Which counties have the largest discrepancies between provisional ballots and registered voters in certified results?
Executive summary
A comprehensive, county-by-county ranking of the largest discrepancies between provisional ballots and registered voters in certified 2024 results cannot be produced from the reporting provided: the sources explain how provisional ballots are used and canvassed, cite state-level trends (not exhaustive county-level miscounts), and flag places with unusual provisional-ballot dynamics such as Georgia, but none publish a nationwide list of county discrepancies in certified results [1] [2] [3]. Available materials describe the mechanisms that create provisional-ballot differences and point to states and localities where provisional ballots rose or were litigated, but they do not supply the raw county-level crosswalks that would be required to identify the single counties with the largest mismatches [4] [5].
1. What “discrepancy” means in this context—and why it’s not a simple error count
Provisional ballots are cast when a voter’s eligibility cannot be immediately verified and are stored separately until eligibility is validated during the canvass; that means provisional-ballot totals will often diverge from the number of voters initially shown as checked in, and resolving those differences is a central part of the canvass—not necessarily proof of fraud or administrative failure (Brennan Center; EAC) [1] [4]. The canvass and certification process is explicitly designed to reconcile tabulation totals with pollbooks, provisional ballots, and ballot accounting so that certified results include each valid vote, which complicates any effort to declare a “discrepancy” without reviewing local canvass reports and post-certification audit notes [1] [4].
2. What the reporting does reveal: state-level hotspots and notable local disputes
Reporting and post-election analysis flag states where provisional voting rose or where rules increased provisional use: Georgia experienced a large increase in provisional ballots between 2020 and 2024 attributed in part to legislative and administrative changes, and researchers documented a statewide jump in provisional ballots that materially changed how many ballots were processed after Election Day [3]. North Carolina’s post-election data were analyzed in depth by Election Truth Alliance, which reported provisional ballots made up about 0.44% of all votes cast in 2024 in the state—useful context but not a county-level discrepancy ranking [6]. Local litigation has focused attention on particular counties—Rockland County, New York, for example, is the subject of a lawsuit alleging voting discrepancies even though those proceedings do not equate to a verified list of the largest county-level provisional/registration mismatches [7].
3. Why the sources do not support a definitive county ranking
None of the supplied sources publish an authoritative, comparable dataset of provisional ballots versus registered-voter totals for every county that was later linked to certified results; instead they provide descriptions of procedures, state summaries, case studies, and advice for canvass practices [1] [2] [4]. The U.S. Election Assistance Commission and Ballotpedia describe the canvass and certification steps and report-releases, but those references note that official local canvass reports and audits are the primary documents that would show resolved provisional-ballot counts at the county level—documents not included in the provided reporting [4] [8] [9].
4. How an accurate county-level answer could be produced (and where to look next)
A defensible county ranking requires assembling county-level certified returns and canvass reports (showing provisional ballots cast, provisional ballots counted, and total registered voters) and then calculating the relative and absolute gaps; repositories for raw returns and cast-vote records could supply that data (Princeton’s elections data guide and state board canvass reports are starting points) but those datasets must be compiled and normalized across jurisdictions to avoid apples-to-oranges comparisons [5] [4]. For investigative follow-up, county certification statements, state board of elections post-certification reports, and the raw CVR and canvass spreadsheets are the primary source material missing from the current reporting [4] [5].
5. Bottom line and the reporting limitations readers should note
The reporting establishes that provisional ballots play an important, often post-election role, and it flags states and counties with notable provisional-ballot activity and legal scrutiny—examples include Georgia’s surge and local legal challenges like Rockland County—but the sources provided do not contain the comprehensive, auditable county-level dataset necessary to name the specific counties with the largest certified discrepancies between provisional ballots and registered voters [3] [7] [6]. Any claim naming particular counties as having the “largest discrepancies” would therefore require additional primary-source canvass reports and county-certified returns that are not present in the supplied material [4] [5].