What specific evidence did Georgia audits and recounts use to verify Fulton County’s 2020 vote totals?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Georgia validated Fulton County’s 2020 presidential totals through three separate tallies — the initial machine tabulation on Election Day, a hand risk‑limiting audit that manually counted ballots, and a machine recount that rescanned ballots — using paper and digital artifacts including batch audit sheets, cast‑vote records (CVRs), memory cards, poll‑tape data and signature verification records as its primary evidentiary trail [1] [2] [3].
1. The three independent tallies that formed the backbone of verification
State and county officials relied on the convergence of three independent counts — election‑day machine totals, a statewide hand risk‑limiting audit conducted Nov. 11–19, 2020, and a machine recount requested by the Trump campaign — because each used different mechanics to reach totals; all three produced broadly similar results, which state actors and observers treated as reinforcing that the certified outcome was accurate [1] [2] [3].
2. Paper ballots and batch paperwork: audit board batch sheets and inventory sheets
The hand audit used a “sort and stack” method in which teams grouped ballots by presidential choice, recorded batch tallies on paper Audit Board Batch Sheets (ABBSs) and kept inventory sheets and sign‑in logs to document who counted which batch and when; those paper ABBSs were later transcribed into VotingWorks’ Arlo audit software to produce an audit spreadsheet used to compare totals [4] [5].
3. Digital crosschecks: CVRs, memory cards and machine recount logs
Officials compared machine‑produced cast‑vote records (CVRs) and memory‑card outputs against reported tabulations to ensure that images and electronic vote records matched the totals; the CVRs were used to confirm that certain duplicated or triplicated ballot images existed in the tabulation files and to reconcile machine recount numbers with the initially reported machine totals [4].
4. Signature matching and absentee/advance ballot controls
Signature verification processes — using ElectioNet or matching to voter registration applications — were part of the ballot acceptance and canvass trail: absentee and advance ballots went through signature review and, where unmatched, could be set aside or treated as provisional, producing another documentary record used in audits and in later challenges to chain‑of‑custody and legality [6] [7].
5. What the reviews found and the limits of those paper/digital artifacts
Post‑election reviews documented errors: state investigators and the Fulton County performance review noted double‑scanning and other management mistakes, and the Georgia Election Board later reprimanded Fulton for double‑counting thousands of ballot images and other procedural violations; those findings underscore that while paper tapes, CVRs and ABBSs provided evidence to reconstruct totals, the audit record was not flawless and some ABBS entries or audit inputs were omitted or transcribed incorrectly into the audit database [8] [4] [3].
6. Missing poll tapes, uncertified tapes and continuing disputes over procedural compliance
Poll tapes from some tabulators were later reported as unavailable, and a state review found wide failures to properly sign or verify tabulation tapes at advance voting sites; Georgia officials have emphasized that poll tapes are only one element of the paper trail — memory cards, CVRs, county reports and the recounts themselves were additional corroboration — but opponents argue that missing or unsigned tapes create legal and procedural vulnerabilities even if they do not, by themselves, prove altered totals [1] [9] [10].
7. Litigation, requests for original ballots and what remains unsettled
Because some procedural lapses and duplicated images were documented, later legal actions and requests (including by the Department of Justice and state board subpoenas) have sought the underlying ballots, used/voided stubs, signature envelopes and image files to allow fuller forensic review; reporting shows these records are the same types of evidence audits and recounts relied on to verify totals, but the presence of errors in chain‑of‑custody paperwork has become the focus of new litigation rather than proof that vote totals were incorrect [11] [10].
8. Bottom line and evidentiary balance
The specific evidence used to verify Fulton County’s 2020 presidential totals comprised the paper ballots themselves, ABBSs and inventory sheets, CVRs and memory‑card outputs, poll tapes and signature verification logs, supplemented by independent hand counts and machine rescans; these overlapping sources produced consistent tallies even as state reviews later identified procedural errors and isolated double‑counting in image files that fed subsequent legal challenges — evidence of robust cross‑checks, but not of a perfect chain‑of‑custody [2] [4] [8].