Thursday, October 05, 2006

Graf Momentum; Anticipating More Good News

Graf-ites had a good day Wednesday with the Reuters poll results suggesting something happening with the independents. Either Graf is pulling them away from Giffords, or previous polls were not very accurate. It could suggest Graf momentum or a closer race than previously suggested.

We suggest now that the campaigns are in full swing and the primary is behind us, there is in the minds of the voters a real race to consider. Graf has overcome some of the damage from the primary, shown that the Republican party is behind him and managed to reach likely Graf voters who have begun to come back at least to the point of reconsidering their hesitating for Randy. Although Graf has not picked up many of these voters yet, the number of new undecideds suggests he has made an impact and is gaining momentum, in any case.

Thursday is looking like another good day for Graf. Look for more good news for the Graf campaign.


sirocco said...


I actually don't believe there are any (or many) "new undecideds".

As all the polls have shown, the difference in support hasn't really been any change in Randy's status, it's been how many support Giffords vs. how many are Undecided.

I think the single biggest difference between all the polls has been whether or not they attempt to force a respondent who initially says they are undecided to make a choice, with a follow up question alog the lines of:

"Recognizing you are currently undecided on who you will vote for in the CD 8 House race, if you were said to be leaning toward a certain candidate, who would it be?"

... and then list the candidates along with Undecided as another choice.

The polls in which Giffords scores highest being those that push Undecideds a little harder, those that don't have her score less -- the implication being most undecideds right now are trending towards Giffords.

Until one of these polls shows Graf actually picking up a significant number of these votes Giffords is supposedly "losing", I remain unconvinced. Those "new undecideds" have been undecided all along.

Still, it remains good news for Graf, as it's something he can point to and say "Hey, this race IS competitive!" ... whether it really is or not.

Framer said...


Respectfully, I will have to disagree, of course. There is nothing in the previously released polls that show that a respondant is being "pushed" to make a decision, other than the Emily's list poll which was an actual push poll.

Either Randy has gained serious momentum, or the Kenski poll is suspect. The fact that the Kenski poll:

1. Was held for a week after it was completed.
2. At this date still refuses to publish internals, when the first poll carried them loud and proud.
3. Carried the very minummum of sampled respondants (making their crosstab breakdowns almost useless for that matter, but I will post on this later)
4. Parts of the poll information were gathered with a separate sample universe. I know some people who were called with the "are you pro-life or pro-choice?" question, but were not asked any of the other questions covered in this poll. We are not given this detail in the poll write up.

Again, I don't suspect any nefarous motive with Tucson Weekly. I really believe that this poll was rushed and done on a budget, with no room for re-dos. It happens, see 2004 exit polling for another example.

As far as the Zogby poll, we know that the randomness of the sample was checked by:

1. Selection probabilities are proportional to population size within area codes and exchanges within those area codes.

2. Up to six calls are made to reach a sampled phone number.

3. Weighting by [party, age, race, gender] is used to adjust for non-response.

If we use the other the other two recent press polls, Gabby's lead goes from 10% to 12% to 8% which looks right to me as Gabby would certainly have experienced a post election bump after her dominance in the primary, but a tightening as the rubber hit the road. It would also make her decision to hold 6 debates more understandable if she actually believed she didn't hold a 19 point lead.

Remember that most of the remaining undecideds are Bush voters and Gabby can't afford to let Graf define her to these undecideds, so she will need to remain proactive. She has shown that she is not likely to become overconfident, and will campaign effectively.

Alternatively, we could assume (which I don't) that the Minuteman ads were so devistatingly effective that they dropped Gabby by 7 points and bouyed Randy by 3 in just a few days. That would be far worse news for Gabby.