Tag: PBI

PBI or Party Brand Index is a concept I developed (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. I’ve tackled IN, NC, CO, VA, MO, OK, AR, WV, NH, OH, and Florida. Now I will look at the fast becoming a purple state of Arizona

Like always I would like to post the data, then I will offer some analysis. My basic pattern is to work my way “out” from the “Purple States” to the more Blue and Red ones. (Although once in a while I like to skip my normal pattern of working out from purple states.  I’m often curious on how my model would work in states like that are deeply blue at the local level, but deeply red at the presidential level.) As a reminder a negative number indicates a Republican bias to a district, while a positive number indicates a Democratic bias.

Let’s examine the fast becoming a purple state of Arizona.

ANALYSIS

Arizona’s number are masking some trends. John McCain was basically unchallenged in 2004, his large victory margin has skewed the numbers in most districts PBI by about 2-3 points. Also there are a number of other trends working in team blue’s favor.

47% of Arizona’s voters had a college education. Obama carried all 14 states (and DC) that had higher college educated electorates. In fact of the 23 states plus DC that were over 44% Obama carried everyone except Arizona, North Dakota, and Texas.

Latinos grew from 12% to 16% of the electorate, but Obama had one of the smallest increases in their support levels of any state. They supported Kerry over Bush in 2004 at 56% to 43%, Obama was at 57% to 41%. Consider this a home state advantage.

The biggest obstacle to a Democratic victory in the future is the party registration numbers. Republicans outnumber Democrats by about 7% state wide, Obama only carried one state where Republicans outnumber Democrats (Indiana). Since GOP voters tend to be more partisan than Democrats (they almost always vote at higher rates for the GOP candidate than Democrats do for the Dem) this lead to a major obstacle. In many ways this is the mirror image of the GOP’s problems in New Jersey. Every 4 years looking at NJ demographics they decide to target it, but the huge Democratic registration edge makes it their “Great White Whale.”

Ideologically Obama won 6 states that were less liberal than Arizona (based on the percentage of self describe liberals) including Colorado, Florida, North Carolina, and Ohio. Colorado had a 17% liberal, 46% moderate, 36% conservative breakdown, Arizona’s numbers are 21% liberal, 42% moderate, 36% conservative. Arizona’s numbers fall roughly in the middle of the average swing state. * the demographic information cited above comes from Chuck Todds’ book “How Obama Won”

My best advice to the party would be to focus on party building in Arizona. Get more people registered at Democrats. Focus party building on the young people moving into Arizona’s college towns. This is in addition to the obvious tactic of getting Latinos to register and vote at higher rates.

One final note, as I alluded to earlier, after I come across a few “conservative” Democrats, I run a “correction” factor to account for them being Blue Dogs. The general idea is that the distance they are able to maintain from the national party may help them win over voters who are more reluctant to vote for Democrats. I want to examine another swing state before I “recompute” Ohio’s and Florida’s (my last two states) Blue Dogs.

As a recap, here are the first “batch” of Blue Dogs, and rural Democrats (West Virginia’s Democrats aren’t members of the Blue Dogs) that I examined correcting for partisanship and ideology.

FOUR BLUE DOGS

THREE BLUE DOGS

As a reminder ranking a members ideology is a somewhat subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have. As I explained a few diaries ago I will use ADJUSTMENT FACTOR 2 in all subsequent corrections.

Because there are “only” 50 states (as opposed to evaluating 435 house members), I will at a later date have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have.

Anthology:

PBI (Party Brand Index) Part 8 Florida

PBI (Party Brand Index) Part 7 Ohio

PBI (Party Brand Index) Part 6 WV and NH

PBI (Party Brand Index) Part 5 Nevada and Iowa

PBI (part 4) MO, AR, OK

Party Brand Index (part 3) North Carolina

Party Brand Index (part 2) Colorado and Virginia (updated)

Introducing PBI, Party Brand Index (Updated)

PBI or Party Brand Index is a concept I developed (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. I’ve tackled IN, NC, CO, VA, MO, OK, AR, WV, NH,and OH. Now I will look at the swing state of Florida.

I had to take a break from my analysis for personal reasons. Like always I would like to post the data, then I will offer some analysis. My basic pattern is to work my way “out” from the “Purple States” to the more Blue and Red ones. (Although once in a while I like to skip my normal pattern of working out from purple states.  I’m often curious on how my model would work in states like that are deeply blue at the local level, but deeply red at the presidential level.) As a reminder a negative number indicates a Republican bias to a district, while a positive number indicates a Democratic bias.

Let’s examine the swing state of Florida, the second large state I have examined (Ohio was the first).

FLORIDA PART 1

FLORIDA PART 2

FLORIDA PART 3

ANALYSIS

Unlike most other states I have examined Florida has very little deviation, from the PBI and PVI. This may be the result of the Floridian GOP’s very effective gerrymandering of house districts after the 2000 census. In fact there are only two districts where the PBI and PVI deviate. One is in Democratic Rep. Kosmos (FL-24). Interestingly enough this district was designed to elect a Republican although it’s not represented by a Democratic. The other is in Democratic Rep. Alan Boyd (FL-02), I think the PBI is a little high. This will be examined in the next batch of Blue Dogs I examine.  

The only hope for “new ground” for Democrats may lie in a swing in the voting preference of Cuban-Americans as a younger less “single issue” group of voters come of age. Also the makeup of many of these Latino districts are changing as more Puerto Ricans and Dominicans move in.

One final note, as I alluded to earlier, after I come across a few “conservative” Democrats, I run a “correction” factor to account for them being Blue Dogs. The general idea is that the distance they are able to maintain from the national party may help them win over voters who are more reluctant to vote for Democrats. I want to examine another swing state before I “recompute” Ohio’s Blue Dogs.

As a recap, here are the first “batch” of Blue Dogs, and rural Democrats (West Virginia’s Democrats aren’t members of the Blue Dogs) that I examined correcting for partisanship and ideology.

FOUR BLUE DOGS

THREE BLUE DOGS

As a reminder ranking a members ideology is a somewhat subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have. As I explained a few diaries ago I will use ADJUSTMENT FACTOR 2 in all subsequent corrections.

Because there are “only” 50 states (as opposed to evaluating 435 house members), I will at a later date have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have.

Anthology:

PBI (Party Brand Index) Part 7 Ohio

PBI (Party Brand Index) Part 6 WV and NH

PBI (Party Brand Index) Part 5 Nevada and Iowa

PBI (part 4) MO, AR, OK

Party Brand Index (part 3) North Carolina

Party Brand Index (part 2) Colorado and Virginia (updated)

Introducing PBI, Party Brand Index (Updated)

PBI or Party Brand Index is a concept I developed (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. I’ve tackled IN, NC, CO, VA, MO, OK, AR, WV, and NH. Now I will look at the swing state of Ohio.

I had to take a break from my analysis for personal reasons, but I will try to return to my pattern of one diary a week. Like always I would like to post the data, then I will offer some analysis. My basic pattern is to work my way “out” from the “Purple States” to the more Blue and Red ones. (Although once in a while I like to skip my normal pattern of working out from purple states.  I’m often curious on how my model would work in states like that are deeply blue at the local level, but deeply red at the presidential level.) Let’s examine the swing state of Ohio, the first large state I have examined.

OHIO Part 1

OHIO Part 2

Based on the difference between PVI and PBI I will conjecture the following. Rep. Tiberi R-OH who according to PVI is in a D+1 District will continue to survive as PBI shows him to actually to have a 4% GOP edge in a house race. Representatives Driehaus (+1D PVI, – 2 PBI), Kilroy (+1D PVI, – 2 PBI), and  Boccieri (+4 R PVI, – 8 PBI) will be in for slightly tougher than expected races, while Rep. Space will win by a slightly larger than expected margin (+7 R PVI, – 5 PBI) My source for the election result data for Ohio is the Ohio Secretary of State.

One final note after I come across a few “conservative” Democrats, I run a “correction” factor to account for them being Blue Dogs. The general idea is that the distance they are able to maintain from the national party may help them win over voters who are more reluctant to vote for Democrats. I want to examine another swing state before I “recompute” Ohio’s Blue Dogs.

As a recap, here are the first “batch” of Blue Dogs, and rural Democrats (West Virginia’s Democrats aren’t members of the Blue Dogs) that I examined correcting for partisanship and ideology.

FOUR BLUE DOGS

THREE BLUE DOGS

As a reminder ranking a members ideology is a somewhat subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have. As I explained a few diaries ago I will use ADJUSTMENT FACTOR 2 in all subsequent corrections.

Because there are “only” 50 states (as opposed to evaluating 435 house members), I will at a later date have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have.

Anthology

PBI (Party Brand Index) Part 6 WV and NH

PBI (Party Brand Index) Part 5 Nevada and Iowa

PBI (part 4) MO, AR, OK

Party Brand Index (part 3) North Carolina

Party Brand Index (part 2) Colorado and Virginia (updated)

Introducing PBI, Party Brand Index (Updated)

PBI or Party Brand Index is a concept I developed (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. I’ve tackled IN, NC, CO, VA, MO, OK, AR, now I will look at the swing states of West Virginia and New Hampshire.

First like always I would like to post the data, then I will offer some analysis. My basic pattern is to work my way “out” from the “Purple States” to the more Blue and Red ones. Once in a while I like to skip my normal pattern of working out from purple states.  I’m often curious on how my model would work in states like that are deeply blue at the local level, but deeply red at the presidential level. I will offer a refresher on them later. But first let’s examine the swings state of New Hampshire and the “split” state of West Virginia.

WEST VIRGINIA

NEW HAMPSHIRE

Although PBI shows some level of pessimism for Carol Shea-Porter in the NH 1st, the trend in her district is truly astounding. From -26% to a +6% Democratic in 2 election cycles. Also during that time NH has seen a 5% rise in the percentage of the electorate that calls itself liberal (conservatives fell 2%) and the party split went from 32:25 Republican to 29:27 Democrat. Although this district isn’t completely safe the trend is good for Team Blue. She is still the top GOP target in the New England.

After I come across a few “conservative” Democrats, I run a “correction” factor to account for them being Blue dogs. The general idea is that the distance they are able to maintain from the national party may help them win over voters who are more reluctant to vote for Democrats. Interestingly enough both West Virginian Democratic Reps despite coming from a Red, Socially conservative state are not members of the Blue Dog caucus. I never the less ran there data to see how my model would respond. I actually was surprised where they fell on the partisan and ideological scale. As I explained a few diaries ago I will use ADJUSTMENT FACTOR 2 in all subsequent corrections.

THREE BLUE DOGS

As a recap, here is the first “batch” of Blue Dogs that I examined correcting for partisanship and ideology. Notice the differences in both partisanship and ideology between the West Virginian congressman, and the AK and OK ones.  Nate Silver has been hopeful that Obama could recapture West Virgina for the Democrats. This data shows West Virgina although often lumped in with the other highland/Appalachia region that turned most heavily against Obama is quite a bit more progressive then other states in that area.  If they become comfortable with Obama’s race he could maybe overcome the recent red-tide there.

FOUR BLUE DOGS

As a reminder ranking a members ideology is a somewhat subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have. As I explained a few diaries ago I will use ADJUSTMENT FACTOR 2 in all subsequent corrections.

Because there are “only” 50 states (as opposed to evaluating 435 house members), I will at a later date have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have.

PBI or Party Brand Index is a concept I developed (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. I’ve tackled IN, NC, CO, VA, MO, OK, AR, now I will look at the swing states of Nevada and Iowa.

First I would like to post the data, then I will offer some analysis. My basic pattern is to work my way “out” from the “Purple States” to the more Blue and Red ones. Two weeks ago (I took one week off for Netroots Nation) I skipped my normal pattern of working out from purple states.  I became curious on how my model would work in states like Arkansas that are deeply blue at the local level, but deeply red at the presidential level. I will offer a refresher on them later. But first let’s examine the swings states of Iowa and Nevada.

IOWA

NEVADA

Both of these states have fairly “straightforward” Democratic or Republican leaning districts, only Iowa Rep. Boswell is a Blue Dog. Rep. Titus in Nevada’s 3rd district was designed as a swing district where there would be a equal number of Democrats and Republicans. PBI maintains this is still the case, while PVI suggest a slight Democratic lean. Also Republican Rep. Lantham of Iowa who is frequently mention as a target of Democrats because of his district’s PVI of only +1 Republican, is actually much safer based on PBI -8 (the equivalent of a PVI of 8), which is much closer to his winning margins.

Just a quick note as a reminder on Blue Dogs like Boswell, I developed a “correction factor” that allows for a better “explanation” for congresspersons who win districts that are a “mismatch” for their party’s majority ideology.  I would prefer to run a “batch” of these at a time. So I will simply republish the results for the four Blue Dogs I have encountered since I started this research.

FOUR BLUE DOGS

I developed a formula based on standard deviations. Basically I can figure out how much the average rep deviates from their district.  If I then compare where a reps voting pattern falls (in what percentile) and compare it to their district’s PVI, I can develop a “standard deviation factor”. Inside the standard deviation will get a bonus, outside a negative. The idea is that if a Blue Dog has a very conservative record, they may be surviving not because of a districts Democratic leanings but because they deviate from Democratic policies.  I showed all three variation of my formula but for future examples I will stick to ADJUSTMENT FACTOR #2.

For example, if Rep X is the 42 most conservative rep, that would place her in the 90th percentile. But if her district’s PVI was “only” the in the 60th, their is a good chance her margins would be effected. Using a few random samples I found most reps lie within 12% of their district’s PVI.

Using these dummy numbers I then came up with this.  

   SQRT[(30-12)^2 /2] = about 13%

    Her factor would then be 100 – 13 = 0.87.

So her victory margin would be weighted by 0.87 because she is more than 12% beyond her acceptable percentile range it making the victories in her district approximate 13% less “representative”.

    My theory yields the following formula:

        If rep’s voting record is > PVI then

            100 – SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

        else if rep’s voting record < PVI

             100 + SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

I then repeated this formula to calculate a partisanship correction factor. Ranking a members ideology is a subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have.

Because there are “only” 50 states (as opposed to evaluating 435 house members), I will at a later date have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have.

Continuing on with a concept I developed called PBI or Party Brand Index (with some much appreciated help from pl515) as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging the percentage of the vote from the last two presidential elections in each house district, and comparing it to the nation as a whole, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But PVI in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVIs indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Lastly PVI can miss trends as it takes 4 years to readjust. The purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY should be expected to perform. Last week I tackled NC, this week I’m tackling MO, OK, AR.

This week I skipped my normal pattern of working out from purple states.  I became curious on how my model would work in a state like Arkansas that is deeply blue at the local level, but deeply red at the presidential level. Also I started to develop the first “holes” in my model with my PBI numbers for representatives like Ike Skelton (D-MO). If I want to replace PBI with PVI ignoring “obviously” flawed results doesn’t help. First I will publish my results, then my proposed corrections. As a reminder a Democratic lean to a district get a positive number, a GOP lean gets a negative number.

MISSOURI:

ARKANSAS:



OKLAHOMA:

Ike Skelton in the Missouri 4th with a PBI of – 3 compared to a PVI of – 14 just didn’t seem right. So I jumped ahead of my schedule and did the numbers for Arkansas and Oklahoma, I got a PBI number of 27 compared to – 7 for Rep. Ross D-AR, and a PBI of 3 versus a -14 PVI for Dan Boren of Oklahoma. Ross at least could be argued as possible. Sen. Pryor didn’t even have a Republican challenger, Blanche Lincoln crushed her opponent, and Democrats in Arkansas have the largest margin in state house outside of Massachusetts in the entire country. Ross’ numbers are as much a reflection of the utter incompetence of the Arkansas GOP at the local level, that bares nor resemblance to their strength at the presidential level.  But Boren in Oklahoma seems a little weird. pl515) suggested correcting for ideology. Since Rep. Boren nearly refused to endorse Obama a case could be made he is getting elected in ruby red Oklahoma (a state where Obama loss every single county) because he is barely a Democrat.

I developed a formula based on standard deviations. Basically I can figure out how much the average rep deviates from their district.  If I then compare where a reps voting pattern falls (in what percentile) and compare it to their district’s PVI, I can develop a “standard deviation factor”. Inside the standard deviation will get a bonus, outside a negative. The idea is that if a Blue Dog has a very conservative record, they may be surviving not because of a districts Democratic leanings but because they deviate from Democratic policies.

For example, if Rep X is the 42 most conservative rep, that would place her in the 90th percentile. But if her district’s PVI was “only” the in the 60th, their is a good chance her margins would be effected. Using a few random samples I found most reps lie within 12% of their district’s PVI.

Using these dummy numbers I then came up with this.  

   SQRT[(30-12)^2 /2] = about 13%

    Her factor would then be 100 – 13 = 0.87.

So her victory margin would be weighted by 0.87 because she is more than 12% beyond her acceptable percentile range it making the victories in her district approximate 13% less “representative”.

    My theory yields the following formula:

        If rep’s voting record is > PVI then

            100 – SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

        else if rep’s voting record < PVI

             100 + SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

I then repeated this formula to calculate a partisanship correction factor. Ranking a members ideology is a subjective decision. Potentially what’s one person “liberal” position, is another person “conservative” ones, remember the wingers developed a model that ranked the Sen. Obama as more liberal than Bernie Sanders or Russ Feingold. But partisanship, how often a member votes with their party is an absolute number. A Democrat who represents a “republican district” would be expected to “break with their party” on votes that don’t reflect their districts values.

I couldn’t find a website that ranks all the districts based on their PVI (I only could find list of them by state not rank, help please anyone), therefor I substituted a PVI ranking with where each member ranked in the Democratic caucus. In the 110th Congress the average Democrat had an ideological ranking of 170 (by the way this is a result of several members being tied, this is the medium not the midpoint). The average of members towards the center was 191, former Daily Kos celeb Ciro Rodriguez fell at exactly 191. The average of members towards the liberal side was 121, which falls between Rep. Larson of Conn. and Rep. Eshoo of CA. As or partisanship in the 110th Congress the average Democrat voted with their party 92.3% of the time.

FOUR BLUE DOGS

Better but still not perfect. PBI Adjustment number two is the model I’m currently most happy with. My model “predict” an Ike Skelton “loss” with a PBI of – 4, Rep. Boren of Oklahoma is at even money with a 0 PBI. Once I’m happy with my model I will do a back run to see how it accesses past members who loss election bids last cycle. As I said earlier, Ross’ numbers are as much a reflection of the utter incompetence of the Arkansas GOP at the local level, that bares nor resemblance to their strength at the presidential level. In other words the Democratic brand is very strong in Arkansas at the local level.

Rep. Baron Hill of Indiana numbers are an indication of the Democratic brands resurgence in the Hoosier state. Rep. Hill won his three races by an average of 11%, with his medium victory being 10%, my model predicts he represents a district where a Democrat should win by 9%.

As a clarification in Adjustment #1, I used a deviation factor based on how far each member was from the center of the Democratic caucus. Adjustment #2 was based on how far each member was from outside the standard deviation of the caucus. In Adjustment #3 I removed the partisanship factor to see what effect it would have.

Because there are “only” 50 states (as opposed to evaluating 435 house members), after Netroots Nation I will have all the states ranked by PVI so I can adjust the Senator’s rankings. I developed Senate factors for the four states the four blue dogs came from. In the interest of full disclosure, my source for ideological rankings is Voteview, and for partisanship it was the Washington Post. This is still a work in progress, I’m making adjustments, and continuing to crunch numbers for more states. I also will use the adjustment factor on a liberal member of congress to see what effect that will have. Finally all future charts will incorporate color schemes.

I have been working on a concept I’m calling PBI or Party Brand Index, as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging voting percentage from the last two presidential elections in each house district, and comparing it to how the nation as a whole voted, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVI indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Secondly PVI can miss trends as it takes 4 years to readjust. The main purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY would be expected to perform. This week I’ll tackle North Carolina.

Last week I tackled Colorado and Virginia in PBI part 2. My general strategy is to work my way “out” from swing states.

Some of the divergences between PBI and PVI in North Carolina are, Rep. Etheridge who has won by margins of between 29-36% goes from being in a lean GOP district to a safe Democratic district. Similarly Rep. McIntyre who has won with between 44 – 37% of the vote goes from being in a lean GOP district (-5 PVI) to a moderate democratic one of (+11 PBI).

For more North Carolina fun in 2008, the Civitas Institute premiered the North Carolina Partisan Index using data from the 2004 General Election. Even though the Civitas Institute is libertarian their numbers and calculations are open sourced, and are valid.

This year, we have updated the NCPI to reflect voters’ choices in the 2008 General Election.

Modeled after the Cook Partisan Voting Index developed for congressional districts, the North Carolina Partisan Index compares the political leanings of voters in each state house and senate district with the partisan voting tendencies of the state as a whole. The end result is a letter (D or R) followed by a number, indicating the extent to which each district leans one way or the other.

The new NCPI was developed using adjusted 2008 data on the elections for Governor and other council of state offices – Lieutenant Governor, Attorney General, Commissioners of Agriculture, Labor, and Insurance, Secretary of State, State Auditor, State Treasurer, and Superintendent of Public Instruction.

______________________________________________________________________________



As a reminder I will review how I calculate Party Brand Index.

To compute PBI I basically did the following. I weighed the last 3 presidential elections by a factor of 0.45. Presidential preference is the most indicative vote since it’s the one politician people follow the most. The POTUS is the elected official people identify with or despise the most, thus illuminating their own ideological identification. I then weighed each house seat by 0.35. House seats are gerrymandered and the local leader can most closely match their districts make up in a way the POTUS can’t. So even though they have a lower profile I still gave them a heavy weight. Lastly I gave the last two Senate elections a weight of 0.2. Senatorial preference can make a difference, although I think it’s less than that of the President or the House members. Also (more practically) because I have to back calculate (estimate) Senate result totals from county results, a smaller number helps lessen the “noise” caused by any errors I may make. Under my system Democratic leaning have a positive number, the GOP has a negative number.

I then developed a way to weight for incumbents.  The reelection numbers for incumbents is so high it would be a mistake to weight a district solely on the fact that an incumbent continues to get elected. There is a long list of districts that have PVI that deviate from their incumbent members, whom none the less keep getting elected. These districts then change parties as soon as the incumbent member retires. This is evidence that incumbency can disguise the ideology of voters in a district.

Next I added a weighting of about 7% for House members. I remember reading that incumbency is worth about 5-10%. Nate wrote in a 538.com article that a VP pick from a small state was worth about a 7% swing, a house seat could in fact be thought of as a small state, that seems as good a number as any to start from. Conversely I will deduct 7% from an incumbents win. I think this will score them closer to the natural weight of a district. By the way I’m weighting the win 7% less, not actually subtracting 7% from the number.  Open seat races will be considered “pure” events and will remain neutral as far as weighting goes.  A seat switching parties will also be considered a neutral event. The 1st defense of a seat by a freshman house member will be given a weighting of 2%. The toughest race for any incumbent is their 1st defense. I decided to adjust for this fact. Note: Indiana’s bloody 9th was a tough call a case could be made that when a seat keeps flipping, and the same two guys run 4 straight times in a row each election should be a neutral event.

Senate weighting is as follows. In state with a single House seat the Senate seat will be weighted the same as a house. In states with multiple seats, the Senate will get a weighting of 2%. Nate Silver stated that a VP pick in a large state is worth this amount. An argument could be made for a sliding scale of Senate weighting from 2-7%, this added complexity may be added at a later date. I will give incumbent presidents a 2% weighting, until I get better data on how powerful a “pull” being the sitting POTUS is, I will give them the same weighting as a senator.

I have been working on (with some much appreciated help from pl515) a concept I’m calling PBI or Party Brand Index, as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging voting percentage from the last two presidential elections in each house district, and comparing it to how the nation as a whole voted, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVI indicates a Democrat shouldn’t win, yet Democrats (outside of the presidency) win quite handily. Secondly why is this the case in Arkansas but not Oklahoma with similar PVI rated districts?

Secondly PVI can miss trends as it takes 4 years to readjust. The main purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but compared to how their generic PARTY would be expected to perform. Last week I calculated PBI for Indiana, this week I tackled Colorado and Virginia.

My best case for arguing against PVI is Indiana.  Bush won Indiana quite easily in 2000 and 2004. The PVI of a number of it’s districts showed them to be quite Red. Yet in 2006 democrats won several districts despite their PVI’s. Also Obama won Indiana in 2008 a state, which based on the make up of the districts PVIs, made little sense. I therefore chose Indiana as my first test case for PBI:

Donnelly in the Indiana 2nd is a perfect example of my issues with PVI. Under PVI Donnelly is in a Republican district with a PVI of -2. But look at how Democrats have recently performed in this district. In 2008 Donnelly won reelection by 37%! Obama won this district by 9 points, and Bayh won it by 22%! Does this sound like a lean GOP district? Under PVI it is, under PBI it’s not it’s a +11 democratic district.

This week I tackled Colorado and Virginia. My general strategy is to work my way “out” from swing states. Both these states have undergone noticable ideological shift. Yet the PVI of their districts haven’t moved as much. This made them ideal candidates.

COLORADO

The big difference in Colorado is that Salazar’s district goes from being a lean Republican one under PVI (-5 Republican), to a lean Democratic one (+4 Democrat), considering that a Salazar has held this same seat for some time this makes more sense. Remember I measuring total party preference not just the presidential preference of a district like PVI measures.

VIRGINIA

Virginia was the first time I had doubts on my ability to compute rough Senate numbers for House districts based on county totals. My estimates from Mark Warner Senate run yielded results of 3540% in Tom Perriello’s (VA-5th) district. This seemed way to high, even though now Senator Warner won the state with 65% of the vote.  At the time Virgil Goode was the representative from the VA-5th, and he lost by only a few hundred votes.- This lead me to do some additional research to try and discover if these numbers were published anywhere. Boy was I wrong Warner actually won the VA-5th by 65%!. Also several of the large victory margins were the results of representatives who ran unopposed. Fixed Party ID, and election results

________________________________________________________________________________

As a reminder I will review how I calculate Party Brand Index.

To compute PBI I basically did the following. I weighed the last 3 presidential elections by a factor of 0.45. Presidential preference is the most indicative vote since it’s the one politician people follow the most. The POTUS is the elected official people identify with or despise the most, thus illuminating their own ideological identification. I then weighed each house seat by 0.35. House seats are gerrymandered and the local leader can most closely match their districts make up in a way the POTUS can’t. So even though they have a lower profile I still gave them a heavy weight. Lastly I gave the last two Senate elections a weight of 0.2. Senatorial preference can make a difference, although I think it’s less than that of the President or the House members. Also (more practically) because I have to back calculate (estimate) Senate result totals from county results, a smaller number helps lessen the “noise” caused by any errors I may make. Under my system Democratic leaning have a positive number, the GOP has a negative number.

I then developed a way to weight for incumbents.  The reelection numbers for incumbents is so high it would be a mistake to weight a district solely on the fact that an incumbent continues to get elected. There is a long list of districts that have PVI that deviate from their incumbent members, whom none the less keep getting elected. These districts then change parties as soon as the incumbent member retires. This is evidence that incumbency can disguise the ideology of voters in a district.

Next I added a weighting of about 7% for House members. I remember reading that incumbency is worth about 5-10%. Nate wrote in a 538.com article that a VP pick from a small state was worth about a 7% swing, a house seat could in fact be thought of as a small state, that seems as good a number as any to start from. Conversely I will deduct 7% from an incumbents win. I think this will score them closer to the natural weight of a district. By the way I’m weighting the win 7% less, not actually subtracting 7% from the number.  Open seat races will be considered “pure” events and will remain neutral as far as weighting goes.  A seat switching parties will also be considered a neutral event. The 1st defense of a seat by a freshman house member will be given a weighting of 2%. The toughest race for any incumbent is their 1st defense. I decided to adjust for this fact. Note: Indiana’s bloddy 9th was a tough call a case could be made that when a seat keeps flipping, and the same two guys run 4 straight times in a row each election should be a neutral event.

Senate weighting is as follows. In state with a single House seat the Senate seat will be weighted the same as a house. In states with multiple seats, the Senate will get a weighting of 2%. Nate Silva stated that a VP pick in a large state is worth this amount. An argument could be made for a sliding scale of Senate weighting from 2-7%, this added complexity may be added at a later date. I will give incumbent presidents a 2% weighting, until I get better data on how powerful a “pull” being the sitting POTUS is, I will give them the same weighting as a senator.

______________________________________

Still to come:

The last major issue is how to deal with the “wingnut” factor. Sometimes a politician like Bill Sali (R-Idaho) or Marylin Musgrove (R-CO)lose because their voting record is outside of the mainstream of their district. I decided to try and factor this in.  

First I had to take a brief refresher on statistics. I developed a formula based on standard deviations. Basically I can figure out how much the average rep deviates from their district.  If I then compare where a reps voting pattern falls (in what percentile) and compare it to their district’s PVI, I can develop a “standard deviation factor”. Inside the standard deviation will get a bonus, outside a negative.

For example, if Rep X is the 42 most conservative rep, that would place her in the 90th percentile. But if her district’s PVI was “only” the in the 60th, their is a good chance her margins would be effected. Using a few random samples I found most reps lie within 12% of their district’s PVI.

Using these dummy numbers I then came up with this.  

   SQRT[(30-12)^2 /2] = about 13%

    Her factor would then be 100 – 13 = 0.87.

So her victory margin would be weighted by 0.87 because she is more than 12% beyond her acceptable percentile range it making the victories in her district approximate 13% less “representative”.

    My theory yields the following formula:

        If rep’s voting record is > PVI then

            100 – SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

        else if rep’s voting record < PVI

             100 + SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

To really do this I need to compute the standard deviation for all 435 reps, which is a pretty large undertaking. Instead  I will do a google search  to see if anyone has already done this. If not well it will take some time. But this would deal with the wingnut factor. Since politician tend to vote relatively close to their districts interest (even changing voting patterns over time) this may not be a major issue. But developing this factor may eventually allow the creation of a “reelection predictor”, so I am still going to work on it.

One last note, the corruption factor (for example Rep. Cao (R-LA) beating former Rep. Jefferson) is outside of any formula I can think of. The only saving grace here is that because my formula uses several elections, the “noise” from a single event will eventually be reduced.

NEXT UP: NC and MO

I have been working on (with some much appreciated help from pl515) a concept I’m calling PBI or Party Brand Index, as a replacement for PVI.  PVI (Partisan Voting Index), which is measured by averaging voting percentage from the last two presidential elections in each house district, and comparing it to how the nation as a whole voted, is a useful shorthand for understanding the liberal v. conservative dynamics of a district. But in my opinion it falls short in a number of areas. First it doesn’t explain states like Arkansas or West Virginia. These states have districts who’s PVI indicates a Democrat would be in a hard position to win, never the less Democrats (outside of the presidency) win quite handily. Secondly why is that the case in Arkansas but not Oklahoma with similar PVI rated districts?

Secondly PVI can miss trends as it takes 4 years to readjust. The main purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but rather compared to how their generic PARTY would be expected to perform. I’m calling this Party Brand Index.

My best case for arguing against PVI is Indiana.  Bush won Indiana quite easily in 2000 and 2004. The PVI of a number of it’s districts showed them to be quite Red. Yet in 2006 democrats won several districts despite their PVI’s. Also Obama won Indiana in 2008 a state, which based on the make up of the districts PVIs, made little sense. I therefor chose Indiana as my first test case for PBI.

Indiana also had a number of other oddity that made it an interesting test case. Indiana has Senators from opposite parties that each won election by large blowouts. Lugar’s in particular was enormous as he was essentially unopposed. Indiana also had a number of districts that flipped in the 3 election cycle expanse that I’m examining. Finally it makes the best case for why PVI can be misleading.

To compute PBI I basically did the following. I weighed the last 3 presidential elections by a factor of 0.45. Presidential preference is the most indicative vote since it’s the one politician people follow the most. The POTUS is the elected official people identify with or despise the most, thus illuminating their own ideological identification. I then weighed each house seat by 0.35. House seats are gerrymandered and the local leader can most closely match their districts make up in a way the POTUS can’t. So even though they have a lower profile I still gave them a heavy weight. Lastly I gave the last two Senate elections a weight of 0.2. Senatorial preference can make a difference, although I think it’s less than that of the President or the House members. Also (more practically) because I have to back calculate (estimate) Senate result totals from county results, a smaller number helps lessen the “noise” caused by any errors I may make.

I was then left with this chart:

I now began to look at the results. Under my system Democratic leaning have a positive number, the GOP has a negative number. Donnelly in the Indiana 2nd is a perfect example of my issues with PVI. Under PVI Donnelly is in a Republican district with a PVI of -2. But look at how democrats have recently performed in this district. In 2008 Donnelly won reelection by 37%! Obama won this ditrict by 9 points, and Bayh won it by 22%! Does this sound like a lean GOP district? Under PVI it is, under PBI it’s not it’s a +11 democratic district.

I then decided to go all Nate Silvaish and gave more recent elections a greater weight. I gave an addition 5% weight to each election as it got closer to the most recent election. To be honest I pulled 5% out of my dairy air but Nate gave a similar weighting to poll results as fresher ones came in 2008, so I copied this formula. This resulted in the following:

The next issue I decided to tackle was to develop a way to weight for incumbents.  The reelection numbers for incumbants is so high it would be a mistake to weight a district soley on the fact that an incumbat continues to get elected. There is a long list of districts that have PVI that devate from their incumbant members, whom none the less keep getting elected. These disticts then change parties as soon as the incumbant member retires. This is evidence that incumbancy can disguise the ideology of voters in a district.

I decided on a weight of about 7% for House members. I remember reading that incumbency is worth about 5-10%. Also Nate wrote in a 538.com article that a VP pick from a small state was worth about a 7% swing, a house seat could in fact be thought of as a small state, that seems as good a number as any to start from. Conversely I will deduct 7% from an incumbents win. I think this will score them closer to the natural weight of a district. By the way I’m weighting the win 7% less, not actually subtracting 7% from the number.  Open seat races will be considered “pure” events and will remain neutral as far as weighting goes.  A seat switching parties will also be considered a neutral event. The 1st defense of a seat by a freshman house member will be given a weighting of 2%. The toughest race for any incumbant is their 1st defense. I decided to adjust for this fact. Note: Indiana’s bloddy 9th was a tough call a case could be made that when a seat keeps flipping, and the same two guys run 4 straight times in a row each election should be a neutral event.

Senate weighting will be as follows. In state with a single House seat the Senate seat will be weighted the same as a house. In states with mutiple seat, the Senate will get a wighting of 2%. Nate Silva stated that a VP pick in a large state is worth this amount. An argument could be made for a sliding scale of Senate weighting from 2-7%, this added complexity may be added at a later date. I will give incumbant presidents a 2% weighting, until I get better data on how powerful a “pull” being the sitting POTUS is, I will give them the same weighting as a senator.

The main purpose of Party Brand Index is to give a better idea of how a candidate does not relative to how the presidential candidate did, but rather compared to how their generic PARTY would be expected to perform. I’m calling this Party Brand Index.

______________________________________

The last major issue is how to deal with the “wingnut” factor. Sometimes a politician like Bill Sali (R-Idaho) or Marylin Musgrove (R-CO)lose because their voting record is outside of the mainstream of their district. I decided to try and factor this in.

First I had to take a brief refresher on statistics. I developed a formula based on standard deviations. Basically I can figure out how much the average rep deviates from their district.  If I then compare where a reps voting pattern falls (in what percentile) and compare it to their district’s PVI, I can develop a “standard deviation factor”. Inside the standard deviation will get a bonus, outside a negative.

For example, if Rep X is the 42 most conservative rep, that would place her in the 90th percentile. But if her district’s PVI was “only” the in the 60th, their is a good chance her margins would be effected. Using a few random samples I found most reps lie within 12% of their district’s PVI.

Using these dummy numbers I then came up with this.  

   SQRT[(30-12)^2 /2] = about 13%

    Her factor would then be 100 – 13 = 0.87.

So her victory margin would be weighted by 0.87 because she is more than 12% beyond her acceptable percentile range it making the victories in her district approximate 13% less “representative”.

    My theory yields the following formula:

        If rep’s voting record is > PVI then

            100 – SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

        else if rep’s voting record < PVI

             100 + SQRT[({Record percentile – PVI} – Standard PVI Sigma)^2 /2] = factor

To really do this I need to compute the standard deviation for all 435 reps, which is a pretty large undertaking. Instead  I will do a google search  to see if anyone has already done this. If not well it will take some time. But this would deal with the wingnut factor. Since politician tend to vote relatively close to their districts interest (even changing voting patterns over time) this may not be a major issue. But developing this factor may eventually allow the creation of a “reelection predictor”, so I am still going to work on it.

One last note, the corruption factor (for example Rep. Cao (R-LA) beating former Rep. Jefferson) is outside of any formula I can think of. The only saving grace here is that because my formula uses several elections, the “noise” from a single event will eventually be reduced.

Next Up: Colorado ( have the data done already) and Virginia