Note: This blog's templates will be updated this afternoon to a responsive design bringing it in line with

At that time, we will also change to the Facebook commenting system. You will need to log in with a Facebook account in order to comment.

« Nearly 400,000 Floridians have already voted. And counting. | Main | Doh! FL union uses "vote early, vote often" fraud mantra as election slogan »

Mack's consultant, Arthur Finkelstein, makes case for why every pollster has it wrong

We're really not too sure if this is news but Republican Connie Mack's political consultant Arthur Finkelstein thinks he can win.

Nontheless, his analysis is telling and the fact they he's posted on the candidates's web site for the world to hold him accountable for it come Nov. 7 is worth noting.

Here it is:

In this year’s election most polls are surveyed off of 2008 turnout models.  This is standard even though the turnout for the Democrats – specifically Obama – was exceptionally high and turnout for the Republicans was lower than average.  The combination of high Democratic turnout and low Republican turnout resulted in Obama and the Democrats winning in states they would normally lose– like Florida.

It is clear that few –if any– McCain voters are angry about not voting for Obama, and we all know many people who voted for Obama who are now not sure that that was the right thing to do.  Democratic turnout surely will be significantly less than it was in 2008.  And some of the Democratic vote that went for Obama will this time vote for Romney and the Republicans.  In short, it means that the sampling should more reflect 2000 and 2004 turnout models than 2008.

In Florida in 2008 the Democratic turnout was +4 percent more than Republican.  Our own previous poll had Democratic turnout at +5.  When the survey is indexed so that it more reflects 2000 and 2004 turnout models, the race becomes 1 point with Nelson still leading.  But this was a week ago.  This week, in the last few days, we have received a PPP poll, a Mason Dixon poll and a Rasmussen poll.  Let me take them one at a time.

The PPP poll had it 45 Nelson, 37 Mack but its poll was +4 Democrat, a stunningly wrong +12 female, and 18% undecided.  Since the undecided have been breaking in every poll 60- 65% for Connie, Connie can expect 10-12% of the 18% undecided to vote for him, which would bring his number to a more realistic 47 to 49.  And that is without any tweaking of the +4 Democrat or the +12 women.  PPP’s poll should be seen as a probable +1 Connie Mack.

Mason Dixon – the most respected statewide polling firm in Florida—had Connie behind 47 – 42.  They too used the 2008 model and therefore had Democrats +4.  Once again, if you index for a proper turnout model, about 3 of the 5 points margin disappears and if Connie were to receive his 60% of the undecided, once again, he would be about one point in the lead.

Finally, the most recent poll was Rasmussen, which used a reasonable turnout model where Republicans were +1 over Democrats.  In that poll, Connie was losing by one point – 46-45—with 4 points going to third parties and 5 points undecided.  Again, if Connie gets even the low end 60% of the undecided, this race would be exactly tied 50-50

So, there are four polls this week. We are anywhere from  -1 to +2 and clearly moving along with Romney.  Now is the time for everyone to step up and win this seat.  Nelson has never moved from the mid-forties. He started there; he’s still there.  There is a clear path to victory.  Now is the time to redouble our efforts.