Why The Tennessee GOP SEC Has A Duty To Invalidate The House District 33 Primary

Image Credit: Tennessee Republican Party & Canva

Submitted by Norman Bobo –

The TN GOP SEC members must make a difficult decision this Saturday that affects thousands of Republicans in Anderson County and has repercussions across the state.

Invalidating a primary is serious business. It should not be done lightly.   

On the flip side, allowing an invalid primary result to stand would be just as serious.  Neither should that be done lightly.

I want to first complain that the GOP SEC has set aside just ONE HOUR on Saturday to make this momentous decision. They have provided each candidate only 7 minutes to present their case (plus a 2 minute rebuttal by Ragan). That is ridiculous. Further, the meeting allows only one minute for each speaker in the public comment period. That’s outrageous. MUCH more time should be allocated for all of the items on the agenda. The meeting should be at least 2 hours, if not 3 or 4 hours. Ok, I’ll hop off my soap box now.

In this House District 33 race, Mr. Ragan, the challenger, provides three sets of information, each of which demonstrate that this primary was invalid because of illegal crossover voting.   [Note: I have not had the time to independently verify Mr. Ragan’s numbers.  I’m accepting them as accurate. The SEC members should validate them.]

I want to dive into the numbers quickly. But before I do, I need to point out some important discussions below regarding the imprecise nature of this decision, HOW this decision should be made and why the information I’m about to discuss should be used to determine this primary was invalid.  I urge everyone to read beyond the conclusions below to better understand the nuances of this decision.

• What is the deciding factor in whether a primary should be declared invalid?

• Should we just count the number of 3/4 and 4/4 Democrats to determine if the primary is invalid?

• The information below uses “estimates”, “factors” and “percentages”. Should we not be more precise in making this determination?

• When evaluating the information, should SEC members lean one way or the other when deciding how many Democrats or independents voted in this primary?

• What does the Recent Williamson County Crossover Voting Analysis Tell Us About Rampant Crossover Voting?   Should ANY Close Republican Primary Should Be Challenged and Found Invalid?

• Should the SEC use the criteria that voters must be “Bona Fide” Republicans according to the TN GOP Bylaws?

• Once the Primary is Invalidated, How Should the House District 33 Nominee Be Determined?

• How to Fix Crossover Voting Completely

The Data Which Clearly Demonstrates That This Primary Was Invalid

There are three sets of information discussed below, each of which leads to an estimated number of illegal votes in this primary.   To be clear, SEC members only have to find that ONE of these estimates is compelling enough to determine that this primary is invalid.   I find that ALL THREE are compelling.

1. There were 1,135 voters who do not qualify as “bona fide” Republicans.  

a. Note:   There is a question being bantered around about whether a voter must be “bona fide” according to the definition in the TN GOP Bylaws.  It is clear to me that a voter does not need to be bona fide as defined in the TN GOP Bylaws in order to vote in a primary.  My reasoning is discussed in “Should the SEC use the criteria that voters must be a “Bona Fide” Republicans according to the TN GOP Bylaws?” later in this commentary.   Having said that, TN GOP SEC members may still choose to use the bona fide status as a way of determining whether voters have allegiance to the Republican party or not, a requirement to participate in a party’s primary.  If voters do not have allegiance, they are voting illegally.   In the final analysis, there is enough compelling information to invalidate this primary without strictly requiring the voters to be “bona fide” using the definition in the TN GOP Bylaws and the following information does not depend on the voters being “bona fide”, though some SEC members may choose to require that voters be bona fide.

b. Whether you agree voters should be bona fide or not, the high number of voters who are not bona fide Republicans still informs the decision, especially when combined with other information provided by Mr. Ragan.

c. A commonsense conclusion (from the fact that so many voters are not bona fide Republicans) is that some percentage of them were Democrats or independents illegally crossing over.   The question is what percentage?

d. 75% of that number would be 851 voters.

e. 50% of that number would be 567 voters.

f. 25% of that number would be 283 voters.

g. All three of those percentages are higher than the 258-vote margin of victory.   But do we have any basis for applying any of those three percentages or should a different percentage be applied?

h. Mr. Ragan points out that only 6 of these voters had ever attended an Anderson County GOP event over the last three years.   Think about that.   I would expect that most of the 1,135 voters who supposedly have allegiance to the Republican Party – and who are active enough to participate in a Republican primary – would have shown up to at least one local GOP event over a 3-year period.  But that is not the case.

i. That shocking number of only 6 voters demonstrates that the vast majority of these voters have no true allegiance to the Republican Party.   (Note this is not about trying to define “active” – it is about whether their behavior demonstrates their allegiance).

j. When you consider the low level of activity among these 1,135 voters, what percentage do you think have no allegiance to the Republican Party and are illegal crossover voters?   I would place it at 50% or more.   But surely it must be 25% or higher – more than enough to find that the primary should be declared invalid.

2. There were 619 voters who had never voted in a Republican primary.   

a. Some of these are new Republicans.   Welcome to the party!  We look forward to you attending events and donating to the party and our candidates.

b. While we welcome the new members, there is simply no way that they are all new Republicans.

c. So, the question is how many of them are new Republicans?   75%?  50%?  25%? 

d. If only 50% of those voters are actually Democrats or independents illegally crossing over, it would mean 309 of them were illegal votes – more than the 258-vote margin of victory.

e. Now, consider again that these 619 voters are a subset of the 1,135 voters discussed above in which only 6 of the 1,135 voters had ever attended an Anderson County GOP event in the last three years.  I don’t have the information about whether the 6 voters in the 1,135 voters above are in this set of 619 voters, but let’s be conservative and assume that all 6 are in this set of 619 voters.

f. It is still shocking that among 619 voters only 6 had attended a Republican event over three years.  Again, you would expect that if they were new Republicans, MOST of them would have attended at least one event in the last three years.  If, God forbid, I chose to become a new Democrat, I would probably start attending Democrat events – at least one event over a three-year period.  Remember that primary voters are in an elite group of voters.   They pay attention MUCH more than other voters and are generally more active.  That fact that only 6 attended county GOP events tells us that a very high percentage of these voters do not truly have allegiance to the Republican party.

g. So, what percentage of these 619 voters would you say we could safely assume have no allegiance to the Republican Party?   I would say 75% or higher.   But all it takes is 42% to exceed the 258-vote margin of victory.

3. 170 of the primary voters are confirmed bona fide Democrats.

a. This is the most damning evidence of all.   

b. Importantly, we must assume that they are all Democrats crossing over.  I discuss why later in this commentary in the question “When evaluating the information, should we lean one way or the other regarding how many Democrats or independents voted in this primary?”

c. But these are not all of the voters who crossed over.   Which voters are NOT included in this count of 170?

• i. – The 170 number does NOT include Democrats who flip back and forth so frequently that they fail to meet the definition of “bona fide Democrat”.
• ii – The 170 number does NOT include independents who flip back and forth and who illegally voted in this primary.

d. To find this primary invalid, we must find at least another 88 voters who crossed over in addition to these 170 Democrats.

e. 88 is only 1.6% of the remaining 5,498 voters in this primary.   If you have examined crossover voting statistics in Tennessee (including the recently completed and very compelling Williamson County crossover voting analysis discussed below) you will know that crossover voting is AT LEAST 2% of the vote in every Republican primary.  In fact, it averages around 5% and can go as high as 8% to 10% in some primaries depending on which partisan races appear on a ballot.

f. So the question is what percentage of crossover voting likely happened in this primary?   There are two factors which inform the percentage that should be applied:

• i. – In this primary, there were no high-profile Democrat primary races on the ballot.  (The United States Senate Democrat primary was on the ballot, but Gloria Johnson was a shoo-in statewide and thus there was no need for Anderson County Democrats and independents to pull a Democrat primary ballot to vote for her).     The point is that this particular ballot was ripe for crossover voting by Democrats and independents.   In situations where there are no consequential Democrat (or independent) candidates on the ballot, crossover voting is typically on the high end of the percentages – 8% or more.
• ii. -We must again consider that only 6 of the 1,135 voters discussed above had participated in a Republican event in the last three years. This also indicates a low level of Republican allegiance and thus a high level of crossover voting.

g. If we apply a VERY LOW estimate of 4% crossover voting percentage to the remaining 5,498 voters, we get another 218 voters who crossed over, in addition to the 170 confirmed Democrats, which combined is WAY more than enough to exceed the 258-vote margin of victory.  But surely the percentage is higher than 4%!

h. The commonsense view of these numbers is that AT LEAST 258 Democrats and independents illegally voted in this election.

Again, all it takes is for ONE of the three evaluations above to be accurate to invalidate the primary. (It does not take all three.)

My assessment is that all three of them are compelling because the resulting estimates of illegal voter counts are significantly higher than the margin of victory. Though this information does not identify specific voters, it does identify ENOUGH voters with confidence.

Though most rank-and-file Republicans may not “run the numbers” like I have done above, they innately know what the numbers are telling them. They are applying their common sense and are confident that this primary was stolen.

[ As an aside, I hear rumors that even more data is being gathered to provide even further proof that Democrats and independents made the final choice on the winner in this primary.   In my assessment, no further proof is needed, but further proof would be welcomed. ]

Because the information above is getting distributed, hundreds of Republicans from across the state are emailing their SEC members (or all of the SEC members) asking that they find that this primary was invalid. If you have not informed your SEC committeeman and committeewoman member regarding how they should vote on Saturday, I encourage you to do so!   You can find their email addresses HERE.

SEC members, I suggest that you not be frustrated by your inboxes being blown up with people asking you to invalidate this primary, but rather embrace the judgment of your fellow Republicans and factor it into your own judgment concerning the available information.

These Republicans innately understand that this decision is not precise and that common sense judgment must be applied.   These Republicans, and particularly the leading Republicans in Anderson County, are telling you that, based on the information they see, their judgment is that this election was clearly stolen by Democrats and independents voting in this primary.  That is how they are interpreting the data and how they are encouraging you to interpret the data.

If you are confident that you can dispel the doubts of all of these Republicans that the result of this close primary was not affected by illegal voting, then you should uphold the primary and defend that decision to the Anderson County Republicans and all Republicans across the state — including the Republicans in your district at your next election.

But if you cannot confidently put to rest the doubts of your fellow Republicans who are calling for justice, you will be doing a strong disservice to the true Anderson County Republicans who voted in this primary, and true Republicans across the state, by allowing the primary result to stand.

In the end, whether you choose to invalidate the primary or not, I highly recommend that you do everything in your power to redirect any Republican voter frustration over the final decision to the General Assembly. They put the voters, the candidates and you in this predicament.  You should lay this at their feet (though you are still responsible for your vote on Saturday).

Further, I strongly recommend that a communication be sent to the members of the Anderson County Republican Party explaining your decision and why you decided the way you did. They are especially owed this explanation.

Answers to Key Questions

1. What is the deciding factor in whether a primary should be declared invalid?

The answer is simple, but still worth some discussion.  

The answer is that if more Democrats and Independents (more precisely, voters without allegiance to the Republican Party) voted in the primary than the margin of victory, the primary is invalid.  Period. And the SEC should simply recognize that truth by voting to invalidate it.

If the number of Democrats and independents who voted in the election was lower than the margin of victory, then upholding the election would mean that NO Republicans are affected by the illegal votes.   You might have had a different margin of victory, but no Republican would be hurt by the result.

However, if the number of Democrats and independents who voted in the election was HIGHER than the margin of victory, then upholding the election would mean that EVERY Republican voter in that primary was damaged.

Why is that? Because you cannot not know for whom the Democrats and independents voted. We honor secret balloting in our Republican primaries. Because of that, there is no way to separate out the invalid votes from the valid votes. Thus, ALL of the Republican ballots have been tainted.

In past challenges, I have heard people suggest we should assume some percentage of the Democrats and independents voted for one candidate and the remaining voters for the other candidate.   This should never be done.  In order to fully protect our Republican primaries from illegal crossover voting, and to be fair to the challenger, the SEC members MUST assume that ALL of the invalid votes went to the winner.

Thus, there is only one question the SEC needs to decide:  “Did more Democrats and independents cast illegal votes in this primary than the margin of victory.”

If the answer is “No”, the primary should be upheld.

If the answer to that question is “Yes”, then the primary is invalid and the SEC has a duty to protect the Republican primary voters by invalidating the election and choosing an alternative method of selecting the nominee.

2. Should we just count the number of 3/4 and 4/4 Democrats to determine if the primary is invalid?

In past primary challenges, SEC members have (tried) to identify how many Democrat voters participated in a primary and used only that criterion to determine if the primary was invalid.  

Most often, the deciding criterion that a voter was a Democrat is their voting record, usually including those who pulled Democrat ballots in 4 of the last 4 statewide primaries or 3 out of the last 4 statewide primaries.

There are three reasons this criterion should NOT be used:

Reason 1:  It undercounts the amount of crossover voting.

Some Democrats flip back and forth so many times that they are never “4 out of 4” nor even “3 out of 4”.   Thus they are not counted in the “4 out of 4” or “3 out of 4” criteria.

This criterion also excludes independents illegally crossing over, which is just as immoral and illegal as a Democrat crossing over.   

Limiting the number of illegal votes to only those voters who voted as a Democrats 4/4 or 3/4 times significantly underestimates the actual level of illegal crossover voting.  By one SEC member’s estimate, only approximately 70 Democrats would be counted.   That is a ridiculously low estimate of the crossover voting in this primary.

The more accurate way to estimate crossover voting is to count the number of voters who flip parties back and forth – “party flips”.

If a voter changes parties only once, this can be assumed to be a genuine change in party. This change should not be counted as an illegal vote, unless there is other information about the voter which tells us otherwise.   

Note:  some would say that we need to wait until the next election to see if they flip back to determine if it was an illegal vote. That is true, but ONLY if they flipped just one time!   If they have a history of flips, then EVERY vote, including the most recent one, should be assumed to be illegal.

If a voter flips twice, particularly twice in a row, it is likely NOT a genuine party change either time, but it may be in a few cases. You should likely count most of these votes as illegal votes – including the most recent vote – and some smaller percentage of them as genuine party flips.

If a voter flips three times or more, they are definitely gaming the system with illegal votes, and you should count ALL of their votes as illegal votes.

When you do this type of analysis, which I just did for the Williamson County voter file discussed below, you will see that the amount of crossover voting is much higher than counting only 3/4 and 4/4 Democrat voters.

Reason 2:  It is an arbitrary criterion.

Why should it be only 4/4 or 3/4?    What’s so special about these numbers?   Why not include 2 out of 4?   Or 3 out of 5?    In fact, it is the 2/4 and 3/5 Democrats who are most likely flipping back and forth with illegal votes and are not being counted!   This is discussed more in the next question below.

Reason 3:  It ignores other information.

As you see from the three sets of information at the beginning of this commentary, there is other, higher-level, statistical information that can and should be applied to a decision on whether a primary was invalid.   

This is not just about counting specific individual voters with one limited set of criteria, but rather estimating with some level of confidence how many voters crossed over based on a variety of factors.  This IS NOT and CANNOT be a precise analysis!  You can have confidence without precision!  This observation is expanded upon in the next question and answer.

In the end, the approach of trying to identify specific Democrats sets the bar too high.  Using this criterion:

• Has discouraged candidates from challenging primaries and the TN GOP SEC from invalidating primaries that rightly SHOULD have been invalidated.

• Will lead to even MORE crossover voting.

• Is the logic the Democrats in our state hope the SEC will apply – because it will let them get away with more treachery and allow more illegally-elected moderates into our General Assembly.

• Will effectively embolden the moderates in the General Assembly to continue to kill party registration bills.

I have never agreed with the line of thinking that counting Democrats by their Democrat primary voting records is the way to make this decision. And I believe most Republicans don’t agree with it either.

3. The information above uses “estimates”, “factors” and “percentages”. Should we not be more precise in making this determination?  

I understand the desire to have precision in this very momentous decision. But the truth is that precision is IMPOSSIBLE, particularly given the short timeframe in which this decision must be made.

Probably the most precise way to make this decision is to thoroughly review the backgrounds of all 5,668 voters who participated in this primary, looking at, for instance, the yard signs they have in their yards, the donations they have made to political candidates, their social media posts, their attendance at partisan political events, and many other factors. But obtaining that level of detail is impossible in this short timeframe.

Thus, the decision has to be made based on some form of statistical, summary level data.   The only question is which statistical, summary level data should be used.   SEC members, you are making judgment calls, not black and white decisions, when you accept or reject various statistics regarding how many illegal votes likely occurred.

As an example, let’s consider the approach of counting only 4/4 or 3/4 Democrats.  This criterion is a statistic just like other statistics, all of which require judgment calls.   What judgment calls are being made by using this criterion?

• It is a judgment call to say that voting records are the only information we should use and that we should ignore all other information about the individual voters, or the summary statistics discussed at the top of this commentary.

• It is another judgment call to use only 3/4 and 4/4 Democrats. Why not only 4/4?  Why not also include 2/4? or 3/5? or even 2/5?   Some people believe that if you voted Democrat even once in the last 4 statewide primaries, you are likely still a Democrat.  Why not include them?

• It is also a judgment call that illegal votes from independents did not affect this election.

As discussed earlier, I suspect many Republicans strongly disagree with the assumptions behind using this criterion for making the decision.  It just ignores too much other compelling information.

In the end, SEC members should examine ALL available information and make commonsense judgments about the information – even if that information that is imprecise.  All you need is confidence that the information is directionally accurate and the numbers large enough to give you confidence in your decision.

Let me draw a parallel. The criteria should NOT be “beyond a reasonable doubt” as is used in a criminal trial. Rather, the criteria that should be used is a “preponderance of evidence” as is used in a civil court case. This decision is more akin to a civil case than a criminal case.

(Note:  I debated about discussing these parallels because some readers may imply too much from them. To be clear, the TN GOP  Bylaws explicitly state that courtroom procedures do NOT apply to this process and I am not implying that they do by drawing the parallels of standards of proof.)

In the end, the question is NOT “Did enough Democrats vote in this election to affect the results.”   

The question is (and always should have been) “Was it likely that enough Democrats and independents voted in this election to affect the results”.

SEC members, you need to use the best information available to you. I assert that the information provided at the start of this commentary is the information you should use – because it is compelling and the number of illegal votes is large enough to have confidence that the primary was decided by illegal voters.

4. “When evaluating the information, should SEC members lean one way or the other when deciding how many Democrats or independents voted in this primary?”

As stated at the start of this commentary, invalidating a primary is serious business.   It should not be done lightly.  On the flip side, allowing an invalid primary result to stand would be just as serious, if not more serious.  Neither should that be done lightly.

Put another way, the “default” position of SEC members should NOT be to assume the primary was valid and then have to be pushed hard to be moved off of that position.    The fact is that a candidate has challenged the result. What’s to say the challenger is wrong? Why should you automatically assume they are wrong?

The default position of SEC members should be neutral (particularly with a primary decided by so few voters).  You should assume neither that the primary was valid nor invalid.  From this neutral position, you should then be swayed one way or the other by the available information.

The fact is that crossover voting happens. Everyone knows it. The Democrats (particularly the TEA) have actively promoted Democrats voting in Republican primaries all across this state for over a decade.

One of the most important jobs of the State Executive Committee is to act as the primary board to protect the sanctity of Republican primaries.   

In my view, in order to do that, and with the knowledge that Democrats actively and aggressively participate in our primaries, you must likewise be aggressive in protecting our primaries.    

Thus, in evaluating whether a primary was tainted by illegal voting, if there is any reasonable way that you can conclude that a primary voter was a Democrat or independent, you must find that they ARE a Democrat or independent.

In other words, you MUST resist illegal voting in our primaries by leaning towards protecting our primaries as you evaluate the information available to you.

This means you must give the benefit of the doubt to the primary challenger if the information they provide is even close to compelling.  And you should NOT reject that information just because it may be imprecise. You only need to find that it is compelling and the numbers large enough to warrant finding the primary is invalid.

5. What does the Recent Williamson County Crossover Voting Analysis Tell Us About Rampant Crossover Voting?   Should ANY Close Republican Primary Should Be Challenged and Found Invalid?

I recently completed a thorough analysis of crossover voting which examined the last 14 state and county primaries in Williamson County (as far back as data is maintained).

That analysis identified 1,435 primary voters who flipped parties back and forth across state and county primaries three or more times in their Williamson County voter history.

Of those 1,435 voters, here is the number of times they flipped parties:

9 Party Flips:         3 voters

8 Party Flips:       12 voters

7 Party Flips:       34 voters

6 Party Flips:       82 voters

5 Party Flips:     178 voters

4 Party Flips:     382 voters

3 Party Flips:     744 voters

                Total:  1,435 voters

To give you some idea of the severity, there were voters who flipped parties:

• 9 times across 14 primaries.
• 6 times across 8 sequential elections.
• 5 times across 6 sequential elections.
• 4 times across 5 sequential elections.

The percentage of voters flipping parties varied from 2.1% up to 8.3%. Let that sink in.  At least 2.1% and up to 8.3%.

This proves that crossover voting is RAMPANT and is NOT a minor issue. SEC members, you should know with certainty that ANY close Republican primary elections is absolutely stolen by illegal voting.

Because of this analysis, and many other prior analyses, I call on ANY candidate who loses a Republican primary in a majority Republican district by less than 5% of the vote to automatically challenge that primary.   And I call on the SEC to invalidate any such primary – automatically, without debate.  That may seem extreme, but that is the true reality we face with crossover voting in our state.  Until the General Assembly takes action to stop crossover voting, the GOP SEC must protect our primaries!!

[ By the way, this does not mean that if the margin of victory is more than 5%, a primary should not be challenged.   In some counties, the numbers may easily support a challenge where the margin of victory is more than 5%. ]

6. Should the SEC use the criteria that voters must be “Bona Fide” Republicans according to the TN GOP Bylaws?

In his original complaint, Mr. Ragan makes the assertion that voters must be “bona fide” using the definition found in the TN GOP Bylaws. That definition has two parts. The first part is that the Republican must have voted in 3 out of the last 4 statewide Republican primaries. The second part is that the person is “active” (see the bylaws for the details on that).

A careful reading of the TN GOP Bylaws show that a person is required to be “bona fide” in three circumstances:   1) to be a candidate, 2) to challenge the bona fides of a candidate and 3) to participate in a convention (caucus).    

Shockingly, the bylaws are actually silent regarding whether a voter needs to be bona fide to participate in a Republican primary.   

In my view, Mr. Ragan incorrectly assumes that the term “bona fide” as used in the law is the same “bona fide” as used in the party rules for 1) candidates, 2) those who would challenge candidates and 3) those who want to participate in a convention (caucus).  They have different meanings in the two different contexts.

How do we know the same term has two different meanings, one meaning in the T.C.A. and a different meaning in the GOP bylaws?   

The answer is simple: If Mr. Ragan’s interpretation that the T.C.A. refers to the “bona fide” definition in the GOP bylaws, a new Republican voter could never meet the three-out-of-four requirement – because they could never vote as a Republican in the first place in order to develop that bona fide Republican voting record.   This interpretation creates an impossible chicken-and-egg scenario. Thus, that interpretation must be rejected.

All of this discussion on “bona fides” does highlight the extremely confusing set of laws and rules regarding who is a “Republican”.  This must be fixed – by both the General Assembly and the SEC.  But any possible future fixes do not change the current situation regarding the House District 33 primary.

I will point out that even though voters do not have to be strictly bona fide to vote in a primary, they must still have “allegiance” to the party according to the law.  Their voting records and their activity levels are still strong indicators of that allegiance or lack of allegiance and should still inform the decision about whether they are valid Republican primary voters.

7. Once the Primary is Invalidated, How Should the House District 33 Nominee Be Determined?

I am not in the weeds regarding the various options, but have these thoughts:

– 1. The SEC should NOT select the next representative from District 33.

• a. This should be a local decision, not one made by people outside of that district.
• b. Local Republicans better understand their members and the candidates.

-2. Ideally, the AC GOP should run a convention.

• a. If time allows, the local party should execute a convention to select the nominee. – Conventions typically have procedures which are highly effective in preventing Democrats from participating.  This ensures only Republicans are electing Republicans.
• b. The convention should be limited to only the two candidates who ran in the primary.
• c. The participants in the convention should be limited to only those party members who were registered in time to vote in the August 1 primary.
• d. If a convention is used, I highly recommend that the AC GOP follow the procedures used in the highly successful Williamson County District 11 convention.   Note that if the Anderson County party begins planning before Saturday, a convention could be held within a week of the Saturday decision, particularly if they just copy the procedures of the recently completed Williamson County convention.
• e. Note: As I have been editing this document and speaking with others about these issues, I have received information indicating that a convention is probably not possible in the limited time in which a nominee must be named in order to meet the ballot printing deadlines.

-3. The Next Best Option: Poll the Anderson County GOP Members

• a. If a convention is not possible, it may be possible for the Anderson County Republican party to poll its members in District 33.
• b. This process should be limited to only the two candidates who ran in the primary.
• c. Only bona fide members should be polled.
• d. It is likely that a web-based voting solution can be found which would allow the members to vote securely and anonymously. – One such solution is:  https://electionrunner.com/  (disclaimer:  I found this solution using a quick web search … I am not confirming that it will meet the need).
• e. Manually poll the members: • The polling may be done by phone, with representatives of each candidate monitoring the calls on speaker phone. • Of course, this would not allow for anonymous voting.
• f. Note: from my discussions with others, it appears this option may also not be possible in the time available.

-4. The Last Resort: The Anderson County CEC Picks the Nominee

• a. If the other options are not possible, then the CEC, as the elected leaders of the Anderson County Republican Party should select the nominee (from the two candidates only) by a majority vote.    They should, however, take in as much input as possible from their members prior to that vote.  Though not as good as holding a convention or polling the members, having the local CEC select the nominee is MUCH better than having the SEC select the nominee.

8. How to Fix Crossover Voting Completely

Those of us who have studied this issue deeply believe that “party registration” will go a long way to fixing crossover voting. But implementing party registration by itself will not completely fix the issue. We must go one step further.

Defining Party Registration

If you go by the content of the various party registration bills which have been proposed (and defeated), party registration, generally, means:

1. Each voter decides the party to which they want to be affiliated at the following times:

• a. When they first register to vote.
• b. When they update their voter registration (name changes, moves, etc.)
• c. If they are an existing registered voter, when they participate in the first primary election (county or state) following the implementation of the law, they will declare their party registration by the ballot they select in that primary.

2. The voter can switch their party registration at any time by completing a new registration form.

• a. However, in order to participate in that party’s next primary, they must switch parties within the voter registration deadline for that primary, which is, generally, 30 days prior to the primary.

3. A voter can only vote in the primaries (county and state) of the party to which they have registered prior to the voter registration deadline.

The main way that party registration stops crossover voting is to stop the “last minute” crossover voters. Most voters do not pay attention to an election until just 2 weeks before the election, or even closer. As the election draws near, they typically turn to others who they know are paying closer attention to ask how they should vote.  Many times, this happens on the day that they actually vote – even on the way to vote!   It is then that they are told by others which ballot to request and for whom they should vote.  My best guess is that 75% or more of crossover voting happens at the last minute like this.

Only the most serious partisans pay deep attention to the races on a ballot more than two weeks in advance, and even fewer pay attention more than 30 days out, which would be the party registration deadline.   Yes, some serious partisans will switch their registration more than 30 days in advance to crossover vote.   This also should be addressed, but so far, as much as I am aware, none of the bills attempt to do anything to address this.   The focus has been on just getting party registration passed in the first place!

The thinking behind party registration is that most people will select their party (or no party at all) and would stay with that choice for many years, or even a lifetime, until they are convinced to switch parties, which would be infrequent. Any such changes would likely stick for a long period of time. 

The second way that party registration would reduce crossover voting is that it becomes a more public declaration of your party affiliation. Today, people’s selection of partisan ballots in primaries is how we determine someone’s party affiliation.  But that is a fairly arcane way to assign a party to a voter. If your party affiliation appears on your voter card and on public lists, that would be MUCH more transparent and “real” to most voters. Most people will want that party affiliation to speak about who they are.

For the reasons above, party registration should be implemented immediately, at least as a first phase.

By the way, I am STRONGLY opposed to watered-down versions of party registration which would allow independents to vote in partisan primaries. Allowing an independent to vote in a Republican primary is, morally, the same as allowing a Democrat to vote in a Republican primary. Neither has allegiance to the party principles. Why should they be picking the party nominees?

There is one major hole in party registration. As discussed above, some highly partisan people in districts where they are a small minority, may purposefully (and permanently) choose to register with a party to which they do not have allegiance. (This is why we currently have some Democrats with better Republican primary voting records than Republicans!)  It is possible that over time the number of people who do this could grow and become a real concern.

There needs to be some form of validation that a person who is registering as a Republican truly is a Republican. It is because of this lack of validation in the party registration bills proposed so far that some Republicans have been against party registration. They state that, effectively, the party gives up control of defining who can participate in their primaries.

But there is a way to fix that.

The law could provide a way for the parties to confirm who is a member and thus who gets to register as a member of their party. There are several ways to do that:

1. Require that a person obtain the permission of the state or county party prior to registering as a Republican.  

• a. In other words, at the time the voter registers (or re-registers), they would need to provide their voter registration form AND some kind of signed form from the local or state party allowing them to register with that party.
• b. In such a case, each party could provide a deadline by which requests for confirmations must be made prior to the next primary, to give the parties time to confirm the party registrations in time for the voter to register by the voter registration deadline.

2. Allow voters to select their party without a form but make all party declarations “provisional” until confirmed by the state or local party.  

• a. If the political party fails to confirm them in time for the election, then they could be either automatically removed or automatically confirmed.  
• b. It could be that all party registrations, say, more than 60 days out from an election would be automatically confirmed while, say, any new party registrations within 30 days would be automatically rejected until after the primary if they remain unconfirmed by the local or state party.

In both of the techniques above, each political party would still be able to define its own rules by which a person can be a member and their party registration confirmed.   Note that the rules for becoming a new member of a party cannot be based on voting records! 

How would a person become a new registered member until they have been allowed to vote as a member?

Note that being a “registered” member of a party could have a different meaning than being a “bona fide” member of the party.  A “registered” member could be a lower level of membership and “bona fide” could be a higher level of membership. For instance, it may take several election cycles for a registered member to become bona fide based on their voting history (if voting history remains a criterion to be bona fide.).  Also, it may take a few years for donations to opposing parties or candidates to “roll off” a person’s record and thus allow them to become bona fide.

I encourage the SEC to pass a resolution calling on the General Assembly to not only pass party registration, but to also include some form or method by which party registrations can be approved by the parties in advance of any voter registration deadline.

I encourage the SEC to modify the bylaws now so that when party registration passes, the rules already include different levels of membership – e.g., registered versus bona fide.  This may also help clarify the “bona fide” issues discussed earlier in this commentary.

I also encourage the SEC to more clearly define “active” as it pertains to the “bona fide” definition.  In the end, there must be some kind of activity tracking system in which members are rated based on their level of activity and only those who achieve a certain rating would be considered “active” and thus bona fide. These records could be maintained in a NEW database maintained by the state (as some county parties do not have the volunteers and infrastructure to develop and maintain this information). Also, if the state maintains the information, then when a person moves within Tennessee, their “active” information would move with them.   

Importantly, I would NOT include donations as a criterion of being “active”.  That looks too much like a poll tax!