Wednesday, November 08, 2006

Weak Turnout Numbers and Pool Update

I had two predictions in my Guess the Turnout office pool. The first was that turnout would set a new record for a Gubernatorial election -- over 61%. That prediction was flat out, dead busted, butt-lickin' wrong.

The second prediction was that if turnout was in the mid-50's Dems would win big. Well, turnout looks to be in the mid-50's and Dems nearly ran the table in the statewides. Of the six Dems in partisan races, five won. Plus eight or so state house pickups and nearly perfect defense of all seats held. My guess is that part of the Dem wave was unmotivated Republicans staying home. We'll have to wait for some more detailed results and exit poll cross-tabs to be sure.

In the meantime, it looks unlikely that the vote count will grow enough to pass Lori who had the lowest guess. As of right now the count stands at 3,868,486, with three counties yet to report. Stark is the biggest of those by far with 270-odd thousand voters. I expect the count to grow by another hundred fifty thousand or so when those counties report, plus another point or two when the provisionals are in.

Meanwhile consider these sad turnout numbers:

Cuyahoga County: 39.44%
Franklin County: 44.74%

All the other major urban counties posted respectable numbers, mostly in the mid-50's with Hamilton County just shy of 50%. Assuming that the sad numbers in Franklin (and Hamilton for that matter) reflect center city residents not voting, the Dems have some serious work to do here. If those voters had turned out in numbers comparable to say Dayton or Toledo, Barbara Sykes would probably be our new Auditor.

Finally, it can't be a coincidence that the two counties with the worst reported voting glitches also had the worst turnout. Jennifer Brunner says that she will administer elections with an eye toward fairness to all and advantage to none, and I believe her. I also believe that if we have that in Cuyahoga and Franklin, it will be a boon to Dems.


Anonymous said...


Judging turnout performance on percent-of-registered in an urban county (at least Cuyahoga)is not terribly helpful... there are still way too many nonexistent registered voters. Personally I think it's more useful to compare this year's turnout to voters who we know exist, i.e. those who actually voted in 2004. By this standard, and assuming the SOS numbers are valid (which is questionable IMHO, see why below) Cuyahoga's turnout was a lot better than average. The statewide ratio of 2006 votes to 2004 votes was about 73%. If Cuyahoga really turned out 562,000 voters, as the BOE and SOS charts claim, then its 2006-to-2004 ratio was more than 80%.

The problem with this is that less than 420,000 of those 562,000 Cuyahoga voters actually seem to have voted for anybody or anything. If you assume, therefore, that the real Cuyahoga turnout was around 430,000 (with overseas and provisionals added in) the county's 2006-to-2004 ratio falls to 62%, which is comparable to Franklin and Hamilton... and your basic point stands.

Why were the urban turnouts so low? I worked GOTV in several West Side Cleveland precincts. Here's my conclusion: a) The coordinated campaign operation for contacting and motivating 2004-only voters, whether by canvass or phone, was just ineffectual -- too little, too late, too incoherent; and b) the failure to integrate Issue 2 in the main campaign message and put it on TV minimized its impact on turnout.