New State Senate Prediction 5/17/06

Our latest State Senate Prediction calls for the Democrats to hold the Senate 18-17.

There actually is a 40% probability that they will pick up another seat instead of just holding all of their seats.

It may be too much of a hill to climb for the GOP to defend basically five open seats (there is a major rebellion going on against former Senator Norma Anderson in SD-22) and two seats that are basically toss-ups.

2006 Colorado State Senate Competitive Race Chart 05/17/06

This chart is based upon our earlier work.

You can look at it over at Polstate.

HD 26: Toss-up

I'm going to move the race in HD-26 (Lakewood) into the Toss-up category. GOP and Democratic registration is basically tied in that district. This district opened up when Rep. Betty Boyd (D) was appointed by a vacancy committee to SD-21 after Sen. Deanna Hanna resigned from office in the wake of a scandal.

The current matchup is:

Kerr (D) vs. Rhodes (R)

Financials:

2004: Coming Soon.

2006:

Kerr (D) coming soon.

Rhodes (R) -$272.57

HD 55: Behind Enemy Lines.

House District 55 is the single most likely district to change hands this fall.

It is currently held by Rep. Bernie Buescher (D).

In 2004 the Democrats picked up HD 55 with 55% of the vote in a two way race.
In 2002 the Republicans held HD 55 with 78% of the vote.

As of April the GOP holds a 21% registration advantage in this district.

Our current model predicts that this seat has an 81% likelihood of returning to Republican hands.

Finance Info:

Current:
Buescher (D) $42,372
Bob Caskey (R) $341

2004:
Buescher (D) $139,892
Opponent (R) $171,578

Current Financials from Colorado Sec. of State
2004 Financials from Follow the Money.org

New State House Prediction 5/15/06

Our latest State House Prediction calls for the GOP to take back the House 33-32.

Well it really isn't that specific.

It actually calls for the Democrats to win 32.4 seats, so we can basically call it a toss up.

2006 Colorado State House Competitive Race Chart 05/15/06

This chart is based upon our earlier work.

You can look at it over at Polstate.

Maps to the State House.

Here is a raw map of the Colorado State House Districts in Metro Denver. A color coded map will be out soon.
State House Districts:Metro Area Raw

Also, the guys at Soap Blox have done a great job with some interactive maps. Here is a link to their page.

Brian Peppers!!!

We are taking today off, look for a finished profile of all 65 house seats and all of the state senate seats by monday, plus you will get a competitive race chart and maps.

Delphi on the Colorado State House 2006

Denver County (HD 1-9), Boulder (HD 10-13), Colorado Springs (HD 14-21), Jeffco (HD 22-29), Adams, Arapaho & Douglas Counties (HD 30-45), Parts of Pueblo (HD 46 & 47), Larimer and Weld Counties (HD 48-53), Mountains and West Slope (HD 54-62)and the Eastern Plains (HD 63-65).

Partisan Differential=Registration advantage as a percentage between the two parties. For example, if 35% of the Registered voters are Republican and 30% are Democrat than the Partisan Differential will be R +5.

Any time the model spits out a probability either equal to or above 100% we call the race "Safe" and any probability between 59% and 50% is a "TOSS UP".

This model is woefully incomplete right now. It only uses two variables (PDiff and a bivariate code for who won) and it only uses data from 2004. Version two will be out within a fortnight and it will include 2002 data plus other variables if they are statistically significant.

Colorado House Phase 1 Chart

Phase one done, predictions for each State House District coming...

Our current model looks like this:

Y=4.5675 * X + -.3446

Y= the probability that either a Democrat will win in the district (-1) or that a Republican will win (1).

X=the registration advantage one party has over the other.

According to this model the Democrats have six seats that, logically, if independents voted 50-50, should be Republican due to registration advantages. The Republicans have no seats that have a Democratic advantage.

It also appears that during the 2004 election the Democrats won on average of 60% of the unaffiliated vote.

However, this still needs some work, since this model only explains about 60% of the variation in Y.

Make the Colorado Legislative Update Sunday.

Somebody bit off more than he could chew...

Colorado State Legislature Projection Update Coming on Friday.

We've set up a Photobucket account so that we can give you maps a la Election Projection.

I'll be completely redoing our state legislative projections. This will be phase one. In it I will build a model that will predict the likelihood of either a Democrat or a Republican winning in a given district based upon a model created from the October 2004 voter registration numbers.

This will give us a base-line for competitive and non-competitive districts.

In phase two I'll be adding in the 2002 voter registration numbers (as soon as the SOS's office gets them to me).

In phase three I'll add in fundraising numbers.

Throughout I'll be making sure that the statistics I use are statistically significant.

These models will be powered by Open Office.