Posted by 
By now, you likely have seen reports about the recent Anzalone-Liszt poll that shows some strong numbers for Congressman Artur Davis in the 2010 Governor’s race. The polling summary prepared by Anzalone-Liszt is here (.pdf).
In case you managed to miss the release as it has been posted elsewhere (e.g., Left in Alabama), I’ll post it below.
Some notes and observations, most of which came from John Anzalone on a conference call:
- The sample of African-American voters in the polling for the Democratic primary was a conservative 46%, in an effort to ensure that the numbers are credible and realistic.
- Davis polled 5 points ahead of Bradley Byrne, “widely seen as the GOP frontrunner.” There were no numbers for other GOP candidates.
- Byrne has not announced, but these #’s are a starting point that won’t change for a candidate until the candidate starts to use paid communications, according to Anzalone. “Until you do tv, mail, radio, it’s not going to move the #’s. Higher name ID now gives you an opportunity to solidify your advantage.”
- The numbers below are from head to head matchups, but when asked about Davis, Cobb, and Sparks in a three-way race, Anzalone replied that the poll #’s show Davis with “a plurality and with three times the vote of either Cobb or Sparks.”
- Anzalone added that Democratic candidates have to hold the “bleed” on “self ID Democrats” to below 20% to win. He explained that you get these “cultural Democrats” who identify themselves as Democrats because they and their family historically have been Democrats, they vote Democratic for the local sheriff, but who go to the voting booth and vote Republican on these big races. Democratic candidates need to keep that bleed under 20%, and right now Davis is holding the bleed to 5%.
- Any Democratic candidate in Alabama needs to get about 36 to 38% of the white vote to win.
- In the poll, Artur Davis outperforms a generic Democrat by 7 points. Byrne underperforms a generic Republican by 3 points.
While the numbers are in keeping with other polls that have been in circulation, this one has gotten attention. Stories have appeared at MSNBC’s First Read, The Economist, and Washington Post’s The Fix.
Here is the release from the campaign:
DAVIS HOLDS CLEAR LEAD ON DEMOCRATIC RIVALS; STRONG SUPPORT ACROSS RACIAL LINES GIVE DAVIS 2-1 LEADS OVER DEMOCRATS AND 5 POINT LEAD OVER LEADING REPUBLICAN BRADLEY BYRNE
In a survey of 600 Democratic primary voters conducted from May 5 to 9, 2009, Congressman Artur Davis (D-Bham) holds a clear edge over all of his potential rivals in the 2010 gubernatorial primary. In prospective head-to-head matchups against each of his potential Democratic rivals, Davis holds decisive leads: against Agriculture Commissioner Ron Sparks, Davis leads 56 to 26 percent; against Supreme Court Chief Justice Sue Bell Cobb, Davis’ edge was 54 to 25 percent.
In each prospective matchup, Davis shows overwhelming strength among black voters (who were 46 percent of the sample) and strong support among white Democratic voters, as well. Davis split the white vote with Cobb, 37 to 38 percent, while he wins the black vote 74 to 11 percent. In a match with Sparks, Davis led 77 to 9 percent among blacks, while narrowly trailing with whites 37 to 41 percent.
Davis’ across the board popularity yields him a total favorable to unfavorable ratio of 59 to 6 percent among Democratic primary voters, the strongest popularity ratio of any potential Democratic candidate. In the 22 county Birmingham media market, the largest voter share in the primary, 72 percent of Democratic voters hold a favorable opinion of Davis while 83 percent of the Democratic voters in that market approve of the job he is doing in Congress.
In a separate sample of 600 general election voters conducted during the same time frame, Davis has opened up a five point lead over the likely Republican front-runner, Two-year College Chancellor Bradley Byrne. Byrne, who is expected to announce his candidacy for governor next week, trails Davis 38 to 43 percent. Davis leads Byrne by double digits in the Birmingham and Montgomery media markets, and holds a narrow 5 point edge in the Huntsville media market. White Democrats support Davis over Byrne by an 81 to 5 percent margin, while independents break for Davis by three points.
In separate findings in the survey, President Obama’s popularity statewide has risen to far higher levels than his showing in the 2008 election in Alabama. Fifty-eight percent of voters approve of his performance, including 46 percent of white voters. In addition, 87 percent of white Democrats and 57 percent of independents approve of Obama’s performance.
Davis spokesperson Anna Ruth Williams stated, “While Artur Davis may not be the preferred candidate with political insiders at the state capitol, it is clear that his message of strengthening our economy and cleaning up state government is catching on with voters around the state. At this stage, Congressman Davis is in a very strong position in the primary and general elections against his leading Democratic and Republican rivals.”
###
Legislative Dispatch
Purple Dot Connection
2010 Big List
2010 Senate Elections
2010 House Elections
Press Releases
That black percentage might still be high. A state election primary is not going to result in turnout anywhere close to the 2008 Presidential primary. 46% is on the high end of realistic for that.
However, these polls are actually more reflective of runoff contests (if they were tests of the initial primary then all three would be in the same heat). The runoff is going to be much lower turnout and much whiter than the primary.
What does this mean? Not much. Davis is still the frontrunner, and a lot would have to change to knock him off that perch.
I respectfully think you are wrong on the Black turnout. I think it would be very high due to the wave of black patriotism set by Obama. The problem in the past has always been getting the younger generations to turn out, and I think an Artur Davis type energizes that gruoup. Just a thought.
So Byrne is the GOP front runner and is underperforming a generic Republican by 3 points. Ouch. Maybe I’ll change my name to “Generic Republican” and run.
46% Black turnout is very close to the 45% turnout in the 2008 Primary.
Assume a minimum 3% Black Republican vote leaves 42%
That 42% still would be enough for a majority of the votes in the Democrat Primary if the rural white Democrats flee the Primary as they did in th Presidential Primary
interesting that they poll against byrne but not james…
Sparks needs to run for state treasurer. He will get smoked by Artur.
The percentage is too high. There will be a lot more white people voting in the Democratic primary because of local elections. That will greatly reduce the AA vote percentage in comparison with the 2008 Presidential Primary. Even if Artur gets the same number of AA voters to the poll the percentage won’t be higher than 40% because of the huge influx of white voters to the Democratic primary that Sheriff races bring in.
If they redid their polling to show a more realistic primary turnout his margin would greatly decrease. Now, is that enough to keep him from winning is the question.
It is interesting that Cobb got more of the AA vote than Sparks. That to me is the interesting number. Everyone assumes that all AA’s will vote for Artur and that is not the case. So, if 40% of the vote is AA and Cobb can get 5% of that then Artur has to get 15% of white vote. Can he do that? Probably in a primary if it is only those 2 going head to head.
However, if it is a runoff then there is an even bigger issue for Artur cause his voters have not been consistently voting in the past. There is no way to project the turnout in a special election of people who at that point will have probably voted twice in their lives. If you reduce his AA vote to 30% in a runoff he looks extremely beatable.
Right now I would say Artur’s best bet is to get the party to make it a head to head battle between 2 candidates in the primary. He will win that.
Its premature to call Byrne the frontrunner and due to his low id he might not be at the top of any Republican primary poll right now.
The expectation of his strength is that he was raise huge amounts and buy high ID and that the media will give him great coverage.
But how well the Republican base will respond to him in very much in doubt in my mind.
And the new Davis is in line with other polls?
And the link was to the earlier poll done by the same company?!
Another thing, if black primary turnout was 45% with Obama in the primary organizing like crazy and spending big bucks etc…
And with Hillary writing off the state.
That means 46% is rather high. But it also may be at this point the turnout model for the Democrats may be pretty settled.
Wasn’t the turnout something like that in 2006?
The 45% Black was the Primary. Obama ran in the Presidential Primary, different election with different percentages.
The question is does having a Black Gov candidate rasie the black Democrat percent above 50 instead of the normal 42%
I mentioned this on Brian’s blog the other day. I think a conservative black could win the govs race. But it would take a pretty special circumstance(Roy Moore winning GOP nod) for Artur to win if you ask me. The voting record is just too liberal. A conservative black could go in to these rural white areas like Blount,Lawrence and Colbert Counties and calm the nerves of a lot of the rural whites who are still a little apprehensive to vote for blacks out of old bad habits. By pointing out that they agree on a lot of the problems facing the state today. Artur can’t do that. Sure he’ll try to blur his record but it is what it is at the end of the day.
Some thoughts on this poll…I’m lost on #7. This poll shows Davis running almost even with the white vote against either Sparks or Cobb. If he even held the roughly 40% he has now, and the rest all went to a white opponent in a two way, he could lose almost 30% of the black vote and still win.
As for the turnout number, let’s not forget that there will be Republican primary races and the white turnout has been declining in Democratic primaries for two decades. Also, there are a lot more contested black primaries in legislative races than in white districts, where the general is the ballgame. I think 45 is probably the low.
I buy that Davis has a lead on Byrne, or likely any Republican, right now. If Obama’s numbers are up to 50% or plus here, and every published poll shows they are, wouldn’t almost anyone who approves of Obama (stimulus and all) likely say they will vote for Davis or any other Democrat over a Republican. Of course, i would expect that Cobb or Sparks would probably be up too (no surprise that Davis did not share those numbers).
To Prince Liberty, John Anzalone is not everybody’s cup of tea, but his polls were dead on in the congressional races in CD2 and CD3. He is hardly dependent on Davis for money or clients and its kind of crazy to suggest he cooks polls for Davis and nobody else.
One moderate Democrat’s conclusion: the Republicans have a bad field and if the Democratic money machine figures that out and dumps money in here next year…
The 57% favorable for Obama, does draw this poll into question. Since most of the white people you ask about Obama, spit before mentioning his name, it seems a bit high.
Davis definitely wins the early name recognition race. But, I think that after Byrne formally announces and starts touting what he’s done in the two-year college system, these numbers will change.
If blacks poll at 35% of the General Election, that is 35% of 15%. If blacks poll at 63%, that is 48% of the Democrat Primary. Disproportionate impact in a small meaningless sample cannot be statistically corelated with any meaningful conclusions. Translation? Artur Davis takes the Democrat Primary with 65% of the vote. That of course has no relation to the mass of Alabama citizens. If Roy Moore runs and is the nominee, or a Tim James type fringe fruit loop type candidate, Artur will win. Otherwise, it aint gonna happen.
underestimate or dismiss Anzalone’s numbers at your own peril.
Easily the best pollster and political mind in the entire state.
I’d take one of him over 100 Swatekss, Chris Browns and Scott Beasons.
Agreed politiki, I just find the numbers for Davis and Obama among white voters a little high at this point. And I think the sample may not take into account the fact that primary turnout in rural counties will be driven by local races not statewide ones. If I see another poll showing similar numbers for Davis and Obama, I can buy it, but right now I think this might be an outlier.
One needs to consider that Artur Davis won election to Congress by garnering a strong white vote in the Birmingham metro area and in the Black belt. Davis is a moderate and his opposition will have a hard time painting him as a radical.
Davis will cut into the blue blood vote in areas like Mountain Brook. Frankly, this man could be your next Governor. This is not promotion, since I do not plan to support Davis, but selling him short would be a terrible mistake.
John, true a white dem would pick Artur over Earl Hilliard, not sure the same would be true against Sparks or Cobb. That being said, Artur doesn’t need blue blood votes; he can win with blacks and the pinhead liberals in the suburbs who think they are being progressive by voting for the black guy. He is unbeatable as of now in a dem primary.
for all those that think Byrne will have a problem for being a “former democrat”….here is a good example of why it will not be a problem.
http://www.youtube.com/watch?v=RRUbwnkEPqc
Politiki – just because John Anzalone is the best politico in the state doesn’t mean we should accept his internal as gospel. He made this for public consumption. What’s he going to do? Tell Artur “hey you can’t win so don’t pay me 26K a month”? Hell no. If he made polls showing Artur losing he’d lose a lot of money when Artur fired him. Underestimate him at your peril? Yes. Underestimate his numbers for Davis at your peril? No.
Mr Larron – you’ve obviously never met Anzalone.
I find a comment like #21 really odd. Public figures hire reputable pollsters because they want to get at the truth. True, when it benefits the clients, the #’s may get released.
A pollster aiming to keep a customer by telling the client something the client wants to hear instead of the truth would be akin to a mechanic telling a car owner that all he needs is a simple repair when he really needs something much more costly. It’s good news that the car owner wants to hear but in the long run, the mechanic will lose the business because he’s not trustworthy.
People hire good pollsters because they want the truth, and good pollsters don’t make their reputations by telling clients only what they want to hear.