For a more complete explanation of the process used, please refer to my initial post on the subject.

Recap of today's official result:

            Recount         Original
Coleman:   1,044,255   -   1,046,101  =  -1,846
Franken:   1,040,285   -   1,042,208  =  -1,923

Franken Net Today: -100 (-20 Yesterday, -5 day before, 14)
Total Franken Net:  -77
Franken Deficit:    292

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,858; the largest precinct that has completed a recount. The dotted lines represent the two-day-ago regression, while the dashed lines represent yesterday's. The thin lines depict the linear regression while the thicker curvy lines present a quartic regression. The black bars emanating from the x axis represent the number of precincts, statewide with x number of total votes. There are a few straggler precincts between 2600 and 6621 that are not depicted due to the resolution of the graph.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 17,270 additional votes, that lie in precincts that surpass the 3,858 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,858 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman: -1599 (+/- 92)
            Franken: -1612 (+/- 95)

            Net Franken: -83.169 (+/- 187)

Franken is now actually projected to lose votes, about 83 to be exact but with all the challenges outstanding, these projections are far from accurate. There are also 17,270 votes excluded from the above extrapolation. If those additional 17,270 votes are applied to the process, nothing really changes but the margin of error:

            Coleman: 1610- (+/- 236)
            Franken: 1623- (+/- 244)            

            Net Franken: -83.703 (+/- 580)

The graph below shows the number of challenges by each candidate relative to how they are performing in a given precinct. The dots represent challenges; a dot to the north of the x-axis represents a precinct that the candidate is currently winning. A dot to the south shows the number of challenges in a precinct that each candidate is currently losing. There are no negative challenges.

The next graph shows each candidate's challenge regression on the same graph:

It appears that a larger percent of Franken challenges occur in precincts he is currently winning, relative to Coleman's percentage. They both appear to be challenging more ballots in precincts they are winning, but Franken at a higher rate. This discrepancy may allow Franken to make up additional votes, but an exact number is impossible to predict.

This analysis is valid as of right now (10:22 PM CT 11/26), after 89.64% of all votes have been recounted. Recounting should resume on the first of December after a Thanksgiving recess.

For a more complete explanation of the process used, please refer to my initial post on the subject.

Recap of today's official result:

           Recount       Original
Coleman:   978,751   -   980,121  =  -1,370
Franken:   976,187   -   977,580  =  -1,393

Franken Net Today:  -20 (-5 Yesterday, 14 day before, 43)
Total Franken Net:   23
Franken Deficit:    192

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,858; the largest precinct that has completed a recount. The dotted lines represent the two-day-ago regression, while the dashed lines represent yesterday's. The thin lines depict the linear regression while the thicker curvy lines present a quartic regression. The black bars emanating from the x axis represent the number of precincts, statewide with x number of total votes. There are a few straggler precincts between 2600 and 6621 that are not depicted due to the resolution of the graph.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 17,270 additional votes, that lie in precincts that surpass the 3,858 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,858 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Franken: -1614 (+/- 63)
            Coleman: -1599 (+/- 63)

            Net Franken: -12.642 (+/- 126)

Franken is now actually projected to lose votes, about 13 to be exact but with all the challenges outstanding, these projections are far from accurate. There are also 17,270 votes excluded from the above extrapolation. If those additional 17,270 votes are applied to the process, nothing really changes but the margin of error:

            Coleman: -1610 (+/- 172)
            Franken: -1626 (+/- 175)            

            Net Franken: -12.928 (+/- 357)

The graph below shows the number of challenges by each candidate relative to how they are performing in a given precinct. The dots represent challenges; a dot to the north of the x-axis represents a precinct that the candidate is currently winning. A dot to the south shows the number of challenges in a precinct that each candidate is currently losing. There are no negative challenges.

The next graph shows each candidate's challenge regression on the same graph:

It appears that a larger percent of Franken challenges occur in precincts he is currently winning, relative to Coleman's percentage. They both appear to be challenging more ballots in precincts they are winning, but Franken at a higher rate. This discrepancy may allow Franken to make up additional votes, but an exact number is impossible to predict.

This analysis is valid as of right now (4:52 PM CT 11/26), after 80.62% of all votes have been recounted. Check back tomorrow for the day 8 result and accompanying analysis.

For a more complete explanation of the process used, please refer to my initial post on the subject.

Recap of today's official result:

           Recount      Original
Coleman:   908,063   -   909,114  =  -1,051
Franken:   889,891   -   890,899  =  -1,008

Franken Net Today:   -5 (-52 Yesterday, 14 day before, 43)
Total Franken Net:   43
Franken Deficit:    172

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,494; the largest precinct that has completed a recount. The dotted lines represent the two-day-ago regression, while the dashed lines represent yesterday's. The thin lines depict the linear regression while the thicker curvy lines present a quartic regression. The black bars emanating from the x axis represent the number of precincts, statewide with x number of total votes. There are a few straggler precincts between 2600 and 6621 that are not depicted due to the resolution of the graph.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 32,186 additional votes, that lie in precincts that surpass the 3,494 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,494 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman: -1356 (+/- 51)            
            Franken: -1300 (+/- 49)            

            Net Franken: 55.679 (+/- 100)

Franken falls short of the pre-recount deficit by about 56 votes, with 32,186 votes entirely ignored and the challenges left uncounted. If those additional 32,186 votes are applied to the process, nothing really changes but the margin of error:

            Coleman: -1374 (+/- 151)
            Franken: -1318 (+/- 147)

            Net Franken: 55.987 (+/- 298)

The graph below shows the number of challenges by each candidate relative to how they are performing in a given precinct. The dots represent challenges; a dot to the north of the x-axis represents a precinct that the candidate is currently winning. A dot to the south shows the number of challenges in a precinct that each candidate is currently losing. There are no negative challenges.

The next graph shows each candidate's challenge regression on the same graph:

It appears that a larger percent of Franken challenges occur in precincts he is currently winning, relative to Coleman's percentage. They both appear to be challenging more ballots in precincts they are winning, but Franken at a higher rate. This discrepancy may allow Franken to make up additional votes, but an exact number is impossible to predict.

This analysis is valid as of right now (3:58 PM CT 11/25), after 74.18% of all votes have been recounted. Check back tomorrow for the day 7 result and accompanying analysis.

There were no new recount results yesterday, so I took the opportunity today, to provide an overview of where the recount currently stands. The images below serve to illustrate the current state of the recount, with respect to the originally certified result. The first image below draws from the official SOS recount result as of 8 PM CT last night.

A complete description of the colorization process for the above image is available for those who are curious.

The following image shows the result of the certified result as of November 18th.

A detailed description of the colorization process is also available for the above image. Both maps are embeddable.

The next series of graphs illustrates the current state of the race in each county alphabetically. There are several counties that have yet to begin the recount process; these counties are all depicted on the last graph in the series. For counties with recount data, the certified result, the current recount result, and the number of challenges by Coleman and Franken are depicted. The vote totals use the left y-axis, while the challenge totals use the right. Also be aware of the inconsistency of the scaling between each graph; if the scale is normalized the data is uninterpretable in most counties.

I talked yesterday about the correlation between challenges and the current vote totals in a given precinct. I concluded that both candidates seem to be challenging more votes in precincts where they are currently leading. The maps and graphs above seem to reaffirm this assertion with one notable exception; Hennepin County.

Franken won Hennepin County based on the certified result, is currently leading and will eventually win the county; but Norm Coleman has challenged 23 more ballots than Al Franken in the County containing Minneapolis. Hennepin County has the largest population of any county in Minnesota and contains at least 6,000 residual votes. It is very likely that both candidates are challenging votes for Al Franken; as these residual votes tend to lean towards the Democratic candidate.

I think its very possible for Franken to make up votes when the Hennepin challenges are finally reviewed. Franken may still have ground to gain in Ramsey County too, judging by their anemic progress, of just less than 50%. Ramsey County has a similar residual vote pattern and it's very possible that the same trend occurring in Hennepin could also be observed in Ramsey County as the remaining ballots are recounted.

The Minnesota Senate Recount is in full swing, and with new numbers posted daily, it can, at times seem overwhelming. But fear not, I've created a snazzy new image that dynamically updates each night (at 8 PM CT) upon the release of new data by the Secretary of State. The maps may initially seem cryptic, but its all explained below. I've also posted html and BBcode for each image so you can help spread the word, and keep others in the know.

The data from the SOS is provided at the precinct level, but our maps are only presentable at the county level. To compensate, our algorithm loops through the entire result adding each precinct's data to the associated county; the result of this computation provides a county level extrapolation that allows for a graphical visualization of the most current recount result.

Minnesota Senate Recount by County (490 x 610 Pixels):

HTML Code:

<a href="http://voteforamerica.net"><img src="http://www.voteforamerica.net/image.aspx?Type=Minnesota_9" alt="VoteForAmerica.net Minnesota Recount Results by County Map" border="0" /></a>

Forum Code:

[url=http://voteforamerica.net][img]http://www.voteforamerica.net/image.aspx?Type=Minnesota_9[/img][/url]

I'll now attempt to explain the peculiarities of each map and the associated colorization process.

Franken vs Coleman (0%, 1, 3)

The upper left most map of Minnesota depicts the percent margin of victory in each county using the unofficial results of the Minnesota Senate election. All totals pertaining to unlisted candidates are included in our analysis. The color purple represents counties decided by less than 3% of the final margin, decrementing by 1 percent for each change in color. The darkest purple represents a voting margin of 1% or less, the middle purple between 1% and 2% and the lightest purple between 2-3%. The blue and red then represent counties that were decisively won by either Al Franken or Norm Coleman, respectively. The red and blue counties follow a 3% scale. The lightest blue (or red) represents a 3-6% margin, followed by 6-9% as the color grows darker, then finally a 9%+ margin of victory for the darkest blue (or red).

% Ballots Counted (0%, 16.6, 16.6)

The upper right most map illustrates the ratio between the number of recounted votes and the certified vote totals for each county. The counties in gray have yet to begin the recount process. If the percent of recounted votes over certified votes is less than 50%, the county is colored purple; if this ratio is greater than 50% the county is colored with respect to winning candidate, either blue or red. The color gradations take place in 16.6% divisions. The tint, for each color represents the highest range. For purple this range is 33.3-50% segment while the darkest blue (or red) represents the 83.3%-100% division. The color grows lighter as the percentage decreases.

Ballots Challenged (0%, 1, 7)

The bottom left map attempts to illustrate the discrepancy between challenged votes, in each county, by each candidate. The purple counties are divided into single challenge divisions while the partisan colors are separated by seven. The partisan coloring begins after a four challenge offset with blue representing a Franken challenge and red a Coleman challenge. Gray depicts counties with no challenge discrepancy; this may or may not include counties that have yet to begin recounting.

Recounted Vote Gain (0%, 1, 5)

The last image provides the net vote gain, per county based on the difference between the current recount result and the initially certified count. If less than 3 votes have been gained by either candidate in a given county that county will be colored purple in single vote divisions. If a candidate has a net gain of more than 3 votes in any county it will be colored accordingly; blue for Franken, and red for Coleman. The partisan coloring is done in 5 vote increments. Gray again represents counties that have yet to begin their recount process.

Preliminary Minnesota Election Results by County (490 x 610 Pixels):

HTML Code:

<a href="http://voteforamerica.net"><img src="http://www.voteforamerica.net/image.aspx?Type=Minnesota_4" alt="VoteForAmerica.net Minnesota Recount Results by County Map" border="0" /></a>

Forum Code:

[url=http://voteforamerica.net][img]http://www.voteforamerica.net/image.aspx?Type=Minnesota_4[/img][/url]

The previous map follows about the same pattern as the first, but a full description is available in a previous article.

For a more complete explanation of the process used, please refer to my initial post on the subject.

Recap of today's official result:

           Recount      Original
Coleman:   809,677   -   809454  =  -669
Franken:   784,310   -   784022  =  -621

Franken Net Today: -52 (14 Yesterday, 43 day before, 43)
Total Franken Net:  48
Franken Deficit:    67

I've added a challenge regression into the mix (at the end), above and beyond the error rate analysis I have routinely conducted.

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,494; the largest precinct that has completed a recount. The dotted lines represent the first day's regression, while the dashed lines represent yesterday's. The thin lines depict the linear regression while the thicker curvy lines present a quartic regression. The black bars emanating from the x axis represent the number of precincts, statewide with x number of total votes. There are a few straggler precincts between 2600 and 6621 that are not depicted due to the resolution of the graph.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 32,186 additional votes, that lie in precincts that surpass the 3,494 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,494 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman Gain: -968.186777451647 (+/- 30)
            Franken Gain: -911.227610123547 (+/- 28)

             Franken Net: 56.9591673280993 (+/- 58)

Franken falls short of the pre-recount deficit by about 160 votes, with 32,186 votes entirely ignored and the challenges left uncounted. If those additional 32,186 votes are applied to the process, Franken less than a single vote:

                
            Coleman Gain: -981.037349428085  (+/- 145)
            Franken Gain: -923.639100212363  (+/- 101)

             Franken Net: 57.3982492157215  (+/- 246)

The graph below shows the number of challenges by each candidate relative to how they are performing in a given precinct. The dots represent challenges; a dot to the north of the x-axis represents a precinct that the candidate is currently winning. A dot to the south shows the number of challenges in a precinct that each candidate is currently losing. There are no negative challenges.

The next graph shows each candidate's challenge regression on the same graph:

It appears that a larger percent of Franken challenges occur in precincts he is currently winning, relative to Coleman's percentage. They both appear to be challenging more ballots in precincts they are winning, but Franken at a higher rate. This discrepancy may allow Franken to make up additional votes, but an exact number is impossible to predict.

This analysis is valid as of right now (10:38 PM CT 11/22), after 65.65% of all votes have been recounted. Check back tomorrow for the day 5 result, if there is one, it's Sunday and I don't know if they continue counting.

For a more complete explanation of the process used, please refer to my initial post on the subject.

Recap of today's official result:

           Recount      Original
Coleman:   751898   -    752438 =  -540
Franken:   723378   -    723818 =  -440

Franken Net Today: 14 (43 yesterday, 43 day before)
Total Franken Net: 100
Franken Deficit:  115

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,494; the largest precinct that has completed a recount. The dotted lines represent the first day's regression, while the dashed lines represent yesterday's. The thin lines depict the linear regression while the thicker curvy lines present a quartic regression. The black bars emanating from the x axis represent the number of precincts, statewide with x number of total votes. There are a few straggler precincts between 2600 and 6621 that are not depicted due to the resolution of the graph.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 32,186 additional votes, that lie in precincts that surpass the 3,494 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,494 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman Gain: -832.977997236315 (+/- 24)
            Franken Gain: -681.983934516065 (+/- 20)

             Franken Net: 150.994062720251 (+/- 44)

Franken falls short of the pre-recount deficit by about 65 votes, with 32,186 votes entirely ignored and the challenges left uncounted. If those additional 32,186 votes are applied to the process, Franken adds an additional vote:

                
            Coleman Gain: -844.072810341355 (+/- 103)
            Franken Gain: -691.688480481963 (+/- 97)

             Franken Net: 152.384329859392 (+/- 200)

As of right now (10:24 PM CT 11/21), after 60.86% of all votes have been recounted. Check back tomorrow for the day 4 result, if there is one, its Saturday and I don't know if they continue counting.

For a more complete explanation of the process used, please refer to my initial post on the subject.

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 3,494; the largest precinct that has completed a recount. The dotted lines represent the previous day's regression.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 32,186 additional votes, that lie in precincts that surpass the 3,494 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 3,494 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman Gain: -514.680649876274  (+/- 8)
            Franken Gain: -295.310532175405  (+/- 4)

             Franken Net: 219.37011770087  (+/- 12)

Franken makes up the pre-recount deficit of 215 votes, while 32,186 votes are entirely ignored. If those additional votes are applied to the process, Franken adds an additional 9 votes:

           Coleman Gain: -539.655665129103  (+/- 1648)
           Franken Gain: -311.596023329002  (+/- 1916)

            Franken Net: 228.059641800101  (+/- 3564)

As of right now (9:20 PM CT 11/20), after 42.33% of all votes have been recounted. Check back tomorrow for the day 3 result.

Using the pretense of my pre-recount audit analysis I've incorporated the initial recount returns into the regression mechanism.

For a more complete explanation of the process used, please refer to my initial post on the subject.

The following two graphs depict the number of votes gained or lost with respect to the total votes recorded for each precinct that has completed their recount. Each dot represents the change between the originally certified result and the post recount tally in a given precinct with x number of total votes:

And the current functions used in conjunction with the regression:

Now again using those functions, the following two graphs further illustrate the regression interpolation. The graph below illustrates the regression on precincts whose vote totals are less than 2,668; the largest precinct that has completed a recount.

The graph below goes further and fits the regression onto all 4,130 precincts statewide; this forces 214,813 additional votes, that lie in precincts that surpass the 2,668 vote threshold, into the predetermined trend.

Using the previous two graphs, and the functions they represent, a projection can be made for the cases covered. The first case simply includes precincts with 2,668 or less, while precincts above and beyond that figure are entirely ignored. For each precinct, the total vote total is taken and applied to the listed function for each candidate. The result is then added to that candidate's sum, and the next precinct is calculated. This process is done using precinct results from the final certification.

            Coleman Gain: -407.600629299817 (+/- 2.4)
            Franken Gain: -194.977755790799 (+/- .62)

             Franken Net: 212.622873509018 (+/- 3.02)

Franken nearly makes up the pre-recount deficit of 215 votes, while 214,813 votes are entirely ignored. If those additional votes are applied to the process, Franken takes a decisive lead:

           Coleman Gain: -540.336882632741 (+/- 4925)
           Franken Gain: -237.764374574729 (+/- 6342)

           Franken Net: 302.572508058013 (+/- 11267)

As of right now (4:16 AM CT 11/20), after 15.49% of all votes have been recounted, Franken stands poised to become Minnesota's next Senator. Later today, after the second day of recounting is over, and the officially unofficial results are posted at the SOS site, I'll post another article with the new data included in the analysis. As each day goes by, the graphs, and projections should begin to converge. If Franken gains another 40 votes throughout the upcoming day, Team Coleman is definitely in trouble. The only way to find out is by counting the votes.

The following analysis uses data provided by Prof. Jeffrey B. Lewis of UCLA who coauthored Residual Votes in the 2008 Minnesota Senate Race. My analysis is entirely independent and not in anyway shape or form associated with the aforementioned individuals or institutions. The data contains an exhaustive compilation of the revised Post Election Audit Review published by the Office of the Secretary of State in Minnesota.

I parsed the data and created the graphs shown below. The three graphs immediately following this paragraph depict the number of votes gained or lost with respect to the total votes recorded for each precinct included in the audit. The dots represent the change between the machine counted votes and hand recorded votes surveyed in the audit. A dot located at (443, 2) represents a gain of 2 votes by the given candidate in a precinct with 443 total votes. Each approximation line uses a least-squares nonlinear regression for arbitrary fit functions as provided by Mathcad's genfit() function:

You'll notice that the graphs range from 0 to 2,645 votes; this is because the largest precinct surveyed in the audit contained just 2,645 votes.

The following functions correspond to the resulting regression's interpolation. The input x represents the number of votes in a given precinct. Although it appears as though the first two terms cancel, the truncated three decimal numbers are in actuality much more precise:

The following graph illustrates the best fit regression functions, developed above, extrapolated out to include the vote totals of all 4,130 precincts statewide. As of right now (12:55 AM CT 11/20) the Otstego precinct of Wright County currently leads the state in total votes with 6,621.

I plan on rehashing this analysis with the current results of the recount within the immediate future. Today Franken gained 43 votes as a result of today's recount over the certified returns published on November 18th:

                       Nov. 18       Recount    Change
       Coleman (R)     195,708       195,638     -70
       Franken (D)     180,950       180,923     -27

                       Franken Net Gain:          43

Franken made up much of his ground in St. Louis County, as I have previously suggested, by gaining 40 votes and losing 16. St. Louis County has recounted just 77 of 232 precincts (33.18%) and has contributed to a 23 vote decrease in Coleman's lead. Overal, Franken gained votes in 50 precincts, and lost votes in 79 while Coleman gained votes in 40 precincts and lost in 92.

While today's result appears to be good news for Team Franken, the result of a Ramsey County lawsuit may be even better. The Franken Campaign sued the St. Paul county for the names of voters whose absentee ballots had been rejected. The presiding judge issued this statement: "With each passing hour, the Franken campaign is irreparably harmed in its efforts to ensure that each valid vote is properly counted." Franken's attorneys said the ruling strengthens their case in other counties and with the canvassing board over the ongoing issue of "wrongfully" invalidated absentee ballots.

On Friday November 14th, the Post Election Audit Review was carried out in 53 (1.297%) precincts across the state. The audit recounted 54,036 (1.872%) Senate votes statewide and included voter intent into some determinations.

Throughout the entire process 106 votes were corrected and Franken accumulated 2 additional votes. The table below depicts the results of this audit for each precinct in which it was conducted. The error rate measures the number of corrections relative to the total number of votes in each precinct. Votes for Dean Barkley and other Senate candidates were included in the math, but are not depicted due to space limitations.

Based upon the random sample, Norm Coleman won 33 (62.26%) precincts, while Franken won the other 20. This ratio slightly exaggerates support for Norm Coleman; using the complete sample of 4,130 precincts, Coleman wins just 60.46% of all statewide precincts. If the result from the audit sample is extrapolated out into the remaining precincts, where Franken accrues 2 votes for every 53 precincts, the Democratic challenger should pick up 156 votes through the recount process. If an adjusted ratio is used, which reflects the statewide precinct ratio, Franken gains approximately 5 votes in addition to the aforementioned 156.

Another item to consider is the high error rate in St. Louis County. St. Louis County is the largest county in Minnesota, and is located in the upper right portion of the state:

Using the information presented in the above image (click on the image for a more thorough explanation) St. Louis County may hold the key to a Franken comeback. Noting the results of the Audit, where Franken accrued 3 votes in just five (three are not depicted in the table because there was no change) precincts from St. Louis County, Franken could gain 120 votes if the aforementioned trend continues in all 232 precincts of St. Louis County.

The Minnesota Senate recount hasn't started, but the already fierce campaign continues to escalate. No amount of complaining will win this election, but both sides seem content to try. Each campaign is spewing propaganda, but the vast majority of media coverage seems to involve Republican talking points, rather than facts. This is an obvious problem; as a result the actual information takes a backseat to spin and deception. Using the current results, as provided by the Minnesota Secretary of State's Office, the reasoning behind the right-wing's focus can be extracted:

There's a lot of information in the image above that may not be directly apparent; I will do my best to explain its meaning; if you don't care about the peculiarities of each maps' coloration, skip to the ellipses.

Franken vs Coleman (0%, 1, 3)

The upper left most map of Minnesota depicts the percent margin of victory in each county using the unofficial results of the Minnesota Senate election. All totals pertaining to unlisted candidates are included in our analysis. The color purple represents counties decided by less than 3% of the final margin, decrementing by 1 percent for each change in color. The darkest purple represents a voting margin of 1% or less, the middle purple between 1% and 2% and the lightest purple between 2-3%. The blue and red then represent counties that were decisively won by either Al Franken or Norm Coleman, respectively. The red and blue counties follow a 3% scale. The lightest blue (or red) represents a 3-6% margin, followed by 6-9% as the color grows darker, then finally a 9%+ margin of victory for the darkest blue (or red).

Obama vs McCain (0%, 1, 3)

Instead of depicting the unofficial Senate Results, the upper right most map shows the result of the Presidential Election between Barack Obama and John McCain; the color scales are identical to those used in the first Senate map.

Franken vs Obama (12%, .166, 1.5)

The lower left most map differs from the previous two, but conveys a combination of their data. Based on the current statewide result, Al Franken received 12.07903% less votes than Obama. For the proposes of this graph that number was rounded to 12% and used as a baseline. If Al Franken performed better than that 12% difference in a given county, the county is designated as blue; if Obama did better, the county was colored orange. The purple represents counties where the adjusted margin, after the 12% is applied, lies within .5%. The darkest purple shows counties decided by .166% or less, the middle purple .333% or less, and the lightest purple .5% or less. The blue and orange counties follow a 1.5% scale. The lightest blue (or orange) represents a 1.5-3% adjusted margin, then a 3-4.5% margin for the next darkest color. followed by the darkest blue (or orange) representing a 7%+ adjusted margin of victory for the associated candidate.

Residual Votes (0%, 200, 2000)

The final map, the lower rightmost, reveals the most interesting conclusion of the currently unofficial result. This map illustrates the difference between the number of total ballots cast and the number cast in the Senate Election; this difference is commonly referred to as the residual vote. The counties are colored with respect to the first map, but the degree of coloration corresponds to the number of residual votes in a given county. The lightest color depicts a county with fewer than 200 residual votes, followed by the middle color with 200-2,000 residual votes, and finally the darkest color representing a county with more than 2,000 residual votes.

...

The right-wing (LetFreedomRingBlog.com, Amanda Carpenter, HotAir.com, RedState.com) seems content to perpetuate misinformation of the residual vote variety. Contrary to their belief, the relationship between Franken's outcome and Obama's outcome has absolutely nothing to do with the current, and likely spread of residual votes. Are they simply refusing to accept the empirical facts or are they already starting to play the blame game?

The raw data, our results, depict a very plausible comeback by Al Franken during the recount. Drawing from our Residual Votes map, there were 34,703 votes cast on a presidential ballot that were not cast in the Senate race. Of those 34,703 residual votes, 17,839 (51.04%) of them occurred in counties that Franken won, a margin of 975. If just 21.13% of the statewide residual votes are classified as either undervotes or overvotes Al Franken should statistically pull ahead during the recount.

If you want a higher degree of residual vote accuracy, perhaps the county and precinct level results should be taken into consideration. Using county level data, Franken needs slightly less than 40% of the 34,703 previously identified residual votes and just 13.2% of that total using precinct data. All modes of interpretation clearly point to a likely surge by Franken during the recount process due to these residual votes; a fact the right seems content to ignore.

If you don't trust my analysis, perhaps you should argue with Dartmouth College or MIT instead.

Update: The obvious discrepancy in my initial result revolved around the exclusion of Presidential residual votes. The MN SOS website happens to provide a count for the total number of ballots cast in each county which allows us to find the number of Presidential residual votes through a simple subtraction. The Presidential residual vote was then added to the Senate residual vote to arrive at the total number of residual votes cast in the Minnesota election. I've updated the numbers within the the article to reflect this change. The inclusion of this new information increased the number of effective residual votes needed, for a Franken victory, by roughly 10%; definitely good news for Norm Coleman's reelection effort.

With the Alaska Senate race still in full swing, a few pieces of information bear consideration. The Democratic challenger, Mark Begich of Anchorage currently holds a 1,022 vote lead over the convicted felon and Republican, Ted Stevens. The percent margin is exceptionally close, just .35%, a slim enough margin to qualify for a state funded recount under Alaska Statutes § 15.20.450:

If the recount includes an office for which candidates received a tie vote, or the difference between the number of votes cast was 20 or less or was less than .5 percent of the total number of votes cast for the two candidates for the contested office...the application need not include a deposit, and the state shall bear the cost of the recount.

The recount is not automatically triggered, as is the case in Minnesota, but rather must be requested under Alaska Statutes § 15.20.430:

A defeated candidate or 10 qualified voters who believe there has been a mistake made by an election official or by the counting board in counting the votes in an election, may file an application within five days after the completion of the state review to the director for a recount of the votes from any particular precinct or any house district and for any particular office, proposition, or question.

As the Alaska count currently stands, a state funded recount is a realistic outcome.

In preparation for the recounts, in both Minnesota and possibly Alaska, I sought to collect geographic information from each state that would allow for the creation of county and precinct level maps. Minnesota went off without a hitch (more on that later) but I ran into some issues with Alaska.

There appears to be an attempt by the Alaskan Government to provide the mapping data I sought, albeit a very sorry attempt. The page appears to provide a plethora of maps and resources, but any effort to actually obtain said files, results in a 505 error. I could always shell over the $150 required to purchase this data from the Alaska Department of Natural Resources, but I don't have the urge. Why does Alaska need to charge for this government funded information, haven't the taxpayers already paid?

Apparently a viable link to an image of Russia remains of paramount importance, while essentially every other link, to the actual data, fails. Unbelievable.

Source: Alaska Department of Natural Resources

To be fair, this image exists in a specific surveying format; the equal area model. That said, there is absolutely no reason why the page could not have been more logically oriented. But ignoring this issue; why is a bizarre image of Russia available on a government sponsored server, while data directly pertaining to the state, and country for which that government seeks to represent, unavailable?

It all makes sense now.

In my previous article on undervotes, which you should probably read before continuing, I looked at the undervote totals in the 87 counties of Minnesota. While counties provide a fair level of resolution, the precinct data could provide an additional level of detail. This time around I used the precinct data, from the MN Secretary of State's Office, for both the Senate and Presidential elections. The resulting calculation was performed identically to the method described in the aforementioned article, just with additional data.

The precinct calculation yielded 17,831 undervotes in precincts won by Obama, while the county result ranged roughly 500 votes higher at 18,328. This deviation initially appears worrisome, for Franken, but the final tally shows a staggering drop off in the number of undervotes necessary for a Franken victory. An undervote total of just 3,280 votes, evenly distributed across each precinct and inline with the current trend, delivers Franken the votes necessary to overcome his 206 vote deficit. These 3,280 votes represent just 13.2% of all potential undervotes, as identified in the preceding article.

Given this result, I have no idea if 13.2% exists in the realm of likelihood, but its better news, for Franken, than our previously determined 40% threshold. Tomorrow (or the next day) I plan on exploring the historical correlation between overvotes, undervotes, and the Presidential to Senate ratio; I'll report back with my findings.

With each passing day, Norm Coleman's lead, currently 206, slips further into obscurity as the final canvassing concludes. No amount of crying, here, here and here, by Norm Coleman will change the fact that the final canvassing total often differs from the election night tally; as evidenced in a document released by the Franken campaign. If you want further evidence, a quick glance to our previous post, illustrates a perceived inequality with relation to the expected trend and the actual result.

	       4:21 AM       Proj       Actual      Diff.
   Franken      80,600      92,990      88,620     -4,370
   Barkley      20,550      23,415      22,652      -914
   Coleman      55,840      63,650      61,513     -2,137

Drawing from the table, Coleman may continue to bleed votes as these four northern counties continue to finalize their vote tally in preparation for the automatic recount. But even if Franken cannot close the 206 vote advantage Norm Coleman currently holds, Franken may be able to make up ground in the undervotes. Undervotes occur when the ballot reading optical scanner fails to record a vote in a given race, in this case the Senate race.

Using data provided by the Minnesota Secretary of State for the presidential and senate elections, I tabulated a massive grid with data from each of Minnesota's 87 counties.

I took this data and calculated the discrepancy between Presidential votes and Senate votes, it turned out to be 24,830; in other words, 24,830 people cast a vote for President, but did not cast a machine defined vote in the Senatorial election. This number represents a maximum number of undervotes that could possibly exist (technically it could be more, but its statistically improbable); in reality it is likely fewer. There are three scenario's that could result in a vote being marked as an undervote by the optical scanner; perhaps the person legitimately did not vote for a Senate candidate, there may have been a random malfunction or perhaps the voter failed to adequately mark their Minnesota ballot as depicted below:

The three above instances clearly demonstrate the intent to vote, but would still none the less be classified as undervotes. It is these instances which could very well allow Franken to surpass Coleman upon the conclusion of these undervotes in the impending recount.

Using the previously collated data table, I concluded that there is a maximum of 18,328 undervotes present in counties won by Obama, but this isn't news, the Associated Press beat me to it, but I still have more to offer. This tilt towards Obama translates into a rather large lead for Franken in counties that have a high number of undervotes. If a uniform distribution of undervotes exists in each of Minnesota's 87 counties and the current trend holds in each county, Franken could pull ahead if just slightly less than 40% of the potential undervotes are infact undervotes. Franken will likely win the recount if the trends hold and there are 9,848 undervotes evenly distributed across the state.

Al Franken stands to gain 4580 votes if the remaining precincts follow their trend. This is essentially a death sentence for Coleman as he only leads by 2568. The table below includes all uncounted counties and the probable vote allocation based on the current percent of precincts reporting and the current results.

St. Louis	Est.		Cass		Est.
Al	59050	70298		Norm	7407	8051
Norm	34812	41443		Al	5699	6195
Dean	12920	16326		Dean	2380	2587
% Rprt	0.84			% Rprt	0.92	
						
Pine		Est.		Itasca		Est.
Al	5824	5943		Norm	7977	8397
Norm	5644	5759		Al	10027	10555
Dean	2471	2521		Dean	2779	2925
% Rprt	0.98			% Rprt	0.95	
						
	Total	Current	Est.	Chng		
	Norm	55840	63650	7810		
	Al	80600	92990	12390		
	Dean	20550	23415	2865		
	% Rprt			4580		
	
	Current Votes
	Franken:	1203315
	Coleman:	1205883

Data taken from CNN at 4:21 AM CT.

Based upon exit polls provided by MSNBC, it appears as though Obama has this election in the bag. Obama will win Virginia, Indiana, Ohio, Missouri, Florida and North Carolina if the exit polls are accurate.

	Georgia			  Florida		
	Obama	McCain	% Total	  Obama	  McCain  % Total
Male	51	56	0.46	  49	  49	  0.47
Female	54	55	0.54	  50	  48	  0.53
							
Male	23.46	25.76		  23.03  23.03	
Female	29.16	29.7		  26.5	 25.44	
Total	52.62	55.46		  49.53	 48.47	
							
	Alabama			  North Carolina		
	Obama	McCain	% Total	  Obama	  McCain  % Total
Male	38	60	0.45	  46	  53	  0.46
Female	45	55	0.55	  56	  43	  0.54
							
Male	17.1	27		  21.16	  24.38	
Female	24.75	30.25		  30.24	  23.22	
Total	41.85	57.25		  51.4	  47.6	
							
	Virginia		  Missouri		
	Obama	McCain	% Total	  Obama	  McCain  % Total
Male	52	46	0.46	  52	  47	  0.46
Female	56	44	0.54	  54	  45	  0.54
							
Male	23.92	21.16		  23.92	  21.62	
Female	30.24	23.76		  29.16	  24.3	
Total	54.16	44.92		  53.08	  45.92	
							
	Indiana			  Ohio		
	Obama	McCain	% Total	  Obama	  McCain  % Total
Male	49	49	0.47	  52	  46	  0.48
Female	55	44	0.53	  54	  44	  0.52
							
Male	23.03	23.03		  24.96	  22.08	
Female	29.15	23.32		  28.08	  22.88	
Total	52.18	46.35		  53.04	  44.96	

Data gathered from MSNBC's exit polls.

Non-Partisan Polling, Final Projections:

Partisan/Internet Polling, Final Projections:

A ton of new polls, with 86 coming from Polimetrix/YouGov. They are marked as a partisan pollster because the internet was used to gather their data. While this in itself does not constitute partisanship, the sample is likely skewed by the internet demographic. Our algorithm only defines partisan and non-partisan polls so in order to filter out these polls (due to the aforementioned questionable sample) a partisan suffix is added.

Other than those 86 polls, the remaining non-partisan polls really provide no new information:

I'm declaring the polling for this election over; I will so be posting out final projections.

Foreclosure Suppression Model:

Using foreclosure data from 2007, the only data set I could find, a suppression model was devised.

It's quite simple, but gives a good result; I take the number of foreclosures in 2007 and divide that by the 2004 popular vote total in a given state. The result represents the percentage of eligible voters [based on 2004] under foreclosure. The table below illustrates the ten states with the greatest percentage of foreclosures relative to the 2004 turnout:

Using this foreclosure percentage allows us to model a potential source of voter suppression. If we assume that all people currently under foreclosure are voting for Obama, we can simulate what would happen if these voters were suddenly prevented from voting in their home state. I've taken the current projection for Obama and subtracted from that the foreclosure percentage in the given state. The result of this calculation projects the following electoral outcome:

Bradley Effect Model:

In order to debunk the affect of the Bradley Effect, I'll try and simulate the potential outcome of the election should the Bradley Effect occur; we'll assume a historically significant racial offset as dictated by the Princeton University graph below:

Source: Princeton Election Consortium

Judging from the graph and more specifically the data points on the far right corresponding to the year 2006, it can be seen that the Bradley Effect was distributed between +5%, (meaning the African American candidate polled 5% lower in the final result than in the public polls leading up to the election) and -5%. For the purpose of this analysis we are trying to find the maximum degree to which the Bradley Effect could occur; in order to achieve this end we will use the maximum racial offset of +5% as prescribed in the 2006 data.

Incorporating the 5% into the projection algorithm presented the following changes:

Obama's Projection:

I calculated his initial projection and then subtracted 5% of it to arrive at his adjusted projection.

McCain's Projection:

I started by figuring out the number of undecided voters at the end of the initial projection; this is done by subtracting 100 from McCain's and Obama's initial projection. I then added this number to McCain's projection; to account for the adjustment made to Obama's total above, 5% of Obama's initial projection is then added to McCain's running sum to arrive at McCain's final adjusted projection.

Translating the pure formula into words results in a much more succinct correlation to the core principles behind the Bradley Effect. I assumed that 5% of Obama's support was racially tinged so I added it to McCain's total while subtracting it from Obama's. I then also assumed that of all the currently undecided voters, 5% are racist and as a result they will cast their vote for McCain. The result of this application can be seen in the map below:

Independent Shift Model:

While altering the outcome of our model, I wondered what would happen if all currently undecided voters suddenly moved to McCain en masse:

The next four model's are almost guaranteed to not happen, but since the means to make the maps exist, the maps corresponding to these unlikely models also exist.

Bradley Effect & Independent Shift Model:

I combined the results of the Independent Shift Model and the Bradley Effect Model by first applying the Bradley Effect scenario and then forcing the independents to support McCain:

Independent Shift & Foreclosure Suppression Model:

I simply subtracted the Foreclosure percentage from Obama and then shifted the independents over to McCain:

Foreclosure Suppression & Bradley Effect Model:

First I applied the Bradley Effect scenario, then the Foreclosure victims were subtracted from Obama's total:

Complete Combination Model:

The absolute worst case for Barack Obama. First the Bradley Effect is taken into account, then Foreclosed voters are suppressed, and then every currently undecided voter decides to support McCain:

The first three models lie much closer to the plane of realism and although their purpose is to depict a worst case scenario, they cannot overcome the ever increasing likelihood of an Obama victory.

I really thought there would be more polls tonight, but I guess I was wrong:

I'm curious to see if any polls are even released tomorrow; this may be our last update before either John McCain or Barack Obama becomes president elect. I'll be making my final projection tomorrow at some point before the polls close in Indiana ; the frequency of polling tomorrow, or perhaps the lack thereof will dictate the timing. If you're bored and can't sleep checkout the new Senate Projection map along with our projections:

I've got a few other things lined up, but I can't guarantee that I'll deliver on all of them, so I'm not going to tell you about any of them.

Busy day and there are still things to do. Here's the second batch of today's polls:

I've updated our Georgia graph to include the Liberatarian candidate Allen Buckley. The Senate election in Georgia will very likely end in a runoff, but I suppose anything is possible. In 2004, 3,301,875 votes were cast in Georgia and this year early voting has already accounted for 60.4% of the that total. This contest is far from decided:

I hope to accomplish three things tonight/morning; the first, another poll update after the late polls surface. I also want to create a Senate Projection map and possibily a probability graph, but we'll see on the latter. Finally I'd like to find a live election data feed to allow for a comparision between our projections and the actual results as they trickle in.

Public Policy Polling (D) and Reuters/Zogby decided that they needed to release their latest numbers at around midnight; and I was all over it. There was also a late straggler in Minnesota by SurveyUSA/KSTP that slipped through the crack earlier in the day.

Here's our late night edition:

Of the fourteen polls included in the roundup, the Republican party leads in just two; Indiana and the Minnesota Senate race. The results of any SurveyUSA poll taken in Minnesota have been suspect throughout the election cycle and this result is no different. Based on the cross tabs Barack Obama is garnering 50% of his support from voters over 50 and just 49% from voters aged 18-50. Can you say Halloween sample skew?

In fact SurveyUSA's results have been consistently awkward; to prove this I highlighted (with gray circles) SurveyUSA's polls on our Minnesota Senate graph below:

You'll notice that every single SurveyUSA poll result for Norm Coleman lies outside of our upper 95% confidence band with the exception of their latest result, but even then its an endpoint The Local Regression algorithm we use always fits the most recent data point within the specified confidence level, so this really isn't an exception, but rather a technicality. That said, I may be able to choose another pollster to illustrate a similar bias towards Franken.

The bottom line is that the Minnesota Senate race remains extremely tight and any additional polling is unlikely to sharpen the direction of this race.

You can listen to the debate at MPR.

7:02 CT: There will be no rules, but candidates should keep their answers concise. The candidates were then wheeled out in alphabetical order: Dean Barkley, Norm Coleman and Al Franken.

7:03 CT: Coleman starts first, the lawsuits will take the stage. Coleman stated that he has not received any gifts from family friends or allies. "The plaintiff to the lawsuit said that their was not basis for this claim." I highly doubt that if they took the time file a lawsuit.

7:04 CT:Norm was then asked if Al Franken has any connection to the lawsuit; Coleman did not answer the question but detoured to talk about how "Al Franken is targeting [Norm's] family."

7:05 CT: "Mr. Franken, do you have any connection to the lawsuit?" Franken responded, "None," and then paused for a moment. "The lawsuit is not about Norm Coleman's wife it's about Norm Coleman's sugar daddy. Norm Coleman cannot blame this on me; this is Norm Coleman's problem." Franken would not go so far to say that Coleman is corrupt, but called it a serious allegation.

7:08 CT: Barkley has had no part in this debate so far.

7:09 CT: Coleman: "There is a line to be drawn." The ad Norm Coleman continues to referrer to, the one where Al Franken attacks his wife follows. Nothing in the ad directly attacks his wife.

7:10 CT: Franken responded by stating that the lawsuit "is not about Norm Coleman's wife, this is about money."

7:13 CT: Now onto Dean Barkley, "now you know one of the reasons I'm running. This is probably the most negative campaign in Minnesota history...I may end up getting sued too."

7:14 CT: What will be the biggest issue to face America in the next six years? Dean went first and stated that he wants to turn America around and put it on a new direction. He also wants to clean up Washington; "Lets take the direct bribery out of politics."

7:16 CT: Norm Coleman next: "If you play to the partisan divide you accomplish nothing."

7:17 CT: Franken then stated that he wants to fix the economic situation; "Washington has not been working for Minnesota families." Franken then went on to list several issues that he believes he can fix.

7:18 CT: Question from a women in the audience. She asked each of the candidates to name one program they support that could be set aside while the nation gets its financial situation in order.

7:20 CT: Franken gets to respond first, "every item on my agenda has a return on my investment...that's what Clinton did."

7:21 CT: Barkley then got his typical line of "I've been working on the issues since 1992 when I first said it was broke...We need a four year spending cap to stop the bleeding; how are we going to pay for these things Al." Barkley then went on to question the massive military spending that contributes to our nations 11 trillion deficit.

7:22 CT: Coleman then stated the standard Kerry line: "You can't support something if you voted against it." The moderator then essentially said that's not the question, please answer it. "I would give up my raise," the crowd started booing and then he again got side tracked after briefly mentioning a spending freeze."

7:24 CT: Another question from the audience; the college question. How will you make it more affordable?

7:25 CT: Coleman essentially stated that he will continue his work with Pell grants. He was also sure to point out that his IRA has taken a hit and that he has two kids in college.

7:26 CT: Franken discussed his intention to give a $5,000 tax credit for kids seeking higher education. Franken then told a story about a student from Mankato State University who has to donate his plasma to pay for his education and he will still graduate with debt. He then went on to mention Obama.

7:28 CT: Onto Barkley, "If you want a guy who can make empty promises, I'm not your guy." He later went on to say that he will not pander, but will focus on the issues.

7:30 CT: Barkley gets to ask a question to Al Franken: "I know you support Obama, but can you name a position McCain supports that you also support."

7:31 CT: "I agree with Sarah Palin and John McCain...on their counter terrorism stance. You don't kill everybody. You make friends and then your friends tell you who to kill."

7:32 CT: Norm Coleman then asked Franken to "name three things Franken has done for Minnesota."

7:33 CT: Franken started but Coleman didn't like the fact that Al was actually answering the question. It got testy for a moment. "Franken went on to say that he worked with chemical dependency" patients. Franken then cited his participation in Operation Helmet, a joint venture with a Republican from Texas with the intention of providing additional helmets for the troops overseas. The third activity he cited was his numerous charitable events.

7:36 CT: Franken then asked Coleman "how many years should a politician have to debate until they can be a lobbyist."

7:37 CT: Coleman responded by saying "what would that accomplish" in reference to a lifetime ban.

7:38 CT: Franken responded by answering Norm's question; Franken roughly stated that Senators vote on certain bills because they believe it will benefit them later down the road.

7:39 CT: Barkley finally got another question. Would you cut off funding for the war? Barkley said "Yes," Coleman said "No" and then gave a story about mothers in his office, and then Franken said "No." Barkley stated that the question was not a direct question about "the Iraq war, but rather a hypothetical immoral war."

7:42 CT: Another question from the crowd; it sounded like a young boy. How would you help push alternative energy?

7:43 CT: Coleman went first and stated his support for the 2005 energy bill citing the fact that "85 senators voted for this bill and Franken said he would not."

7:44 CT: Barkley next; "I will not support tax breaks for Oil Companies." Barkley then cited the difference between the passive income stipulation included in the 2005 energy bill "that only benefits the wealthy."

7:45 CT: Franken gets the last word on the energy issue. "Norm Coleman basically said that if people agree blah blah blah [sic]." Then went onto say that if people actually agreed this election wouldn't look so bad for the Republicans.

7:46 CT: Another question from South Minneapolis. Is health care a privilege, right or responsibility?

7:47 CT: Barkley first; "16% of our GDP [Gross Domestic Product] is taken over by health care. That is much higher than other European countries." Barkley believes that health care should be a right and the ultimate solution is a single payer plan, although he has other solutions that are less drastic..

7:49 CT: Al then agreed with Barkley on the "right" to health care. Al stated that "I support Barack Obama's health care plan. If you like your health care now you can keep it, if you don't have health care it will be less expensive."

7:50 CT: Coleman followed citing "Franken's support for a single payer plan during his primary run with Cirisii...Under that plan you take the healthy kids and make them unhealthy because they aren't allowed to take Liptor and Prilosec." These are not drugs, kids are taking, or need to be taking under any plan.

7:51 CT: Yes and no question time. Is Medicare better than no Medicare?

7:52 CT: Coleman: "Yes", Barkley: "Yes', and Franken "Sort of."

7:53 CT: Raise taxes? Barkley: "No", Franken "No", Coleman "No."

7:54 CT: Is our current plan for children's health care adequate? Barkley "Yes", Coleman "Yes", Franken "we can fix it."

7:54 CT: Franken wants to do a quad-partisan bill with Bernie Sanders and Joe Lieberman.

7:54 CT: Approve a Supreme Court Justice who differs with your view on abortion.

7:54 CT: Barkley: "I would never ask that question". Coleman: would ask "if they're qualified." Franken: responded that he would not press the issue because "privacy emanates from the constitution."

7:55 CT: End Iraq war?

7:55 CT: Barkley "Yes", Franken "Yes". Coleman "Yes because we're not gonna be in Iraq in six years because gen. Petreaus has achieved such great success."

7:56 CT: Final statements, Franken first.

7:57 CT: "Elections are a time to hold our politicians responsible. If you believe George Bush was right 90% of the time I'm not your guy." Franken then did his "I ask for your vote" bit. Ending with his desire to end the war.

7:57 CT: Now Norm: "These are the most challenging times in my 37 years of public service. The question is not who do you blame, but rather who will fix it...We can bridge the partisan divide."

7:58 CT: Barkley concludes the night: "There are a lot of people who believe that Al or Norm are their worst nightmare...Do you think that Al and Norm will change the direction of Washington...Send Washington a message by sending Dean Barkley to the Senate."

Quick update for now, the Minnesota Senate debates starts in an hour and a half and I've got things to do before then. Here are the polls so far today. I'm still convinced a few will trickle in later:

Here's just a quick thought; judging from our Senate Projections its pretty much a given that the Democrats will end up with at least 58 members in their caucus. But acquiring the final two needed to push them to 60 seems to be a stepper climb. The two best opportunities seem to be Georgia and Minnesota. I discussed the Minnesota race earlier but the Georgia race is also deserving of attention; or is it? I said it yesterday, but I'll say it again: If the Democrats are serious about reaching 60, which I assume they are, it would be a wise move to pseudo concede Georgia, and then later fight in the runoff.

Be sure to comeback later and checkout out our recap of the final Minnesota Senate debate.

The three way Senate race in Minnesota took another bizarre turn this weekend, leaving all three candidates with a viable shot at becoming Minnesota's next Senator. Before I delve into the details, lets take a look at the current state of the race:

Donate: Al Franken (D), Dean Barkley (I), Norm Coleman (R)

And our Senate to Presidential correlations:

           VFA Projection [10/18]    VFA Projection [11/2]
            Dem    Rep    Dem-Rep    Dem    Rep    Dem-Rep
President  49.98  43.71     6.27    53.61  40.67    12.94
Senate     38.61  40.15    -1.54    40.90  39.32     1.58
Pres-Sen   11.37   3.56     7.81    12.71  -1.35    14.06

Norm Coleman is now polling slightly worse than John McCain, while Franken continues to lag behind the levels established by Obama. If you take the numbers above at face value, Coleman is losing 1.35% of McCain's vote and Barkley appears to be making up the difference between Franken and Obama. If these trends were to continue, Barkley would likely be eliminated but recent events have left the election wide open.

Heading back to last Monday (10/27), Paul McKim, founder and former CEO of Houston-based Deep Marine Technology, filled a lawsuit alleging that an ally funneled $75,000 to the Colemans. Reading directly from the Star Tribune article on the subject, "A Texas businessman has filed a lawsuit alleging that Minnesota multimillionaire Nasser Kazeminy used his Houston marine company to funnel $75,000 to Sen. Norm Coleman last year via a Minneapolis insurance company that employs the senator's wife." The Coleman campaign immediately went to work trying to kill the story, but the DFL may have gotten the best of him:

This suit was later withdrawn, but it didn't end there.

The Coleman campaign responded with a lawsuit of their own; on Thursday October 30th, Norm Coleman filed a defamation lawsuit against Al Franken. The Coleman campaign concluded that "Mr. Franken ha[d] chosen to push the lines of believability far beyond the bounds of the truth." The "truth" Coleman is questioning relates to an ad released by Franken citing Norm Coleman as the "fourth most corrupt senator in Washington." Franken responded by saying this: "Our ads are factual and true, even if Norm Coleman doesn't like being held accountable for his conduct. Every time someone tries to hold Norm Coleman accountable, he runs to court to try to weasel his way out of it. And none of the three prior times he's done this has he been successful, and he won't be this time, either." To be fair the report Franken appears to be citing does not explicitly rank corrupt Senators, but they did place Norm Coleman, along with three other Republican senators in the "Dishonorable Mention" category.

Here's where it gets interesting. The lawsuit filled on Monday by Paul McKim was refiled again Thursday evening, just hours after Coleman's defamation charge. The Coleman campaign responded by calling the suit "false and defamatory...My wife has been devastated by this. She's angry -- and she has a right to be -- and so am I. So Coleman's wife can be angry but Al can't? Seems semi-hypocritical but that's not the most interesting nugget from the quote. The use of "defamatory" was likely intended to cascade into the suit filed against Franken, and it may have worked. If you do a google search for "Coleman Franken Defamatory" the results are roughly split between the two cases, but the issue still lingers.

The Coleman campaign tried to kill the issue again Friday night by saying the premise of the suit was "baseless and [contained] false claims ... being used to influence the outcome of the election.'' Still sensing the urgency Coleman tried to link the funneling charge to Franken; "If my opponents have any shred of decency left in this campaign -- stop attacking my family." Franken's campaign, specifically Colleen Murray released a statement rebutting Coleman's charge; Franken had nothing to do with the lawsuit. In fact the plaintiff justified the timing of the suit by stating that it has "absolutely nothing to do with politics." In any case the Coleman campaign released a new ad overtly linking Franken to the money funneling suit:

If the Coleman campaign is going to make a connection that just blatantly doesn't exist they should at least be consistent. "Yesterday, Coleman's statement segued from the [Star Tribune] receiving a pre-filing 'copy of these false allegations' to Franken 'running vicious, untrue attacks against me' on filing day." I also thought Norm was done with negative ads, but apparently it's not politically convenient anymore.

Al Franken and Norm Coleman clearly have a fight on their hands, but Barkley remains unscathed. Within the entire body of this article Barkley's name was not mentioned. When the last debate occurs later today, Barkley can reside above the fray while Franken and Coleman undoubtedly duke it out. If Barkley strikes the right cord he may very well be able to peel off 5% from each of his opponents and sneak away with a victory. We likely won't know the result of the whole lawsuit situation or the debate until the actual votes are counted.

This is pretty much all you need to know:

Yes, Obama is now winning Indiana; by .02%. While this is likely insignificant, the election could very well be over when the first phase of polls in Indiana close at 5 PM CT; and the exit polls are revealed. If Obama's within 5% or so, the Evansville and Gary regions may very well push him over the edge when their polling closes an hour later. But even if Obama does not in fact win Indiana, the result should be very telling of where the election is headed.

Just eleven pro-Republican polls today, definitely not a good sign for the GOP. Here were today's polls:

If the Dems want to reach 60, there are really only three viable options; they must win two of the following: Kentucky (4.58%), Minnesota (20.4%) and Georgia (7.72). Georgia law requires a runoff election if the winner recieves less than 50% of the vote; this could legitimately happen, but more on this possibility Monday afternoon. If the Democrats are serious about reaching 60, which I assume they are, it would be a wise move to pseudo concede Georgia, and then later fight in the runoff. What an epic battle that could be.