polling the Alaska race?

Yeah, Palin's going to win.

[Sam Spade]

This week you will start to notice me starting to take races off the table.  The first ones will be those where the incumbent or challenger is consistently over 50% and has a 10+% lead, unless I see some type of exception which warrants me keeping it otherwise.

Governor

Safe D
Arizona
Colorado (R)* (from Likely D to Safe D)
New Hampshire
New Mexico
New York (R)*
Ohio (R)* (from Likely D to Safe D)
Oklahoma
Pennsylvania (from Likely D to Safe D)
Tennessee
Wyoming

Likely D
Arkansas (R)*
Kansas
Massachusetts (R)*

Lean D
Illinois - The scandal problems for Blago are keeping him here.
Maine - No polls from here for a while.
Maryland (R)
Michigan - Slowly slipping away from Republicans.
Oregon
Wisconsin

Toss-up
Iowa (D)*
Minnesota (R)

Lean R
Nevada*
Rhode Island - If a poll shows Fogarty up, I'll move.
South Carolina

Likely R
Alaska
California
Florida*
Texas

Safe R
Alabama (from Likely R to Safe R)
Connecticut
Georgia (from Likely R to Safe R)
Hawaii
Idaho*
Nebraska
South Dakota
Vermont (from Likely R to Safe R)

Present prediction (As of October 15, 2006)

Democratic gains
Arkansas
Colorado
Maryland
Massachusetts
New York
Ohio

Republican gains
None

Total changes:
+6 Dems, 28D, 22R total