Introduction

Center for Election Confidence, Inc. (“CEC”) is a non-profit organization based in Arlington, which is recognized as a key “civil society group”^[1] that promotes ethics, integrity, and professionalism in the electoral process. CEC works to ensure that all citizens can vote freely within an election system of reasonable procedures that promote election integrity, prevent vote dilution and disenfranchisement, and instill public confidence in election systems and outcomes.

CEC submits these comments concerning the State Board of Elections’ Proposed Final Regulation, 1 VAC 20-100, Ranked Choice Voting, 42 Va. Reg. of Reguls. 2031 (“Proposed Final Regulation”). CEC previously submitted comments concerning Chapter 100 during the State Board’s periodic review.[2] In those comments, CEC urged the State Board to update outdated statutory references and complete other rulemakings required by Title 24.2 of the Code of Virginia to the extent such rulemakings would be codified in Chapter 100.[3]

CEC does not support ranked choice voting (“RCV”) as a policy matter. CEC has supported research by Professor Nolan McCarty, the Susan Dod Brown Professor of Politics at Princeton University, analyzing the effects of RCV on racial and ethnic minority electorates. That research found that RCV may disproportionately decrease representation and electoral influence among minority electorates because minority voters are disproportionately more likely to “exhaust” their ballots, thereby removing those ballots from decisive tabulations.[4] CEC believes the core mechanics of RCV, including transferable votes and exhausted ballots, make RCV fundamentally confusing, insufficiently transparent, harmful to voters’ confidence in elections, and wholly unfit for use in Virginia.

For that reason, these comments should not be read as support for RCV, as support for expanding RCV, or as a roadmap for making RCV sound election policy. CEC’s position is more limited. If Virginia law permits localities to use RCV, the State Board should not adopt regulations that make an already flawed system less transparent and results harder to reproduce and/or more dependent on software-driven calculations that voters cannot readily understand. To the extent the State Board is not required to adopt regulations facilitating RCV, the State Board should decline to do so. Better still, the State Board should advise the General Assembly of the underlying issues caused by RCV that served as the impetus for the workarounds and shortcuts proposed here. 

CEC’s Position on the Proposed Final Regulation

The State Board explains that the existing Chapter 100 framework requires “the elimination of one candidate in each round” of RCV tabulation and that, in races with many candidates “including independents and write-ins,” this process may extend tabulation during the canvass. The State Board therefore proposes to add an option for “batch elimination” of multiple candidates in a single-winner instant-runoff contest where it is “mathematically impossible” for candidates set for batch elimination to be elected.[5]

CEC opposes the Proposed Final Regulation as written. The fact that RCV tabulation may be lengthy or difficult to complete is not a reason to make RCV less observable or the results it offers less transparent or harder to reproduce. It is evidence of the deeper problem inherent with RCV. Batch elimination may reduce the number of visible rounds, but doing so also compresses multiple tabulation steps into a single administrative or software-driven action. Rather than improving RCV overall, this proposed solution to lengthy tabulation risks worsening the very transparency problems that make RCV objectionable in the first place.

CEC therefore urges the State Board to recall the Proposed Final Regulation and otherwise not to proceed with language introducing batch elimination. Batch elimination would trade an alleged reduction in tabulation time for the decreased transparency and increased administrability problems inherent in RCV. The introduction of batch tabulation compounds these transparency and administrability problems by making the tabulation process less observable, less intuitive, and more dependent on mathematical or software-driven determinations that are harder to reproduce and that ordinary voters are unlikely to be able to follow in real time. If sequential elimination makes RCV lengthy or difficult to administer, that problem weighs against the use of RCV, not in favor of compressing multiple elimination rounds into a less transparent process that magnifies more serious issues. The State Board should recall and not implement the Proposed Final Regulation or other proposals that incorporate batch elimination.

I. Proposed 1 VAC 20-100-10

Definition of “Batch Elimination”

The Proposed Final Regulation defines “batch elimination” as “the simultaneous defeat of multiple candidates for whom it is mathematically impossible to be elected”.[6] That definition is not merely incomplete; it authorizes a tabulation method that appears contrary to the structure required by Virginia law. [7]

Va. Code Ann. § 24.2-673.1(A) defines RCV as a method by which “tabulation proceeds in rounds such that in each round either a candidate or candidates are elected or the last-place candidate is defeated”. (emphasis added). Chapter 100 similarly defines RCV as a process in which, after first-choice tabulation, “tabulation proceeds in rounds such that in each round either a candidate is elected or the last-place candidate is defeated”.[8] (emphasis added). In both the statute and the existing regulation, the defeat mechanism is singular: one “last-place candidate” is defeated per round. The proposed final regulation’s definition of “batch elimination”, by contrast, would authorize the simultaneous defeat of multiplecandidates per round.

That change does not merely clarify the existing RCV process. It substitutes the statutory round-by-round single-candidate elimination structure with one that removes multiple candidates from tabulation in a single round. The State Board’s general authority to implement RCV does not permit it to substitute a different elimination procedure for the statutory single-candidate elimination sequence. If the General Assembly intended to authorize simultaneous elimination of multiple candidates in a single round, it could have done so expressly or written the Code to allow for such a system. It did not.

CEC recommends that the State Board remove batch elimination from the proposed final regulation.

Definition of “Mathematically Impossible to Be Elected”

The Proposed Final Regulation defines “mathematically impossible to be elected,” with respect to a given candidate in two ways: first, where the sum of votes credited to that candidate and all candidates with the same or fewer votes is less than the votes credited to the candidate with the next greatest number of votes; and second, where the candidate has a lower vote total than a candidate described in the first clause.[9] This definition may be mathematically meaningful to tabulation specialists, but it will not be readily understandable to many voters, candidates, observers, or local officials.

That matters because RCV is already less intuitive than ordinary plurality or majority elections. A rule that allows several candidates to be defeated at once based on a technical calculation will further reduce public understanding. Evincing the magnitude of the issues at play, the following items vital to any effort to implement such a definition either remain unresolved or require deeper explication:

• whether “all votes credited” refers only to votes from active ballots credited to active candidates in the current round;
• whether exhausted ballots are excluded from the calculation;
• whether ranked choice overvotes, duplicate rankings, and skipped rankings are resolved under 1 VAC 20-100-40 before the calculation is performed;
• whether the calculation occurs before or after transfers in a round in which batch elimination is considered;
• whether unresolved write-ins or grouped write-in totals may or must be included in the calculation; and
• which person, and, if applicable, which approved tabulation tool is responsible for determining that the mathematical-impossibility standard has been satisfied.

II. 1 VAC 20-100-50

Existing Sequential Instant-Runoff Structure

The existing instant-runoff provisions proceed sequentially: first, an active candidate with a majority of votes from active ballots is elected; second, if two or fewer active candidates remain, the candidate with the greatest number of votes is elected; and third, if more than two active candidates remain and no candidate has a majority, the active candidate with the fewest votes is defeated and that candidate’s votes are transferred before a new round begins.[10]

That sequential structure is important because it is one of the few features of RCV that can be explained to the public in an orderly way: count, check, eliminate, transfer, and repeat. Batch elimination disrupts that order by allowing multiple eliminations in a single round. As described above, this change trades an alleged increase in tabulation speed for greater transparency and administrability issues, a bad bargain.

Operative Batch-Elimination Provision

Proposed subdivision (B)(4) provides that, “notwithstanding subdivision (3)(b) of 1 VAC 20-100-40, more than one candidate may be defeated in a round by batch elimination, unless batch elimination would result in only one continuing candidate, in which case no batch elimination shall occur”.[11]

First, the State Board should reconsider whether batch elimination should be adopted at all. The stated problem is that ordinary RCV can take too long to tabulate. The proposed final regulation does not resolve that issue. Indeed, a voting method that requires shortcuts to remain administratively workable may be too complicated for voters and officials in the first place. 

Second, the phrase “may be defeated” is ambiguous. (emphasis added). Consider this: If batch elimination is optional, similarly situated localities could tabulate otherwise identical elections differently. If batch elimination is mandatory whenever a mathematical condition is satisfied, the State Board would need to specify when batch elimination is permitted, when it is required, and who makes that determination—key provisions on which the proposed final regulation is silent.

In a nutshell, batch elimination is a not-yet-ready-for-primetime proposal that is not fit for purpose.

Cross-Referenced 1 VAC 20-100-40 Tie-Breaking Issue

Proposed 1 VAC 20-100-50(B)(4) would operate “notwithstanding subdivision (3)(b) of 1 VAC 20-100-40”. That cross-reference is significant. Existing 1 VAC 20-100-40 (3)(b) provides that when two or more persons have an equal number of votes for a seat and the fewest number of votes in a round, tabulation cannot continue until one of the persons is defeated, and that the person continuing as the active candidate is determined by lot as prescribed by Va. Code § 24.2-674.[12]

CEC opposes the use of batch elimination to override this lot procedure primarily because the General Assembly has provided expressly otherwise. The existing regulation incorporates a public, statutorily grounded tie-breaking process. That process is observable, understandable, and capable of public verification. By contrast, batch elimination would, by administrative edict, allow a software-driven or mathematical calculation to displace the statute’s public lot procedure in the very circumstance where transparency is most important: when tied candidates are competing to remain active in the tabulation.

Notwithstanding its non-compliance with Virginia law, this change would make RCV even less transparent and less accountable. If the statutory and regulatory framework requires a lot procedure to determine which tied candidate or person continues, the State Board should not authorize that process to be bypassed by regulation. CEC therefore recommends that the State Board remove the “notwithstanding subdivision (3)(b) of 1 VAC 20-100-40” language to make clear that the ordinary lot procedure applies unless the Code of Virginia expressly provides otherwise.

IV. Write-In Candidates and Persons

The State Board’s Background section expressly identifies contests with many candidates, including “write-ins”, as a reason for adding batch elimination.[13] But the new definition and operative text refer to “candidates” and existing Chapter 100 uses both “candidate” and “person” in relevant provisions. Chapter 100 also provides that, at all RCV elections except primary elections, a voter may vote for “any person other than the listed candidates” by writing or hand printing the person’s name on the official ballot.[14]

That distinction matters because write-in votes may not be resolved in the same way as votes for listed candidates. If write-ins are grouped, unresolved, or pending adjudication, their treatment could affect the lowest vote totals, tie-breaking, transfer calculations, and any determination that a listed candidate or write-in person is mathematically impossible to elect.

This demonstrates again the reasons why RCV is not suited for use in Virginia elections.

This section of the proposed final regulation is not ready for public evaluation or implementation and should be recalled. Remaining unresolved are vital, integral, and obvious questions, such as whether write-ins would be treated as individual persons, a grouped write-in category, or candidates only after adjudication for purposes of batch elimination. Further, the proposed final regulation does not explain whether batch elimination would be implemented before or after all write-in votes have been resolved for tabulation purposes.

V. Public Reporting, Software Approval, and Reproducibility

Batch elimination would make RCV even more dependent on calculations that voters are unlikely to be able to perform on their own and on tabulation tools that may be controlled or approved outside public view. The Proposed Final Regulation states that tabulation tools, including software, used for tabulation under subsections B and C must be approved for use by the State Board.[15] That approval requirement is important, but would not be sufficient in its current form. Instead, the Proposed Final Regulation should be recalled in order to include a requirement for disclosure of the standards by which tabulation tools are approved, the person or persons responsible for testing and certifying such tools to the State Board’s standards, and the records that must be preserved after such tools are employed.

VI. Statutory Conformity and CEC’s Unresolved Periodic-Review Recommendations

CEC renews its periodic review recommendation that the State Board update outdated statutory references and complete other rulemakings required by Title 24.2 to the extent such rulemakings would be codified in Chapter 100. The State Board’s decision to amend Chapter 100 makes that recommendation more urgent, not less.

To the extent the State Board concludes that Virginia law does not require it to adopt additional rules facilitating RCV, the State Board should exercise its discretion to decline to do so. CEC also encourages the State Board to provide candid information to the General Assembly regarding the administrative complexity, voter confusion risks, and voter confidence concerns associated with RCV, including its position that technical shortcuts such as batch elimination are necessary to make RCV function at a basic level.

Conclusion

CEC does not support RCV because it risks confusing, discouraging, and disenfranchising voters. The Proposed Final Regulation does not resolve CEC’s concerns; as written, it risks making an already nontransparent voting method less transparent by compressing multiple rounds into a single batch elimination action based on ambiguous provisions and definitions; further, the Proposed Final Regulation is inconsistent with the Code of Virginia. The State Board should recall the Proposed Final Regulation.

Respectfully submitted this 1st day of July 2026, 

/s/ Caleb J. Hays

Chief Policy Counsel 

Center for Election Confidence, Inc.