Kerry 291 - Bush 222 - ?? 25 (Ohio and West Virginia)

If I project the two toss-up states on the basis of the latest polls there, I get:

Kerry 311 - Bush 227

(You'll find notes as to my thinking in these assignments in the "comments" to the various versions stored here.)

As for other sites:

Chris Bowers (7/19):
Kerry 322 - Bush 216

Chuck Buckley (7/19):
Kerry 311 - Bush 227

Charlie Cook (7/6 - not updated since last survey):
Kerry 207 - Bush 211 - ?? 120

Dales' EC Breakdown (7/20):
Kerry 282 - Bush 242 - ?? 14

DC Political Report (7/17):
Kerry 251 - Bush 183 - ?? 104

Election Projection (7/19):
Kerry 327 - Bush 211

Electoral-vote.com (7/20):
Kerry 322 - Bush 205 - ?? 11

Kevin's Marginalia (7/15):
Kerry 291 - Bush 247

LA Times (7/20):
Kerry 169 - Bush 154 - ?? 215

Leip Atlas (764 user predictions compiled) (c.7/14):
Kerry 264 - Bush 274

Pollkatz (7.15):
Kerry 356 - Bush 182

Race2004 (new) (7/21):
Kerry 286 - Bush 252

Rasmussen (7/12):
Kerry 254 - Bush 197 - ?? 87

Larry Sabato ("June" - not updated since last survey):
Kerry 274 - Bush 264

Samboni's State-by-State (7/19):
Kerry 337 - Bush 190 - ?? 11

Wayne in Missouri (7/16):
Kerry 322 - Bush 216

Dave Wissing (7/20 - updated):
Kerry 306 - Bush 232

2.004k.com (7/18):
Kerry 302 - Bush 236

Two weeks ago, with 18 projections surveyed (including my own), 11 had Kerry winning, 4 had Kerry leading, 1 had Bush winning, and 2 had Bush leading. This time, of the 19 projections I surveyed, 14 show Kerry winning (although one of these had not been updated from 2 weeks ago), and 3 others show Kerry ahead. Only 1 shows Bush winning, and 1 other (not updated) shows Bush with a lead.

(Note: Many of these sites differentiate between "solid", "slightly" and "leaning" states, but I've combined them all together -- for specifics, and the site's methodology, use the links and take a look.)

If anyone has links for any other sites that regularly track Electoral College status, please feel free to send them my way and I'll be glad to add them to the list.

Update: Updated the numbers for Dave Wissing.


Another update (7/20): Found another site:

Federal Review (7/20):
Kerry 292 - Bush 246

And again (7/21): I just realized that I had included Tradesports in a previous survey, but not in the last one (because the site was unavailable at the time). Also, someone mentioned Intrade in a comment on Swing State Project -- although it seems to be a separate site from Tradesports (they have different Irish license numbers) their trading figures are precisely the same, as far as I can tell.

Tradesports (7/21):
Kerry 264 - Bush 274

Updating my count:

21 sites

Kerry winning: 15
Kerry ahead: 3
Bush winning: 2
Bush ahead: 1

Addendum (7/21): It's interesting to note that the Tradesports/Intrade results are precisely the same as those compiled from 764 user predictions at Dave Leip's Atlas of Presidential Elections. I assume that this means it represents a sort of collective conventional wisdom about the current state of the race.

The result gives Pennsylvania and New Hampshire to Kerry, but Florida and Ohio to Bush -- but it's worth noting that the trading results show an extremely thin advantage for Bush in Florida (50.2% chance of his winning there), and a somewhat stronger one in Ohio (56.0% chance). New Hampshire is also close (48% chance of Bush winning), and Pennsylvania less so (42.3% chance of Bush winning).

So the apparent collective conventional wisdom says that Bush is ahead right now, 274 to 264, but it's a razor thin margin, with Florida teetering.


Note (7/22): On Tradesports, Florida's Bush contract is currently at 49.0, down from 50.2, which means that the state changes hands and the current electoral vote total from Tradesports is

Kerry 291 - Bush 247.

Obviously things are very fluid.

[There's some controversy about what the best method is for converting the Tradesports contract data into an electoral college result. For a sometimes informative and sometimes snippy discussion (mea culpa)on that subject, see the comment thread here. Suffice to say that for my purposes here I've done the conversion by assigning any state in which the value of the Bush-wins contract is over 50 to Bush, and any state under 50 to Kerry. That seems to me to be the simplest and most transparent way to go about it, but there is disagreement about that, and another method has been suggested as being better. At this time, I am unconvinced of its merits.]


More...: I see that Princeton Professor Sam Wang (mindgeek on dKos), has added an electoral vote map to his meta-analysis I cited in another post, which wasn't there (I don't believe) the last time I was there, which means he now has a specific projection I can add here (it's the FL & NH but not OH scenario):

Sam Wang (7/18):
Kerry 291 - Bush 247

22 sites

Kerry winning: 17
Kerry ahead: 3
Bush winning: 1
Bush ahead: 1

And again...:

Benjamin Schak (7/17):
Kerry 316 - Bush 222

23 sites

Kerry winning: 18
Kerry ahead: 3
Bush winning: 1
Bush ahead: 1

Update (7/24): There's been a slew of polls, some of which have provoked changes in my prediction, and probably in those of other sites as well (a number of which use the latest poll to assign each state's votes), so I think I'll be doing a new survey in a couple of days.

In the meantime, my newest prediction, along with my notes on the changes, is here.

The executive summary? Florida moves from Kerry to toss-up, assigned to Bush and Missouri goes from Bush to toss-up, assigned to Kerry, which gives us:

Kerry 264 - Bush 211 - ?? 63 (FL, MO, OH, WV)

or, with toss-ups assigned

Kerry 291 - Bush 243