Tuesday, December 20, 2005

State House Closest Races 2004

District Party in Power Win Margin % Win Margin Raw Number
HD 23 Democrat .2% 41
HD 52 Republican 1.3% 484
HD 29 Democrat 1.8% 520
HD 47 Democrat 3.6% 1107
HD 22 Republican 3.6% 1157
HD 31 Democrat 3.4% 1177
HD 27 Republican 4% 1307
HD 50 Democrat 6.2% 1356
HD 30 Democrat 7.8% 1490
HD 64 Democrat 7.4% 1929
HD 33 Republican 5.4% 1972
HD 36 Democrat 36 10.4% 2483

Colorado State House Highest Turnout 2004

HD 43 96.54%
HD 44 96.03%
HD 37 95.58%
HD 45 94.95%
HD 65 93.93%
HD 38 93.68%
HD 33 93.13%
HD 25 92.92%
HD 20 92.64%
HD 31 92.61%
HD 48 92.40%
HD 40 91.66%

Monday, December 12, 2005

Public Service Announcement.

We can't be serious all the time.

Check out this potato song.

Friday, December 09, 2005

Colorado House Homepage.

The Colorado House has 65 members. Currently the Democrats hold the majority 35-30. 33 seats are needed for conrol. Over the next few weeks each district will be examined, triage will occur, and we will come up with a competitive race chart...

...............THIS PAGE IS STILL UNDER CONSTRUCTION......................
Last Update 12/12/2005
Seven County Metro Area (Currently 11-R 25-D):

Adams County: 1-R 5-D
HD-30 Likely Democrat
HD-31 Competitive Democrat
HD-32 Likely Democrat
HD-33 Very Competitive Republican
HD-34 Non Competitive Democrat
HD-35 Non Competitive Democrat

Denver County: 0-R 9-D
HD-01 Non Competitive Democrat
HD-02 Non Competitive Democrat
HD-03 Non Competitive Democrat
HD-04 Non Competitive Democrat
HD-05 Non Competitive Democrat
HD-06 Non Competitive Democrat
HD-07 Non Competitive Democrat
HD-08 Non Competitive Democrat
HD-09 Non Competitive Democrat

Jefferson County W/ Broomfield: 4-R 4-D
HD-22 Very Competitive Republican
HD-23 Very Competitive Democrat
HD-24 Likely Democrat
HD-25 Likely Republican
HD-26 Likely Democrat
HD-27 Very Competitive Republican
HD-28 Likely Republican
HD-29 Very Competitive Democrat

Western Region: 5-R 3-D
HD-54 Non Competitive Republican
HD-55 Competitive Democrat
HD-56 Likely Democrat
HD-57 Non Competitive Republican
HD-58 Likely Republican
HD-59 Likely Republican
HD-60 Non Competitive Republican
HD-61 Likely Democrat

Northern & Eastern Regions: 6-R 2-D
HD-48 Non Competitive Republican
HD-49 Competitive Republican
HD-50 Competitive Democrat
HD-51 Non Competitive Republican
HD-52 Very Competitive Republican
HD-53 Non Competitive Democrat
HD-63 Non Competitive Republican
HD-65 Non Competitive Republican

Southern Region: 8-R 5-D
HD-14 Non Competitive Republican
HD-15 Non Competitive Republican
HD-16 Non Competitive Republican
HD-17 Non Competitive Republican
HD-18 Competitive Democrat
HD-19 Non Competitive Republican
HD-20 Non Competitive Republican
HD-21 Non Competitive Republican
HD-45 Non Competitive Republican
HD-46 Non Competitive Democrat
HD-47 Very Competitive Democrat
HD-62 Non Competitive Democrat
HD-64 Competitive Democrat

Competitive Race Chart:

Non Competitive & Likely Republican: Total 25 HD 14-17, 19-21, 25, 28, 37-40, 43-45, 48, 51, 54, 57-60, 63 & 65.
Competitive Republican: Total 1 HD 49.
Very Competitive Republican: Total 4 HD 22, 27, 33 & 52.
Very Competitive Democrat: Total 4 HD 23, 29, 31 & 47.
Competitive Democrat: Total 5 HD 18, 50, 55, 56, 64.
Non Competitive & Likely Democrat: Total 26 HD 1-13, 24, 26, 30, 32, 34-36, 41, 42, 46, 53, 61 & 62.

BETA Projected Makeup of Colorado State House: R 28-33 D 32-37

Wednesday, December 07, 2005

Southern Region of the United States.

We have divided the United States into four regions in order to more effectively go about our state-by-state analysis.

Our southern region consists of Texas, Louisiana, Mississippi, Alabama, Georgia, Florida, South Carolina, North Carolina, Tennesse and Arkansas. The southern region consists of 10 states that account for 120 of the 435 members of congress (27.6%).

Macro: Of the 120 House members that represent this region, 49 are Democrats and 71 are Republicans. This region's Senate delegation is made up of 16 Republicans and 4 Democrats. This region accounts for 27.6% of the US House and 20% of the United States Senate. John Kerry did not carry a single southern state in 2004.

Micro: This region only has ten competitive House races.

TX-17 Rep. Edwards (D) 8th Term 2004: % of vote 47% R 51% D CPI R+18
TX-22 Rep. DeLay (R) 11th Term 2004: % of vote 55% R 41% D CPI R+15
LA-7 Rep. Boustany (R) Freshman 2004: % of vote 55% R 45% D CPI R+7
LA-3 Rep. Melancon (D) Freshman 2004: % of vote 50% R 50% D CPI R+5
GA-12 Rep. Barrow (D)Freshman 2004: % of vote 48% R 52% D CPI D+5
GA-8 Rep. Marshall (D) Freshman 2004: % of vote NA% R NA% D CPI R+21
FL-9 OPEN 2004: Republican Unopposed CPI R+4
FL-13 OPEN 2004: % of vote 55% R 45% D CPI R+4
FL-22 Rep. Shaw (R) 13th Term 2004: % of vote 63% R 35% D CPI D+4
NC-11 Taylor (R) 8th Term 2004: % of vote 55% R 45% D CPI R+7

Thursday, December 01, 2005

Western Region of the United States.

We have divided the United States into four regions in order to more effectively go about our state-by-state analysis.

Our western region consists of Alaska, Hawaii, California, Oregon, Washington, Idaho, Nevada, Arizona, Utah, New Mexico, Colorado, Wyoming and Montana. Thirteen states that account for 98 of the 435 members of congress.

Macro: Of the 98 House members that represent this region, 53 are Democrats and 45 are Republicans. However, this region's Senate delegation is made up of 15 Republicans and 11 Democrats. This region accounts for 22.5% of the US House and 26% of the United States Senate. This region voted 9-4 in favor of President Bush in 2004.

Micro: This region only has five competitive House races.

WA-8 Rep. Reichert (R) Freshman. 2004: % of vote 52% R 47% D CPI D+2
WA-2 Rep. Larson (D) 3rd Term 2004: % of vote 34% R 64% D CPI D+3 w/ nominal opp.
NM-1 Rep. Wilson (R) 4th Term 2004: % of vote 54% R 46% D CPI D+2
CO-3 Rep. Salazar (D) Freshman. 2004: % of vote 47% R 51% D CPI R+6
CO-7 OPEN 2004: % of vote 55% R 43% D CPI D+2

It also appears that there will be two competitive Senate elections, Montana and Washington.

Monday, November 28, 2005

In the hopper...

Three projects are going on at the same time here at MHD.

First Matrow is finishing his work on determining if there is a correlation between how a county voted in the 2004 Presidential election and how it voted in the Referendum C election. When he's done I'll set up a table over at polstate.com. We are also working on creating a few jpg maps.

Second, we will begin doing our state-by-state analysis of the 2006 election. At the same time we will also do a systematic analysis of the Colorado State House and State Senate races.

Thirdly, we've got a project in the hopper that will be reviewing a report that we keep seeing in the local papers.

We will release the state-by-state analysis beginning this week.

Meanwhile, here is an exclusive look at how Prof. Snape killed Dumbledore.

Tuesday, November 22, 2005

Opening Odds on Control of Congress in 2006...

We are using Tradesports current market prices to set our opening odds for control of the United States Senate and House.

Currently the GOP is favored to keep control of both the House and the Senate.

Senate Probability of GOP Control: 74.2% Fractional: 5/14
House Probability of GOP Control: 72.8% Fractional: 4/11

Over the course of the winter we will be doing a state by state analysis of all 435 House seats and all of the open Senate seats. This analysis will be cross posted at polstate.com.

Monday, November 21, 2005

Referendum C State-Wide Analysis...

Here is our complete initial analysis of the 2005 Referenda C and D election. We are dividing Colorado into five regions, the Denver Metro Area, Northern Region, Eastern Region, Southern Region and the Western Region.

Referendum C:

State-Wide Yes 52.0% No 48.0%
Metro Area Yes 54.1% No 45.9% MHD Index = +2.1
Northern Region Yes 52.1% No 47.9% MHD Index = +.1
Southern Region Yes 48.7% No 51.3% MHD Index = -3.3
Eastern Region Yes 42.8% No 57.2% MHD Index = -9.2
Western Region Yes 56.8% No 43.2% MHD Index = +4.8

Western Region.

Our Western Region consists of 27 counties: Archuletta, Chaffee, Clear Creek, Delta, Dolores, Eagle, Garfield, Gilpin, Grand, Gunnison, Hinsdale, Jackson, Lake, La Plata, Mesa, Mineral, Moffat, Montezuma, Montrose, Ouray, Park, Pitkin, Rio Blanco, Routt, San Juan, San Miguel, Summit.

Referendum C Won in the Western Region 56.8-43.2. The Eastern Region accounted for 17% of the total state-wide vote.

Eastern Region.

Our Eastern Region is made up of 16 counties: Baca, Bent, Cheyenne, Crowley, Elbert, Kiowa, Kit Carson, Lincoln, Logan, Morgan, Otero, Phillips, Prowlers, Sedgewick, Washington and Yuma.

Referendum C Lost in the Eastern Region 42.8-57.2. The Eastern Region accounted for 5% of the total state-wide vote.

The maximum percentage of the vote that Referendum C got in any county was 69.5%, the minimum was 31.4%, with a mean of 51.3%.

Thursday, November 17, 2005

More Referendum C Goodness.

Last week we looked at how the seven county Denver Metro Area voted in the Referendum C campaign.

The Denver Metro Area cast 53% of the total ballots.

Referendum C passed in the Denver Metro Area 54.1-45.9 vs. a State-wide vote of 52-48.

Our Northern Region consists of two counties, Larimer and Weld. In this Region Referendum C passed 52.1-47.1 and it accounted for 11% of the state-wide vote.

Our Southern Region consists of eleven counties. In this Region Referendum C lost 48.7-51.3 and it accounted for 17% of the state-wide vote.

On Monday we will go over the Eastern and Western Regions.

Later in the week we will look for correlations between each county's MHD Diff. and its Cook Partisanship Index.

In the meantime, look here for some G rated Paris Hilton goodness.

Friday, November 11, 2005

Regional Analysis of the 2005 Referenda C and D Election.

Over the next week we will be releasing an analysis of the 2005 Referenda C and D election. We are dividing Colorado into five regions, the 7 county Denver Metro Area, North Central, Eastern Plains, South Central and the Mountains.


7 County Denver Metro Area:

Referendum C:

State-Wide Yes 52.0% No 48.0%
Metro Area Yes 54.1% No 45.9%

MHD Index = +2.1

Source: Colorado SOS.

Thursday, November 10, 2005

In the hopper...

We are going to start doing updates about once a week (either on Mondays or Fridays). Tomorrow we will release a regional analysis of the Referendum C and D election.

After that we will begin our predictions for the Colorado State Senate, Colorado State House, United States Senate and United States House contests and we will start our war on the statistical illiterati.

Friday, November 04, 2005

Looking at the Governor Races of 2005.

In 2005 there are only two races for Governor in the United States, New Jersey and Virginia.

According to Tradesports.com the Democrats have a 1/13 chance of winning the seat. That's a 93% probability.

Also according to Tradesports.com the Virginia race looks like a toss-up.

You can go here to look at Prof. Larry Sabato's analysis of these races. The GOP currently has a 28-22 advantage over the Democrats in Governorships. If the GOP picks up either Virginia or New Jersey it will increase that margin.

Thursday, November 03, 2005

Caldara gives post election analysis; echoes MHD...

As you may know from our pre-election analysis, it is our belief that Independence Institute President Jon Caldara is one of the two consuls for the Republican side in Colorado's current political war.

Last night on his 850 KOA radio show Jon Caldara lead what was a mixture of post election analysis, pep rally and a wake.

He began by stating that he felt like "he let [his supporters] down" by losing the battle over Referendum C while he claimed a partial victory in the defeat of Referendum D. "We didn't do that bad considering that we faced the entire Democratic establishment and a minority of the GOP."

When it came to his actual analysis he said that "the state is moving to the left."

He cited Public Choice Theory as the reason for the Referendum C victory. "Had this election taken place during a even-year this wouldn't have even been close...we matched them in the air war but we lost the ground war...we have no GOTV."

A caller called in and dubbed yesterday the "2nd Black Wednesday" referring to the Caldara legion's loss. Caldara asked what the 1st Black Wednesday was. "A year ago," replied the caller. "Yes, your right" stated Caldara, "A year ago we lost the State Legislature, a Congress Seat and a Senate Seat." We call that election Colorado's Cannae, but "Black Wednesday" certainly has a ring to it.

Caldara also gave special congratulations to two supporters of Referendum C. He stated that without the support of Governor Owens, bringing his GOP legions over to the pro-C side, the ballot item wouldn't have passed. He also stated that the tracking polls showed that when Denver Mayor John Hickenlooper jumped out of the plane, support for C jumped.

Wednesday, November 02, 2005

Split Decision!!!

Update 16:30:

Look to the right for a report card on how Mile High Delphi did this election cycle. Our sophomore effort looked pretty good (simple models work better than big complex ones). Now we need to focus on the State Legislature...

Update 15:45:

We've done a breakdown of our B Market. The predicted results for the percentage of the vote for Referendum C was 51.12%. The actual election result was (with 99.9% of precincts reporting) 52.03%. A difference of .91%. Not too bad, with 58 participants it sure did get close.

Original post:

The results are in and it appears that Referendum C passed and its sister item Referendum D failed.

It is our belief that this will set up a great battle in 2006 (since this election was basically fought to a stalemate).

We will have analysis of the election up throughout the weekend. I'm working on a color-coded map of Colorado (we will put it up both here and at polstate.com), Blogicus is working on a regression model, and our newest partner, Salacious Crumb, will be typing up the analysis.

We will also have the results of our markets and our winners up.

Early next week we will have a look at other elections across the United States.

Tuesday, November 01, 2005


Here are some links to the results of the 2005 elections.

Rocky Mountain News.

Colorado Secretary of State.

Here is a look at how Mile High Delphi did in predicting the results of this election:

Econometric Models:

Ref. C Predicted:49.4% Actual:52%

95% Range: 46.9%-51.9% Off: 2.6

Ref. D Predicted:48.7% Actual:49.4%

95% Range: 46.2%-51.2% Off:.7

Futures Market A

Ref. C Predicted:51.12%Actual:52.03%

Futures Market B

Ref. C Predicted:48% Actual:52%

Ref. C and D Homepage

Early numbers now good for C and D.

Update 9:00:

With over half of the precincts reporting Referendum C looks good. Ref. D still tight.

Looks like a big win in the Metro Area will put C over the top.

Original post:

Early numbers are starting to trickle out. It appears that C and D are going to be about 20,000 votes short of passage according to these very preliminary projections.

In the metro area it appears that C is getting about 51.4% of the vote. About 305,000 votes are projected to be cast for C in the seven county metro region.

However, unless the rest of the state votes 48.76% for Referendum C the issue will lose. It currently appears that the rest of the state is polling more like 45.5% for the ballot issue.

Coincidentally, if the City and County of Denver had an all mail election, with a turnout of 50%, instead of the projected 42%, Referendum C would have gotten about 14K additional votes.

These numbers are very preliminary, keep tuned in for more updates throughout the night.

CO: Analysis of Election 2006; A Modern Punic War.

Let us tell you a story from 22 centuries ago...

During this time Rome was still building its commonwealth, but it only ruled over the Italian peninsula. Eventually a war broke out over control of Sicily. Rome fought Carthage (at that time another great power) for years during the 1st Punic War. Rome didn't utterly defeat Carthage, but they did win the war, and they imposed humiliating terms.

During that time a young Carthaginian, Hannibal, watched as his father, a general for Carthage, was forced to surrender to Rome when he had never lost a battle. Hannibal swore a blood oath to his father, that he would humiliate Rome one day for what Rome had done to Carthage.

Hannibal eventually concurred modern day Spain for Carthage, went to war against Rome, beat the Romans in every battle that he fought in Italy, decimating an entire generation of Roman men. Eventually Scipio the younger arose to face Hannibal, defeating him in Africa (after Hannibal left half his troops in Italy). Even in death Hannibal defied Rome, he committed suicide rather than be taken prisoner.

How the hell does this relate to modern day Colorado; well we think that the different factions in the huge political war that is swelling up in Colorado are not that much unlike the factions and actors in the 2nd Punic War.

Setting the Stage:

After Hannibal crossed the Alps with his army and battle elephants, he defeated Rome several times. At that time Rome was governed by two Consuls. Imagine that the United States had two presidents, each with a separate army, that's about how the Roman system worked. Several times a consul would come up with his army and face Hannibal. Even though he had less troops and was on enemy ground, Hannibal won time and time again, mainly because of his tactical superiority. Hannibal had a professional army, unlike Rome's citizen soldiers. He also was a military genius, unlike the politicians who commanded Rome's military might.

Eventually Rome formed an army of over 80,000 men. As they approached the plains of Cannae, where Hannibal set up with his force of about a fourth of that, Hannibal hatched a battle plan that would strike fear into the heart of Rome for a half a millennium. The Romans surged ahead in a simple attack. Hannibal spread his troops around and encircled the larger force, by the end of the day Rome had lost a generation.

In 2004 the Colorado GOP had its Cannae. The GOP has a demographic advantage in Colorado. But their superior numbers cannot make up for that fact that the Democrats have proven time and time again that they have a tactical superiority. Simply put, the Democrats are better at politics than the Republicans. In 2004 the GOP lost control of the State Senate, State House, the Third Congressional Seat and one of the State's Senate Seats. Today's election is a battle in this war. To understand the importance of the passage of failure of Referendums C and D you have to understand the different players in this battle and what motivates them.

The Players:

We have divided the field into five factions:

The Hoi Polloi is the largest single faction but the weakest. They are simply the multitude. No one side can win without them, but they are fickle and schizophrenic.

We divide the Democrats into two factions:

The McGovernite wing of the party believes that the Democrats can win in Colorado by simply being liberal. They see the defeat of Referendum C and D as a win because they can then go directly after TABOR, since, by their logic, the Democrats rode the state's fiscal woes all the way to victory in 2004. If C and D win that theory will be vindicated, they feel, and that additional victory will add momentum to the 2006 election. Two examples of this line of thought are available here. Colorado Luis believes that the Democrats should abandon working with the GOP, get rid of TABOR and institute a progressive state income tax. The Wash Park Prophet has put forward a plan that really fleshes out some of the goals of the McGovernite wing of the party.

The Hannibal Democrats understand that they are too strong for the GOP to defeat, like the historical Hannibal, and that they are not strong enough to defeat the GOP and build a lasting majority. They are outnumbered, they can win but they cannot concur. For these Democrats the amazingly close Referendum C and D election is a example of how they must tread lightly in Colorado. They were surprised that the hoi polloi, after putting them in power, wouldn't want them to fix the fiscal mess that got them there (this contradiction troubles them). Hickenlooper, Joan Fitz-Gerald and others who either haven't announced for Governor or who have ruled it out are in this camp. They are trying to read the hoi polloi, they figure that after the Referendum C and D election they will be able to evaluate the battleground for the 2006 elections.

The GOP is also divided, but unlike the Democrats. While the Democrats are divided in their beliefs as to how strong they are, they are united in where to go and how to get there. The GOP is like Rome, they have two consuls and two different armies.

The two consuls are the President of the Independence Institute, currently Jon Caldara and the nominal leader of the state GOP, currently Governor Owens.

Caldara and his allies are best defined by this piece by Mike Rosen.

Owens and his allies agree with the McGovernite Democrats, they think that the GOP lost power in 2004 because of the fiscal crisis. Here is a quote from The Economist that outlines the beliefs of Owens and his faction:

The difficulty is that you can squeeze only so much before services are cut beyond what most people- even in libertarian-inclined Colorado -consider sensible. A no vote in November would force Mr. Owens into the kind of budget surgery that makes governors deeply unpopular. For the moment he affects not to worry: "I'm not running for president," he says; "The backlash won't be against me, but against my party."

As you can see the Owens faction and the McGovernites believe that the fiscal crisis issue is what moves the hoi polloi. The Hannibals and the Caldara faction do not.

Monday, October 31, 2005

Colorado Election Projection: Ref. C and D 2005

Referendum C is Too Close To Call

Projected Results
Yes: 49.4% No: 50.6% Probability: See Footnote.

Referendum D is Too Close To Call

Projected Results
Yes: 48.7% No: 51.3% Probability: See Footnote.


The reason why the races are too close to call is because the predicted vote percentages are the mean points of a zone of probability. For each predicted value (excuse the mathematics) the 95% zone of probability is -/+ 2.5.

For example, our current projection on Referendum C is that it will get 49.4% of the vote. However we can only predict to 95% accuracy that the percentage of the vote that Referendum C gets will be between 46.9% and 51.9%. The same spread applies for Referendum D. A good rule of thumb then would be that if the predicted result falls below 47.5% then we will be able to be 95% confident that the issues will fail (and vice versa if the predicted result is above 52.5%.

Final Fundraising Numbers...

Here are the final fundraising numbers from the two sides of the Referendum C and D battle...

Pro Side

Vote Yes on C and D

Total expenditures as of Late September Report (9/28) $3,722,000
Early October expenditures: $1,173,000
Total expenditures as of Early October Report (10/17) $4,895,000
Funds on hand as of 10/17 $477,000

Con Side

Vote No; It's Your Dough

Total expenditures as of Late September Report (9/28) $130,000
Early October expenditures: $123,000
Total expenditures as of Early October Report (10/17) $253,000
Funds on hand as of 10/17 $15,000

If C Wins, You Lose Committee

Total expenditures as of Late September Report (9/28) $222,000
Early October expenditures: $104,000
Total expenditures as of Early October Report (10/17) $326,000
Funds on hand as of 10/17 $104,000

Colorado Club for Growth-Issue Committee

Total expenditures as of Late September Report (9/28) $582,000
Early October expenditures: $626,000
Total expenditures as of Early October Report (10/17) $1,208,000
Funds on hand as of 10/17 $19,000

As of these reports the Vote Yes on C and D campaign has outspent the allied opposition by a factor of about 2.75 to one. Normally to get a ballot measure to pass you need to outspend your opposition by 10-1, however, the Yes on C and D side has spent so much money that they are probably facing as many problems from diminishing returns as they are from the opposition.

Tomorrow we will do a projection based upon county by county projections (we are kinda lazy so some counties, such as the eastern plains will just be grouped together). Look for it around noon.

Also, check polstate.com for an updated version of our Referendum C and D homepage.

Alea iacta est...

Alea iacta est...the die has been cast.

Well not yet, but we are getting close.

With no new polling information out yet our final projection will stay what it is. Its our sophomore effort, lets see how much better we are than last year.

We are being flooded with new requests for the futures market so we've extended the trading deadline until 5:00 pm tomorrow. That's it. No more extensions, don't even ask.

You can also submit your B market guesses up to 5:00 pm tomorrow. Be sure to get them to us, we want to put the mean and the standard deviation up before polls close tomorrow so that we can test the B market in real time.

We will also be putting up an analysis of this election cycle tomorrow during the day. This is a combined effort of a historian, a MBA, a banker and a biology teacher, even I don't know exactly how it will end up.

Thanks everyone for playing, and if you haven't yet, get out and vote!

Wednesday, October 26, 2005

Futures Market Info...Five Days to Go.

The Futures Markets will be open all weekend long and until 5:00 PM Mountain Time on 11/1/05.

The A market is the regular futures market. Volume has picked up of late. The payout for that market is $100 to the winner, $50 to second place and $25 to third. Read the FAQ's to find out more.

The B market is simpler and we really would like at least 100 participants in it. All you have to do is guess what percentage of the vote Referendum C will get at the polls. You must guess to within 1/100 of 1%, for example you would guess that it will get 52.14%. All you need is an e-mail address to register. The winner of this market gets $100.

Thanks for reading and playing.

Tuesday, October 25, 2005

One week to go

There is only one week left until election day in Colorado. We are going to extend our time period for both of the futures markets until 5:00 pm Mountain Time on Monday October 31st.

Also on Monday we will release our final projection for the outcome of the Referendum C and D election.

For those of you who are political junkies here are a few tidbits to chew on...

While sorting through the information in the Denver Post poll from October 17th, we found some details that make the passage of C and D less likely.

The story contained this paragraph:

...the poll uncovered several bad omens for backers of the ballot measures: too many undecided Republicans, slipping support in the Denver suburbs and a majority of voters who say they'd accept across-the-board budget cuts.

Which points out the regional problem that the Yes on C campaign is having. You don't have to take Mile High Delphi's word, listen to Brad Coker:

The narrow margin suggests that the ballot measures are likely to lose because undecided voters typically opt for the status quo, Coker said. Another factor is Republicans compose the biggest group of undecided voters.

"Unless it's over 50 percent, I'm always skeptical," Coker said. "It could pass. It could fail. You can't really make a call."

The poll also confirmed a pattern that, if it holds through to the general election on November 1st, will spell defeat for the C and D campaigns.

Recently poll numbers have shown the ballot measure picking up steam. For example support for Referendum C increased by four percentage points from July until mid October. During that period most other ballot initiatives have lost support in prior years elections. However, it appears that the pickup isn't spread evenly around:

The lead is widening in strongholds such as Denver. In July, Referendum C had 56 percent support; now it has 66 percent.

But the ballot measure is falling further behind in the suburbs.

In July, poll respondents in the north and east suburbs were evenly split at 41 percent on Referendum C. Now, 43 percent support and 46 percent oppose the measure.

In the south and west suburbs, the number of poll respondents planning to vote against Referendum C has risen from 43 percent to 48 percent.

This discrepancy makes the passage of C and D increasingly unlikely.

Simply put, 66% support with 42% voter turnout doesn't get you to 50% state wide when the rest of the state votes 55-45 with 50% turnout.

A Rocky Mountain News story by Stuart Steers demonstrates this fact. The way this election works, it pays for the supporters of C and D to rake up huge majorities in Denver and Boulder counties, however only four of the metro counties will be voting by mail, two of which, Douglas and Arapahoe, are widely expected to vote against C and D. "Typically, in off-year elections, mail balloting brings in a much larger turnout than traditional voting at percincts." Boulder will be voting by mail, but Denver, which the yes on C campaign is depending on to get over the hump is voting by percinct. "'We expect the turnout to be consistent with other off year elections, around 42%,' said Alton Dillard, spokesman for the Denver Election Commission." With Denver unable to rake up a huge Yes count, Douglas, Arapahoe and El Paso Counties should easily be able to neutralize the effects of Denver and Boulder, making the rest of Colorado the real battleground, and basically consigning C and D to defeat.

Another Stuart Steers story available here reveals how each campaign acknowledges this structural problem for C and D.

For more detail into the structural problems facing C and D I suggest you read this Ciruli Issue Brief. Especially go down to the table labeled "Turnout and Required Vote for Proponents." It gives a much more detailed analysis of the structural problems that C and D face.

Thursday, October 20, 2005

Futures Market Info...

For those of you who don't know, we are doing three different experiments this election cycle.

The first one is the typical modeling that we have done before that uses econometric modeling to try and predict election outcomes. We had some success last cycle, but we are hoping to improve dramatically (mainly through better data).

The second experiment is a full on futures market. Like the Iowa Presidential Futures Market, or Tradesports.com, we are looking to see if a market can predict the winner of an election.

The third experiment is where we are asking participants to give us a guess at what percentage of the vote Referendum C gets. We want to see if these guesses will have any predictive power.

We have cash awards for the best three performers in the futures market and a simple $100 cash prize for whoever guesses closest on the third experiment.

For more info click on the links to the right.

Wednesday, October 19, 2005

When Journalists Use Statistics It Gets Fugly.

We are doing some research right now on a recent RMN story about Colorado's "Achille's Heel." The story threw out a bunch of statistics, which, to the average reader probably looked quite impressive, but, look to us like garbage. Right now we are trying to do research (since the reporter's source also uses reams of garbage stats) that should answer if the statistics really are relevant.

What are we doing? We're looking at the rational behind the article, mainly that the cited "weaknesses" of Colorado have a correlation to negative economic growth, or just slower than average economic growth. Since both the News and the Source failed to give us either a correlation coefficient (which tells us if the statistic is even statistically significant) or a coefficient of determination (which tells us how much one variable explains about another) we will do it for you.

Our hypothesis: That the stats thrown out were garbage. That a state's state and local support for higher education, K-12 spending as a percent of personal income or high school graduation rates have no statistically significant correlation to either job creation, unemployment rate, GDP growth or personal income growth.

If you have any ideas tell us. Do you think that these rankings matter? Do you think that we should compare the state spending stats to another variable that would better measure the vitality of Colorado's economy?

The RMN Story is available here.

Reporter's Source.

Here is a gem from the article:

Here's what Colorado can report as strengths, the Metro Denver Economic Development Corp. found: Colorado is the fourth most highly educated state in percentage of college graduates and the 10th most productive work force by one measure. Property taxes for families, as well as overall business tax burden, are the second-lowest in the U.S.

And the state has the lowest obesity rate in the nation.

On the flip side, Colorado is 48th in state and local support for higher education and 49th in K-12 spending as a percent of personal income. While the population as a whole is well-educated - thanks to an influx of well-educated residents from other states - Colorado is 29th in high school graduation rates, down from 26th in 1995.

Tuesday, October 18, 2005

From Tabor to Today...

Here is a quick look at the off year elections from 1993 until today. Only two passed. Both were bonding measures.

1993: Referendum A Reinstatement of Sales Tax on Tourist-Related Purchases Yes: 44.8% No: 55.2%

1995: Referendum A Multiple-Year Financial Obligations - Nonstate Prisons Yes: 45.1% No: 54.9%

1997: Amendment 1 Tax and Fee Increases for Funding Transportation Projects Yes: 15.8% No: 84.2%

1999: Referendum A Financing State Transportation Projects Yes: 61.7% No: 38.3%

Amendment 26 Surplus Revenue to Test I-70 Fixed Guideway Yes: 34.1% No: 65.9%
Referendum A GOCO Bonds for Open Space Yes: 57.6% No: 42.4%

Amendment 32 Taxable Value of Residential Property Yes: 22.4% No: 77.6%
Amendment 33 Video Lottery/Tourism Promotion Yes: 19.1% No: 80.9%
Referendum A Revenue Bonds for Water Projects Yes: 32.9% No: 67.1%

Source SoapBlox Colorado.

Monday, October 17, 2005

Mike Rosen Op Ed Could Foreshadow Defeat Of C and D.

A recent Op Ed by Conservative Talk Radio Host Mike Rosen outlines what, in hindsight, we may agree were the reasons for Referendum C and D failing at the polls on November 1st; if they do.

Here are a few of the highlights:

No more than a handful of voters are well-versed on the details and nuances of the Taxpayer's Bill of Rights and the state budget process. The rest derive their impressions from activists on each side, fair and unfair "news" stories, opinion commentaries, and 30-second radio and TV spots.

While Rosen may be a conservative, he does do excellent, if partisan, analysis:

While C and D opponents are correct in saying that overall state spending has not gone down in any year because of TABOR, it's also true that inflation-adjusted per capita spending is, in fact, down from prior years in some areas. And while C and D advocates are correct in saying that there will be no increase in statutory tax rates, there will be an increase in effective tax rates under their proposal. Your effective tax rate is what you pay net of your TABOR refund. If the state keeps your TABOR refund and spends it on government programs, your effective tax rate will necessarily be higher...Bottom line: C and D constitute a tax increase.

Colorado's fiscal problems were compounded by Amendment 23, which put K-12 spending on autopilot, regardless of overall revenue shortfalls. That contributed to the squeeze in other areas like roads, transportation and building maintenance. This was good for the teachers unions that contrived Amendment 23, but bad for everyone else.

So Mike Rosen throws out a compromise, mired in the memory of the 100/100 plan that the GOP controlled legislature tried to put out before 2004:

So I'm willing to compromise something I like - TABOR - in exchange for which I expect Democrats to compromise something they like - Amendment 23. Such a deal was offered in 2004 by Gov. Bill Owens and Republican leaders in the state legislature. The 100/100 Plan would have reduced the TABOR refund by $100 million for two consecutive years and also slowed Amendment 23-mandated K-12 spending by the same amount for the same period. Democrats, in tow to the teachers unions, rebuffed it.

This echoes closely analysis that Mile High Delphi has recently done.

In 2005, after Democrats took control of the state legislature, Owens found himself negotiating from weakness. Since a change in Amendment 23 can only be put to the voters in an even-numbered year, Owens believes C and D is the best deal he can get in 2005 and doesn't want to risk a death blow to TABOR next year if C and D fail.

Many wavering Republicans seem to be echoing Rosen's rational. Shot down C and D and go for a better offer in 2006:

A compromise with $2 billion in additional taxes and spending should be sufficient to meet the state's essential needs. It should also include a rainy-day fund to cover future shortfalls and a reduction in Amendment 23 increases - which can be put to the voters in 2006...In a fair compromise, both sides give. So far, Democrats and liberals haven't given anything. So I'm voting "no" on C and D and relying on reasonable Democrats in the state legislature to roll up their sleeves and cooperate in forging a more sensible compromise in the next session.

It may be a little premature, but if C and D go down, I'd re-read this Op Ed. The seeds of C and D's destruction may have been planted in 2003.

Closer look at Friday's Rocky Mountain News Poll.

Here is a closer look at Friday's RMN poll:

Referendum C Definitely Yes: 36% Probably Yes: 13% definitely No: 36% Probably No: 10% Undecided: 5%

Referendum D Definitely Yes: 31% Probably Yes: 19% definitely No: 32% Probably No: 12% Undecided: 5%

This more complete breakdown, from page 4A on 10/14/2005, is not available on the internet. It nicely parallels another question in the poll that found that 31% of voters like C & D and will support them, 33% don't like them all that much but can support them because they could be good for Colorado and 31% who don't like C & D and won't support them.

Sunday, October 16, 2005

New polling on Referenda C and D put issues behind. Ritter (D) favorite for Govs race.

(updated 10/18/2005)

A new Denver Post poll out today disagrees with a Rocky Mountain News poll released two days ago.

Here are the results of the Denver Post poll:

Mason-Dixon Oct 11-13
Referendum C Yes: 47% No: 44% Undecided: 9% MOE:4%
Referendum D Yes: 44% No: 45% Undecided: 11% MOE:4%

These results are interesting because they contrast somewhat with the results from the Rocky Mountain News poll. The RMN poll put support for C at 49% and D at 50%. The results of the Denver Post poll are within the margin of error of the earlier RMN poll, but they pull the average significantly down.

As a result we are further modifying our election projections:

Referendum C is Too Close To Call Yes: 49.4% No: 50.6% Probability: See Footnote.

Referendum D is Too Close To Call Yes: 48.7% No: 51.3% Probability: See Footnote.

The poll also asked who you would vote for in the 2006 Governor's race:

Mason-Dixon Oct 11-13
Ritter (D) vs. Beauprez (R) Ritter: 42% Beauprez: 36% Undecided: 22% MOE:4%
Ritter (D) vs. Holtzman (R) Ritter: 41% Holtzman: 30% Undecided: 29% MOE:4%

Update: We've updated our Referendum C and D homepage so that you can look at the polling history of the campaign and see how the campaign has evolved over the last few months.

Our last headline said that Referendum D was dead, that was a typo. Sorry for any confusion.


The reason why the races are too close to call is because the predicted vote percentages are the mean points of a zone of probability. For each predicted value (excuse the mathematics) the 95% zone of probability is -/+ 2.5.

For example, our current projection on Referendum C is that it will get 49.4% of the vote. However we can only predict to 95% accuracy that the percentage of the vote that Referendum C gets will be between 46.9% and 51.9%. The same spread applies for Referendum D. A good rule of thumb then would be that if the predicted result falls below 47.5% then we will be able to be 95% confident that the issues will fail (and vice versa if the predicted result is above 52.5%.

As for the predicted probabilities, both measures have a slightly higher than 50% probability of failing.

Friday, October 14, 2005

New polling on Referenda C and D. C in statistical tie, D is basically dead.

A new Rocky Mountain News/CBS 4 poll out today shows that support for Referendum C is nearing 50%. Support for D is at 50%.

Here is the poll:

Referendum C: Yes:49% No:46% Undecided:5% MOE:4.38%
Referendum D: Yes:50% No:44% Undecided:6% MOE:4.38%

Oct 10-12 POS

Using our model we predict that the Referendum C race is too close to call. Our model predicts that the outcome should be a 50-50 tie. This also moves the probability of passage to 50%.

As for Referendum D, it is statistically tied. Our model predicts that it will win 51.2-48.8. The probability of D passing has risen to 50%.

Other interesting tidbits from the poll:
- It appears that we have been right all along about the "GOP split" over C, it isn't as large as the media makes it out to be, in fact, surprisingly, it seems to be as large as the Democratic split, which is a huge surprise.

Among Democrats, 62 percent said yes. Republican voters polled 39 percent yes, despite the strong support of GOP Gov. Bill Owens.

- It seems that the electorate is split by thirds, with one third supporting C and D, one third opposing them and one third that is confused. This doesn't jive with the results of the poll and bodes well for the opposition.

Which ONE of the following phrases best describes how you feel about these proposals?

I like them and would support them. 31%

Even though I don't like them all that much I could support them because they could be good for Colorado. 33%

I don't like them at all and could never support them. 31%

Don't know 3%

Refused 1%

-The "Yes on C and D" campaign needs to get on message. The voters aren't sold on the "fiscal crisis" narrative. Over 50% of those polled think that the state is not in fiscal crisis.

And would you say the Colorado state budget is in crisis, has some problems but is not in crisis, or is in pretty good shape?

In crisis 32%

Problems, but not crisis 51%

Pretty good shape 12%

Unsure 5%

Overall this poll confirms what Mile High Delphi has been saying for weeks, this race is a statistical tie at best. The subheadline in the Rocky Article "Supporters seen as having an edge 2 1/2 weeks before vote." seems woefully unsupported by the text of the article.

Wednesday, October 12, 2005

For Caldara and Allies, Less is More.

Not that money matters in these off year elections, but it seems that Jon Caldara's Vote No; Its Your Dough campaign has pulled even in spending with the Vote Yes on C and D campaign guys, maybe even ahead.

How can I come to this conclusion?

The thought first occurred to me on 10/07/2005, while watching Channel 4 News at 10:00. The vote no on c commercials outnumbered the vote yes on c commercials by a margin of 5-2 (and the five vote no commercials ran nearly in a row). Not one of these commercials was paid for by Caldara's campaign, instead they were all paid for by two separate groups, Colorado Club for Growth Issue Committee and If C Wins You Lose.

When you put together the monies spent by Caldara and the allies you get a total of $934,000. Vote Yes on C and D has only spent $3,722,000 (all numbers as of the 9/28 Secretary of State filings). That is a margin of about 4-1, well below the conventional wisdom margin of 10-1 that is needed for a vote yes committee to win against an organized opposition.

Additionally, during the month of September Vote Yes on C and D spent $700,000. The allied front spent $917,000. No referenda item has ever passed when it was outspent by its opposition (during a off-year election).

Sabato Speaks on Colorado Politics.

Larry Sabato, of Crystal Ball fame, has begun his handicapping of the races for Colorado's Congressional Seats.

You can look at his excellent website here.

Three of Colorado's seven seats made it onto his "Dirty Thirty" most competitive races.

He lists HD 3 (Western Slope) John Salazar's (D) seat as "leans Democrat".

HD 4 (Eastern Plains) Musgrave's (R) seat as "leans Republican".

HD 7 (Lakewood, Arvada and Aurora) an open seat as "toss up".

Also, we will be adding three new links, Technorati so that you can easily search blogs, Rossputin for some more Colorado focused commentary and Colorado Exposed which is perhaps the most off the wall political website in Colorado.

Additionally, it has come to our attention that the link to the futures market is broken. We will be fixing that later today.

Tuesday, October 11, 2005

Hoi Polloi and Referendum C...

Here is a little musing on the tyranny of the majority...

If by the people you understand the multitude, the hoi polloi, 'tis no matter what they think; they are sometimes in the right, sometimes in the wrong: their judgment is a mere lottery.
-John Dryden (1631-1700) Essay of Dramatic Poesy

Monday, October 10, 2005

Praise Apollo! The Oracle Speaks on Referendum C.

Our new model predicts that Referendum C has a 28.6% probability of passing.

If the election was held today referendum C would LOSE 47-53.

How did we get here? We constructed a model based upon looking at other off year elections. The only statistically significant variables we could find were all related to polling numbers. This is the result of using complete data sets from only 2003, 2001 and 1999 and partial data from 1997, 1995 and 1993. Using these numbers we constructed a formula that predicted the outcome of every election that we had full data for. Using the Student's t-Distribution we calculated the probability that C would garner at least 50% of the vote at 28.6%.

This is the equation that we used = X+(.2*Y))+/-Z = a

X = sample mean
Y = Undecided voters (as a percentage).
Z = estimator for the population standard deviation

On a less mathematical point, the first "If C wins you lose" yard sign was sighted today at 20th and Denver West Parkway.

Wednesday, October 05, 2005

Delphi on the Battle over Ref. C: Pro Side Needs More Money To Win.

Some members of the Colorado political prognosticati believe that since the Vote Yes on C and D campaign has outraised the Vote No on C side (by dramatic margins in fact) then Referenda C and D have a better chance of passing. Looking back at the most recent off year election it appears that the Pro C side may quickly face diminishing returns, especially as extra issue groups, such as the Backbone Issues Committee and the Independence institute, pile on.

As things currently stand now, the main anti C issue group Vote No; It's Your Dough has spent $129,550 as of 10/03. This doesn't take into account hundreds of thousands of dollars in adds by ally groups.

As of 9/19 (they are late on filing their 9/19 report) the main pro C issue group Vote Yes on C & D has spent $3,141,000. That is a margin of 24-1.

However, estimates are that the Independence Institute has purchased $400,000 of radio ads that basically blast Referendum C, and that the allied issue groups have probably spent another $250,000. The best estimate we have at this time is that the total amount the allied groups has spent is probably close to $750,000. That brings the margin down to 4-1, not enough to pass an initiative in an off year.

Looking back to 2003, the most expensive campaign was over Amendment 33 (the racetrack gambling amendment). The Pro side spent $6,711,400. The Con side spent $3,278,000 (almost all in the last four weeks). The Pro side outspent the Con side 2-1.

Polling shows that Amendment 33 was stuck basically in a statistical tie during the months of July and August, polling in the low to mid forties. An October poll, put its support at 19% (after suffering a marketing blitz by the opposition). It ended up getting only 19% of the vote that November.

Amendment 32 (Gallagher Freeze) polled in the low thirties in July (so it was basically doomed anyway). Even though the pro side outspent the con side by 87-1, it lost by a margin of 2-1.

Referendum A (Water Bonds) polled in the mid sixties in the wake of the drought. However, the pro side only outspent the opposition 2-1 overall, and was basically matched dollar for dollar during the last four weeks.

We are working on a model that can account for the effect of campaign fundraising differentials. We will have an updated model predicting the outcome of this election up on Mile High Delphi this weekend.

If you are a supporter of Referendum C you need to understand this, right now C probably is a 5-1 or a 6-1 underdog. Absent a huge GOTV and a media blitz in the last few weeks of the campaign (read as outspending the allies by a factor of 10-1) C will lose.

Cleaning up this page.

This afternoon we are cleaning up this page.

We are fixing some broken links...Such as The Plumber

Adding some new links...Follow the Money..Denver Gas Prices...The Fug Girls

And, after much confusion, we will be moving almost all futures market book info over to polstate, with only the most basic info available here.

Tuesday, October 04, 2005

Closing in on Referendum C! Building a better model.

The first part of building any scientific model (aside from coming up with a hypothesis) is collecting data. My last few days have been filled with the unfortunate experience of collecting new data for the model on Referendum C.

Some of our data raises interesting questions, such as, can we use even year election results to predict an odd year election, or do we need to focus on odd year election results only. Quite frankly we aren't sure yet. If we just go off of the odd year results, the future looks bleak for Ref. C. For example, in 2003 Amendment 33 (the racetrack gambling amendment) polled in a statistical tie in the months of July and August, then suddenly in a late October Ciruli poll it collapsed down to 19%. None of the 2003 ballot issues polled over 35% by October, surprising since Referendum A (the water bonds referendum) polled in the mid 60s in July. Additionally, we saw undecided voters break on average of 4-1 against the ballot measures. That means that for Referendum C to pass it will need to poll about 48% or 49%. Anything lower than that will be the kiss of death.

Can we use 2004 data? Maybe, but until we run the numbers we don't know. Even year elections have higher turnouts than odd year elections, this turnout differential makes a huge difference (h/t to the guys over at soapblox, their election data is great and they turned me on to the drop off that happens between the top of the ticket and the referenda items). Until I can quantify that differential and use it properly, I probably can't use even year data, its just too dissimilar. On example of this is the difference between FasTracks, which passed last year with basically 60% support and Guide the Ride, which was basically the same thing, but it lost with only 40% of the vote in 1997.

Campaign funding also is a big issue. Campaign finances are a statistically significant variable in legislative races (not the most significant however, voter registration numbers rule that roost). Finding the campaign finance numbers for prior year elections here in Colorado, quite frankly, sucks. The Secretary of State's website doesn't allow you to search by campaign issue, so I have to jump through hoops to find any information. Even once I find the name of the issue committee, I have to add up the campaign expenditures because they don't have a running total. I assume that we will end up finding out that the pro side must outspend the con side by a factor of greater than 10-1, that seems to be the conventional wisdom. However, maybe we'll prove the conventional wisdom wrong, I just don't know yet.

Until we work out what data to use, and find the funding numbers, it wouldn't be prudent to update our model. Just from my gut, it looks like C has an uphill battle. I can't find an example of a off year ballot measure winning without polling in the 50s.

New polling information out, Ref. C and the Top Of the Ticket.

Earlier today Floyd Ciruli appeared on 850 KOA's Colorado Morning News. In that interview he stated that polling on Referendum C was showing that support for the issue is basically stalled, and that it is statistically tied with those who will vote against the issue. He added the caveat that most Coloradoans don't seem to be paying attention to the election battle yet.

This seems to be in line with, and re-enforces the polling information that we have gotten over here. The most recent numbers that made it into the sunlight are available at our Referendum C and D Homepage. They show support for C at 48% with 44% against it. 8% are undecided and the poll had a margin of error of 4%. So Ciruli's caveat seems right on target, the issue seems statistically tied.

As for the Governors race...

ColoradoPols put these Wall Street Journal-Zogby numbers up earlier today. You can link directly to them here. They credit Coyote Gulch who in turn credits Taegan Goddard's Political Wire who in turn sends you to the Wall Street Journal Online. Quite a journey but well worth it.

Zogby looked at a variety of different matchups that are all possible next November:

Bob Beauprez vs. Bill Ritter
Beauprez: 43.6%
Ritter: 39%

John Hickenlooper vs. Marc Holtzman
Hickenlooper: 48.2%
Holtzman: 36.1%

Bob Beauprez vs. Joan Fitz-Gerald
Beauprez: 45.4%
Fitz-Gerald: 38.6%

But what is perhaps the most interesting tidbit from this poll is what Zogby gleaned from it:

"With term-limited Gov. Bill Owens set to step down, this race promises to be a shootout...Should the Democrats nominate liberal Senate President Joan Fitz-Gerald, Rep. Beauprez is the GOP's best hope for defeating her. The more conservative Bill Ritter, Denver's district attorney (sic), would be a more formidable opponent and is in a close race with all three Republicans. If Denver Mayor John Hickenlooper -- who has said he has no intentions of running for governor -- enters the race, he is poised to defeat any GOP challenger handily."

Cross posted at Polstate.

Monday, October 03, 2005

Some thoughts on Katrina...

As we work on putting out an improved model for predicting the outcome of Referendum C we haven't lost sight of one of the most responsibilities that the political prognosticati have, helping our readers procrastinate with nifty tidbits.

Here are some thoughts on Katrina...

On the local leadership...
So I'm asking Congress, please investigate this now. Take whatever idiot they have at the top of whatever agency and give me a better idiot. Give me a caring idiot. Give me a sensitive idiot. Just don't give me the same idiot.
-Aaron Broussard, President of Jefferson Parish, La. (hat tip to Vincent Carroll)

On what Charity to pick...

Charity Navigator lists relief agencies that get three or four star ratings-the highest the group gives out.

The Better Business Bureau Wise Giving Alliance offers various info about the charities.

The IRS provides tips on giving.

And on what is the most American game of all; the blame game...

The urge to blame is an innate human impulse dating back a million years or more. It's an impulse that travels through our bodies to our fingertips, as we all saw in the frenzied finger-pointing over Hurricane Katrina.

Just as meteorologists and coastal engineers predicted the hurricane and flooding, there are "blame researchers" who foresaw the storm of words and pointed fingers that followed. They weren't surprised when politicians, victims and the media angrily affixed blame for the inadequate levee system and deadly slow rescue efforts. In fact, they see the tragedy through the prism of an academic question: can our blame impulses weed out ineptitude, improve conditions and save lives?

Modern America is beset by blame-mongering. At ShiftTheBlame.com you can buy a "calibrated blame-shifting device" for $2.95. It's a giant foam had with the words "It's Your Fault!" on the pointer finger. Run by East Bank Communications, an ad agency in Portland, Ore. the Web site offers tongue-in-cheek mantras: "You have everyone but yourself to blame." "It's not you, it's the printer." The jokes ring true because finding blame is an American Preoccupation...

Hat tip to the Wall Street Journal.

Tuesday, September 27, 2005

Referendum C in the News; Updated Legislative Odds.

The Rocky mountain News is running a series of articles that examine how Referendum C would effect a handful of local individuals.

The opening article of the series is here.

The first of the seven parts is available here. This first part featured a struggling west slope family who said that they were going to vote against C because they couldn't afford to forfeit their tax rebates over the next five years.

The second article features a 23 year old single professional who lives in Aurora and works in Downtown Denver. She says that she supports C because "[I]n the end... Her TABOR refund would be better spent on improving roads and higher education." You can link to that article here.

The third part of that series features a look at the unusual Mayor of Sterling and his wife. They support C because they want to increase funding for Sterling's High School and because "[L]osing TABOR refunds with Refs C and D wouldn't devastate them." You can link to that story here.

Overall the score is No on C:1 Yes on C:2. I imagine that the Rocky will try to get a balanced look. We will probably see the storyline end with three families supporting C, three against it and one who still hasn't made its mind up.

Meanwhile in the Campaign over C...

A proxy war has sprung up between Republican Gubernatorial Candidates Mark Holtzman and Bob Beauprez. Each accuses the other of not being enough against C and of abusing campaign finance laws. The latest in a series of articles on the subject is available here. This is simply positioning for the primary next fall, as each candidate vies for support from the Republican Party's anti-tax base.


It seems that the new trend in the media's reporting on the battle over Referendum C is to point out how this side or that has lied. Most recently the No on C side has taken a few big hits. Earlier in the week Jon Caldara, the President of the Independence Institute admitted that the State didn't pay $5,000 for a piece of artwork that featured some dildos hanging on hooks. That charge had been a rallying cry for the anti-C forces. In this article it comes to light that the State didn't give Colorado company Red Robin nearly $100,000 in cash. Instead the State approved Red Robin for a grant that it will receive in the future.

Updating our legislative numbers...

It appears that the GOP is going to have a much harder time winning back the State Legislature next year than we originally thought. While these predictions are only preliminary ones, here is what we think the probability of control for next year is right now:

probability in % terms....................Fractional
GOP Control.............5%......................................19/1
Dem Control.............45%.....................................6/5
Split Control.............50%.....................................1/1

Monday, September 26, 2005

Delphi on the Battle over Ref. C. End of Sept. Edition.

Eric Sondermann, a local political analyst, wrote a great piece in Sunday's Denver Post. The thesis of his piece was that both sides were using overheated rhetoric in the debate over C. Ref. C won't kill TABOR, it will basically maintain the status quo. On the other side Ref. C won't even begin to fill the wish list for the "fiscal crisis" crowd. If it passes they will have a new fiscal crisis that needs to be solved in the fall of 2006.

An exerpt is available here:

These ballot proposals could be approved and yet neither the professed wishes of the proponents nor the desperate fears of the opponents are likely to be realized. Contrary to the proponents' implicit promise, the state's list of unmet needs would remain a lengthy one. Within months, there would be fresh talk of a "fiscal crisis" and impassioned calls for new revenue sources. Contrary to the opponents' claims about the magnitude of the "tax increase," these are essentially status quo measures that would maintain tax and spending at pre-recession levels.

A link to the piece is available here.

As far as the advertisements for and against C go, both sides appear to be lying.

The No on C side has been running ads saying that the average voter will forfeit $3,200 in tax refunds. That number seems to be in doubt:

Based on current estimates, the state would keep and spend an extra $3.7 billion over five years if Referendum C passes. The amount of money that most taxpayers would give up if Referendum C passes is $491, according to the nonpartisan Legislative Council.

Opponents of Referendum C, however, calculate that the average family of four would give up $3,200 in tax refunds.

To get that number, they divide $3.1 billion - another projection of the amount of money the state would keep if C passes - by the number of families.

But that approach assumes each taxpayer qualifies for every possible refund. The legislature has devised 12 special-interest tax breaks for things such as child care and business and capital-gains exemptions. No taxpayer qualifies for every refund.

In truth, the real number is probably somewhere in between. However, the Pro-C adds that accuse the anti-C side of lying about the $3,200 isn't really truthful itself.

In another commercial, the Pro-C side outright lies. Here is what the Denver Post had to say:

The message: Three years of recession have caused deep cuts to education that have hit kids hard. Substandard buildings, crowded classrooms and cuts to programs including preschool, kindergarten and Read to Achieve are the result.

Fact: Read to Achieve, which provides grants to help 2nd- and 3rd-graders who read below their grade level, was cut $11.9 million between fiscal year 2004-05 and 2005-06. Between 2003-04 and 2004-05, 2,000 kids were cut from the state's preschool program. But for 2005-06, money was restored to fund the 2,000 kids and add 1,310 more. In 2001-02, the state cut funding for unsatisfactory schools to send their half-day kindergartners to full-day programs. That funding was restored for 2005-06...Overall, K-12 funding increased between 2001 and 2004 because of Amendment 23, which requires the state to increase funding by student growth and inflation plus 1 percent every year.

Between 2001-02 and 2003-04, the total general fund money to the Education Department increased $149 million, from $2.3 billion to $2.4 billion.

When the Pro-C side says that "Three years of recession have caused deep cuts to education that have hit kids hard. Substandard buildings, crowded classrooms and cuts to programs" they aren't telling the truth.

You can link to the story here.

Tomorrow I'll have more links up. Stay tuned.

Friday, September 23, 2005

Delphi on the Battle over Ref. C.

The Conventional Wisdom is that the debate over Referendum C is creating a rift in the Colorado State Republican Party. An excellent Durango Herald article outlined the basic facts of the rift.

A November ballot proposal to fix Colorado's financial woes has sparked a nasty and expensive campaign that has split the state's Republican base, already reeling from losses in last fall's elections.

Gov. Bill Owens, once touted as presidential material in conservative Republican circles, teamed with Democrats last spring in drawing up the proposal that will ask voters to give up as much as $3.7 billion in tax refunds over the next five years.

TABOR [The part of the state constitution that limits tax increases and government spending and which would be modified by Ref. C] supporters recently brought in former House Majority Leader Dick Armey, who accused Owens in a face-to-face debate of "drinking backslider wine" for supporting the ballot measure. The measure's author, Douglas Bruce, has threatened to sue if voters approve the ballot plan.

Owens acknowledged the plan does not have much support among Republicans, but he said Colorado is facing a crisis next year if voters refuse to give up surplus tax refunds. He said he is simply backing TABOR's provision that requires voter approval involving tax dollars.

And it is true, the State party is split, but the split is more symbolic than real. A handful of elected GOP officials have come out in support of C, however, straw polls, polling and local party officials have revealed that the Republican party is anything but split, rank and file members don't support C. For instance the Larimer County Republican Executive Committee voted against supporting C, angering some former GOP office holders. An earlier GOP straw poll saw support for C at about 25%. The question that I have to ask is why are these Republicans taking a position that seems suicidal inside their party?

The answer appears to be the recent Democratic surge in Colorado. In 2004 the Democrats picked up both houses of the State Legislature, a Senate Seat and a Congress Seat. Many GOP leaders believe that the reason for that surge was voter disgust at the "wither on the vine" attitude many Republicans have towards State spending. I believe that many of these GOP leaders have made a strategic choice. Support C and take away a Democratic issue. Governor Owens is a perfect example of this approach. If C loses he blunts the criticism of Democrats, but if C wins he can take much of the credit as the highest profile Republican in support of C. Perhaps allowing this one time presidential hopeful to restart his career.

The Democrats see C as a win-win. If C wins they can finally do some damage to the hated TABOR amendment. Democratic constituents have been disproportionately affected by TABOR's limits on spending growth. Quite literally TABOR takes money out of their pockets. They also can blunt some of the effects of Amendment 23, which put the K-12 education budget on autopilot, by undoing parts of TABOR, not modifying 23. If C loses then the Democrats can come back in 2006, and with all the state-wide races, including an open governors seat, they can play the same cards that they won with in 2004, namely that the GOP is ideologically rigid and can't respond to the real world needs of everyday Coloradoans. However this isn't my idea, it comes from the American Prospect:

It [the Democratic Da Vinci Code] also echoed Colorado Democrats, who used deficits to win the state Legislature for the first time in 40 years. "The Republicans' obsession with narrow cultural issues while the state's looming fiscal crisis was ignored drove a deep wedge between fiscally conservative live-and-let-live Republicans and the neo-conservative extremists with an agenda," wrote one Denver Post columnist.

More often than not the conventional wisdom is wrong. The state GOP is not deeply split over C. When new polls become available look for anywhere from one quarter to one third of the GOP voters to support C, but no where near the huge divide that you hear in the media. Where their is a divide is between the base of the party and the elected officials who believe that the Democrats rode the budget deficit to victory.

The Democrats are going to believe that they have won no matter what. If C passes any reform to amendment 23 will be off the table. If it loses they can play the same cards that won in 2004. And, unless the GOP learns to play the same game as the Democrats they will win.

The Colorado GOP may not be as divided as the media makes it out to be, but it is infected with a cancer. It held power for too long, as Vince Lombardy used to say, sometimes you need to get beat down in order to win. The Colorado Democrats have found out how to win. They blocked any ballot issues to deal with the budget problems while in the minority in 2003 and then accused the GOP of doing nothing, and they won. For the GOP to win in the future they may have to do something that Republicans in the late 1990s never could have imagined, learn from the Democrats.

Thursday, September 22, 2005

Referendum C and D homepage.


Mason-Dixon Oct 11-13
Referendum C Yes: 47% No: 44% Undecided: 9% MOE:4%
Referendum D Yes: 44% No: 45% Undecided: 11% MOE:4%

POS Oct 10-12
Referendum C: Yes:49% No:46% Undecided:5% MOE:4.38%
Referendum D: Yes:50% No:44% Undecided:6% MOE:4.38%

Dan Willis* 08/05
Referenda C: For 48%. Against 44%. Undec. 8%. MoE 4%.
Mason Dixon 07/05
Referenda C: For 43%. Against 42%.
Referenda D: For 39%. Against 45%.

Update 10/17/2005 Rosen Reveals Part of GOP Strategy.

Update 10/17/2005 Closer Look at Friday's RMN Poll Results.

Update 10/16/2005 New Denver Post Poll Out.

Update 10/14/2005 New Polling Out, C and D within MOE.

Update 10/12/2005 Caldara and Allies Outspending Vote Yes on C and D.

Update 10/10/2005 Mile High Delphi Reveals New Model. Initial Prediction, C and D lose 47-53.

Update 10/05/2005 C and D Need More Money to Win.

Update 10/04/2005 A look at what we look at as we build the Referendum C and D model.

Update 10/04/2005 Ciruli Says C and D Statistically Tied.

Update 9/26/2005 Delphi on the Battle Over Ref. C and D.

Update 9/23/2005 A look at the coming battle.

Update 9/22/2005

The battle over C and D is shifting into high gear. Republican Gubernatorial candidate Mark Holtzman has hitched his horse to the No on C campaign. If C goes down his stock will go up. He is now being featured on Anti-C commercials that are financed by, among others, his father.

A new front has opened in the battle, earlier in the month Jon Caldara and the No on C campaign sued over the title of the issue. Rocky Mountain News political columnist Peter Blake, a fairminded journalist, sided with Caldara...

Earlier this week Jon Caldara and other foes of Referendum C sued the secretary of state and the legislature because, they said, the ballot title erroneously says it will enable the state to spend extra billions on roads, schools and health care over five years "without raising taxes." In fact, the plaintiffs argue, "if Referendum C passed, the measure would raise net taxes."

Under the rules laid down by the Taxpayer's Bill of Rights, says Caldara, the ballot title should ask: "Shall state taxes be increased $3.7 billion over five years" by raising the TABOR baseline. He asked the Jefferson County District Court to force this change in the ballot title. (For what it's worth, Caldara didn't protest Referendum D, since its operative phrase on the ballot title is "Without increasing any tax rates or imposing any new taxes . . ." In other words it notifies voters of the distinction between hiking taxes and hiking tax rates.)

The point here isn't to debate whether or not spending tax money that would otherwise be refunded to the citizenry is in fact a tax hike. The point is that the courts will refuse to address a possibly faulty ballot title before the election.

In other news, Dan Willis reported on Coloradopols that Referendum C had pulled up to 48-44. We have no reason to doubt these numbers, it still is a statistical tie.

* Not really sure how to characterize this poll. Here is what Dan said about it:
The poll numbers I quoted were given in a public forum I attended a couple of weeks ago. They were supposedly from one of the well-known pollsters (Talemy-Drake is in my mind, but I honestly don't remember for sure).

8/30/05 Update.

Original Post 8/21/2005

The big contest this fall is Referendum C (D only matters if C passes).

This page is designed to answer any questions someone may have about C and D and it will provide you with quick links to the supporters and opponents of C and D.

Mile High Delphi has no opinion on C and D.

Lets answer some basic questions. What will C and D do? Who is supporting C and D? Why? Who opposes C and D? Why?

What Will C and D Do?

"What is Referendum C?

C Suspends state spending limits for five years.

C Allows lawmakers to spend about $3.6 billion they otherwise would have to refund to taxpayers.

What is Referendum D?

D Allows the state to borrow against that money to begin transportation and school construction right away."

Source Rocky Mountain News.

Who is supporting C and D and why?

Vote Yes on C and D. The two Referenda are currently (updated on 8/21/05) endorsed by 522 organizations. The Vote Yes on C & D campaign committee had nearly 1.3 million dollars on hand as of July 26th.

Some of the major donors to the Yes on C & D campaign are the AFSCME Council 76 which gave $100,000 on 07/22/05, the Denver Metro Chamber of Commerce which gave $250,000 on 07/21/05, Centura Health gave $195,000 on 07/15/2005 and the Colorado Education Association gave $200,000 on 07/26/05.

Sometimes you feel like you just get a bunch of spin as to why to vote yes or no on something. Here at Mile High Delphi we've decided to just give you a link to the supporter's website (that way you don't have to read our spin). Here is why Vote Yes on C & D thinks you should vote yes.

Who opposes C and D and why?

Vote No; It's Your Dough. The two Referenda are currently (updated on 8/21/05) opposed by Americans for Prosperity, Americans for Tax Reform, The Club for Growth, Colorado Union of Taxpayers, Freedom Works, National Taxpayers Union, Save Tabor.com and the Colorado Republican Business Coalition. The Vote No; It's Your Dough campaign committee had $107,279.49 on hand as of July 26th.

Vote No; It's Your Dough had only three donors over the reporting period. Helen Sweeney gave $25.00. Edward McVaney gave $100,000. Brent Bickel gave $5,000. All on 07/19/05. This only tells part of the story. Two of the major opponents of C and D, Jon Caldara and John Andrews, both have radio shows which they use to attack the referenda items. Also, the Independence Institute is running radio commercials that basically attack C and D as tax increases (but which don't advocate a no vote so therefore count as "educational" and protect the Institute from losing its tax exempt status).

The best "No on C and D" website I've been able to find so far is this one, TaxIncrease.org.


There is only one poll out on C and D so far:

Mason Dixon 07/05
Referenda C: For 43%. Against 42%.
Referenda D: For 39%. Against 45%.

Top of the Ticket!

Poll average Ritter (D) 51.7% Beauprez (R) 35% Spread Ritter +16.7

Update (10/05/2006)

If you want a page that has everything you could want on this race go here. Real Clear Politics has done as good a job as anyone out there looking at this race.

As for this race. The Beauprez campaign cannot seem to get any traction against Ritter. Beauprez cannot even get the GOP voters lined up behind him, nevermind the unaffiliated voters. Don't mind those biased attacks against the Beauprez campaign that come out from the likes of coloradopols which is a liberal attack blog that masquerades as "Politics, News and Inside Information." (Not that we don't like liberal websites, squarestate.net is a daily read, and they state that they are progressives up front.)

The main problem Beauprez has is that this year the public is generally against the Republicans. National scandals and the 6 year political itch give Beauprez a strong headwind that he seems to be making no progress against. Unless his poll numbers can pull to within five points he will lose to Ritter, if it is a blowout the GOP could lose more seats in the State Legislature (which the Democrats already control) and the Democrats could pick up the 7th Congressional District.

Update (8/15/2006)

I've just received a bunch of data from Matrow that has caused us to move this race from Barely GOP to Lean Dem.

The reasons for this are many and varied and others say it better than we can, but this graph of the price for a futures option from Tradesports.com shows it all.

The market has basically shown that the probability of the Bob Beauprez winning the Governor's race has dropped from about 70% in February to about 20% today. Now this market isn't very liquid, but it does have strong predictive powers.

Below you can see how the price on a Ritter to win option has gone up over the same time.

Larry Sabato has decided to put this race into the Leans Democratic category.

A link to Sabato's Crystal Ball and its page devoted to the Colorado Gubernatorial race is available here.

The Cook Political Report has this race listed as a Toss Up. A link to their gubernatorial Race Ratings page is available here.

Here are some recent polls:

Rasmussen Ritter (D) 48% Beauprez (R) 39% Undec. 12% August 14

Mason-Dixon Ritter (D) 42% Beauprez (R) 35% Undec. 23% July 17

Rasmussen Ritter (D) 42% Beauprez (R) 37% Undec. 16% July 10

The guys over at Coloradopols have a page with more polls on it. Click here.

8/19-24/2005 Ritter (D) 44% Beauprez 42% Undec. 14%

Update (4/27/06)

We are going to move this race into the Barely GOP category.

The race breaks down like this. It is going to be a Bill Ritter (D) vs. Bob Beauprez (R) matchup.

Here is a graph of how much a option for the GOP winning the Colorado Governor's race is going for on Tradesports.com:

As you can see the market over at Tradesports has moved the probability of the GOP winning Colorado's open Governor's seat from 70% down to 51% today.

Larry Sabato's Crystal Ball has the race listed as a Toss Up, we think he is hedging, but check his site out.

Update (9/22/05)

Here is a link to a post over at Polstate. It appears that a early poll is out that shows Ritter leading Beauprez. All polling information will be placed at the top of this page.

Original post (8/30/05)

As of right now (over a year out from the election) there seems to only be three big players in the Governor's race.

On the Republican side there is Bob Beauprez and Mark Holtzman.

Bob Beauprez's website is available here. So far he appears like the slight favorite in the race. He has won a straw poll and has better name recognition than Holtzman.

Mark Holtzman's website is available here. While Beauprez seems like a slight favorite right now, Holtzman could easily win. We're waiting for some polls to point the way. Holtzman has the money lead right now, but we are a year out.

Edge...Slight Edge to Beauprez.

On the Democratic side there is only one man standing, former Denver D.A. Bill Ritter.

Ritter is a pro-life Democrat. It has been opined that he should be a strong candidate. If he doesn't have a primary we will get to see. Senator Salazar and Denver Mayor Hickenlooper both have declined to run.


Our predictions for the general election. Colorado had trended Republican during the late 90s and into the early part of this century, however a Democratic tidal-wave hit Colorado in 2004. Our preliminary prediction is that the Republican candidate has a slight advantage.

Wednesday, September 21, 2005

Referendum C Odds Update 09/21/2005

The market book will be moving over to the Political State Report. Check there daily for price quotes and end of week graphs.

We will still have a condensed book on the sidebar.

Referendum C to win continues to surge up in bidding. Over the past two days it has moved up from 48 to 55. Word over at ColoradoPols is that Ref. C has moved to within striking distance. Hat Tip to Dan Willis for keeping his ear to the grapevine. No action on the No on C contract.

The market has begun. Remember it is free to play. Click on the "sign up" link to the right and join in. There is lots of room for new participants. Sign up. Read the FAQs. Win $200.

Here is where the market stands now:
Name of Contract.0-100...Moneyline...Fractional.
Ref. C to lose.....52......-108...........10/11.
Ref. C to win......55......-122.............5/6.

Tuesday, September 20, 2005

Referendum C Futures Market Book 9/20/05

Today's trading starts at 6:00 PM for the Referendum C Futures Market. Remember no trading over the weekend. Below you will see a table that shows the best trading information. The opening prices were established by auction. B Qty = How many contracts are offered for sale at the best price. Bid = the price those contracts are being offered for. Offer = The best price that is being offered for a contract. A Qty = the number of contracts that are wanted at that best price.

Best to SellBest to Buy
Contract NameB QtyBIDOfferA QtyLastVolChange
Ref. C to Win10053501005075+2
Ref. C to Lose505253255200

Below is the C to win book. Green indicates buyers. Red indicates sellers.

Referendum C to Win Book

Referendum C to Lose Book

Referendum C Odds Update 09/20/2005

Referendum C to win has moved up to even money...

The market has begun. Remember it is free to play. Click on the "sign up" link to the right and join in. There is lots of room for new participants. Sign up. Read the FAQs. Win $200.

Here is where the market stands now:
Name of Contract.0-100...Moneyline...Fractional.
Ref. C to lose.....52......-108...........10/11.
Ref. C to win......50.......100...............1/1.

Monday, September 19, 2005

Ref. C Futures Market 09/19/2005

All prices are unchanged over the weekend.

Friday, September 16, 2005

Week in review.

This week the market acted much as we thought it would. The opening prices didn't change at all. We signed up a dozen new traders. It appears that everyone is waiting for some new information before they begin to move.

Referendum C Futures Market Book 9/16/05 (week in review tonight at 7:00)

Today's trading starts at 6:00 PM for the Referendum C Futures Market. Remember no trading over the weekend. Below you will see a table that shows the best trading information. The opening prices were established by auction. B Qty = How many contracts are offered for sale at the best price. Bid = the price those contracts are being offered for. Offer = The best price that is being offered for a contract. A Qty = the number of contracts that are wanted at that best price.

Best to SellBest to Buy
Contract NameB QtyBIDOfferA QtyLastVolChange
Ref. C to Win254847254800
Ref. C to Lose505253255200

Below is the C to win book. Green indicates buyers. Red indicates sellers.

Referendum C to Win Book

Referendum C to Lose Book

Thursday, September 15, 2005

Referendum C Futures Market Book 9/15/05

Today's trading starts at 6:00 PM for the Referendum C Futures Market. Below you will see a table that shows the best trading information. The opening prices were established by auction. B Qty = How many contracts are offered for sale at the best price. Bid = the price those contracts are being offered for. Offer = The best price that is being offered for a contract. A Qty = the number of contracts that are wanted at that best price.

Best to SellBest to Buy
Contract NameB QtyBIDOfferA QtyLastVolChange
Ref. C to Win254847254800
Ref. C to Lose505253255200

Below is the C to win book. Green indicates buyers. Red indicates sellers.

Referendum C to Win Book

Referendum C to Lose Book