🔍 Read the full analysis: The Local Decisions Behind 46 Votes Against Data Centers on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
A public tracker counted 47 local data center measures on Nov. 3 ballots in 10 states; as of Nov. 5, the side favoring restrictions had won or was leading in 46. The measures vary widely, from binding bans to advisory questions, and all results remain unofficial.
Voters in 46 of 47 local data center measures on Nov. 3 ballots favored restrictions or were leading for them, according to the Data Centers on the Ballot tracker, a tally compiled by Redbud Consulting. The measures in 10 states range from outright bans and zoning decisions to temporary pauses and nonbinding questions, reflecting how local data center votes can involve more than a simple yes or no; results are unofficial, and six counts were still being tallied as of Nov. 5.
The tracker lists 47 measures and maps differently worded outcomes into two categories: “Restrict” and “Allow.” At 12:30 p.m. Eastern on Nov. 5, it showed the restriction side leading or winning in 46 measures, with 41 counts complete and six still counting. The tracker’s combined total was 260,177 votes, 72.8% cast for the outcomes it classified as restrictive. That aggregate combines distinct local questions and should not be read as a vote on one uniform policy, particularly given the funding and power costs behind AI data centers.
The count grew after election day. At 6:16 a.m. Eastern on Nov. 4, the tracker showed 45 measures, 40 with results. It later added a nonbinding question in Belvidere, New Jersey, and a charter amendment in Piqua, Ohio. Ballotpedia listed 41 measures before the vote, using a different scope: the tracker includes more Maine, Kansas and New Jersey measures. A separate count cited by 404 Media included recalls and candidate races, as well as ballot questions.
Seven results were checked against official election sites in Kansas, Nevada, Florida and North Dakota, and matched the tracker, according to the source report. Examples include Hubbard, Ohio, where a ban on data centers of any size received 2,622 yes votes and 802 no votes, an unofficial 76.6% in favor. In Boulder City, Nevada, voters rejected a question to allow data centers on city-owned desert land, 4,947 to 2,064. The report says it could not independently check all 47 results; for the rest, it relied on the tracker’s linked sources and local reporting.
Infographic 01 · Data centers on the ballot
Where “45 measures” comes from
One public tracker did the counting. Its numbers grew as results and late additions came in. We checked seven results against official election sites.
local ballot measures on data centers went against them on unofficial counts, 5 Nov 2026, 12:30 p.m. ET. The one exception: Harwood, North Dakota.
- 3 NovBallotpedia: 41 measuresListed before the vote · 10 states
- 4 Nov, 6:16 a.m.Tracker: 45 measures40 with results · 39 against data centers, 1 for
- 4 NovHeadlines: “45 measures”33 won, 6 leading, 1 lost, 5 not reported
- 5 Nov, 12:30 p.m.Tracker: 47 measuresAll with results · 46 against, 1 for · 6 still counting
- Emporia, KSban over 10 MW · Lyon County4,346 : 2,891
- Americus, KSban · Lyon County274 : 127
- El Dorado, KSten-year ban · Butler County2,972 : 1,183
- Ottawa, KSban over 25 MW · Franklin County2,882 : 1,129
- Edgewater, FLcharter ban · Volusia County7,875 : 2,328
- Boulder City, NVallow data centers? · Clark County2,064 : 4,947
- Harwood, NDreview rules · ND Secretary of State105 : 325
Yes : No, unofficial. All seven match the tracker to the vote.
- Ballotpedia lists 4 Maine measures, the tracker 8; Kansas 3 vs 4; New Jersey 1 vs 2. That is the gap from 41 to 47.
- The tracker added Belvidere (NJ) and Piqua (OH) after the first reports, so 45 became 47.
- 404 Media’s “40+ races in 13 states” also counts recalls and candidate races.
- Wilmington, Ohio alone accounts for 4 measures: one vote per rezoned parcel.
Sources: Data Centers on the Ballot (Redbud Consulting), 5 Nov 2026, 12:30 p.m. ET · Ballotpedia, 3 Nov 2026 · official results of Lyon, Butler and Franklin County (KS), Volusia County (FL), Clark County (NV), North Dakota Secretary of State. Nothing is certified yet.
Powered by Thorsten Meyer AI
Local Votes Shape Data Center Siting
The results point to a growing local approval risk for companies seeking sites, land-use changes or public agreements for data centers. In some jurisdictions, a proposed project may now face a ballot test or rules requiring public approval. The practical effect depends on each measure: an outright ban has a different reach from a temporary moratorium, a size limit or an advisory vote.
The source report describes a split in the arguments around development: opponents cite water and energy use, while supporters point to jobs and tax revenue. Those concerns are reflected in ballot language that can specify power capacity in megawatts or water use in gallons per day. The election tally does not establish why voters chose one side, nor does it show that every measure will stop a planned project.
For residents, the immediate consequence is similarly local. Some votes change zoning or city rules; others only register public sentiment. Many measures concern future projects rather than facilities already under construction, according to the source report. Whether a particular vote alters a project’s path depends on its legal status, wording and the steps local officials take after the election.
Products Worth Considering
Bose QuietComfort Ultra Earbuds (2nd Gen), Wireless Noise-Cancelling Earbuds, Black
EcoFlow Delta 3 Portable Power Station – 1024 Wh, 1800 W
Sony WH-1000XM4C (M4 2nd Gen) Wireless Noise-Cancelling Headphones, Black
As an affiliate, we earn on qualifying purchases.
Why the Counts Differ
The headline figure of 45 measures came from the same Redbud Consulting tracker on the morning after the election. The tracker’s stated method counts ballot questions that directly concern data centers, including bans, megawatt caps, zoning referendums, developer agreements, voter-approval requirements and advisory questions. It classifies outcomes as “Restrict” or “Allow” to compare questions with different wording. The tracker says its results are not certified.
Other published totals do not necessarily conflict. Ballotpedia counted 41 measures on election day, including fewer measures in Maine, Kansas and New Jersey than the tracker. The source report also notes that the 404 Media figure of “40+ races across 13 states” includes candidate and recall contests, not just data center ballot questions. The totals therefore depend on what is counted and when.
The measures themselves also differ in force. The source says 37 were binding and 10 advisory. Examples include Hubbard’s ban covering data centers of any size, size thresholds commonly set at 25 megawatts, temporary moratoriums, rezoning referendums and requirements for a public vote before approval. A large “no” total across this collection is not a single national rejection or a uniform legal outcome.
“Nothing on the page is certified.”
— Data Centers on the Ballot, a tracker built by Redbud Consulting
Certification and Local Effects
All results in the tracker were unofficial as of Nov. 5; six measures were still counting. The source report independently checked seven results and said they matched official election sites, but it could not verify every race. Final certified totals may differ from election-night figures.
The overall 46-of-47 count also reflects the tracker’s definition and its grouping of outcomes into “Restrict” and “Allow.” It does not mean all 46 jurisdictions enacted bans. The source material does not provide the full text or legal status of every measure, and it does not establish what effect each result will have on specific projects. The one exception identified by the tracker was Harwood, North Dakota, where a proposal to review rules for large projects received 105 yes votes and 325 no votes, unofficially.
Certification and Local Follow-Up
The next milestones are completion of the six outstanding counts and certification by the relevant election authorities. Until then, the tracker’s figures should be treated as preliminary. Local governments and project proponents will then need to determine how binding measures apply, including whether they alter zoning, create a pause or require another public vote.
For a clearer picture of the development impact, readers will need the final certified results and local follow-up on each measure’s implementation. The source report says the tracker links races to county sources, while noting that its own review covered only seven official results. No single next step applies across all 47 questions.
Key Questions
What does the 46-of-47 figure mean?
It means the Data Centers on the Ballot tracker classified the restriction side as having won or leading in 46 of 47 local measures as of Nov. 5. The results were unofficial, and the questions did not all propose the same policy.
Did voters pass 46 data center bans?
No. The measures included bans, size limits, moratoriums, zoning referendums, public-vote rules and advisory questions. The tracker’s broad “Restrict” category is not equivalent to 46 enacted bans.
Why did reports cite 41, 45 and 47 measures?
The counts used different definitions and were published at different times. The tracker initially listed 45 measures, then added two. Ballotpedia counted 41 under a different scope, while another reported tally included candidate and recall races.
Are the results final?
No. The tracker marked its results unofficial as of Nov. 5, with six counts still in progress. Election authorities must certify results before they are final.
Which measure was the exception?
In Harwood, North Dakota, a measure concerning review rules for large projects had 105 yes votes and 325 no votes, according to the unofficial figures cited in the source report. It was the one measure where the tracker did not show the restriction side leading or winning.
Source: ThorstenMeyerAI.com
